Welcome to PVCDROM

Add new comment

Welcome to PVCDROM
Add new comment
3 comment(s)
Español

Photovoltaics is a most elegant energy source. Light shines on a crystal and produces electricity. Its as simple as that. There are no
moving parts. The fuel source (sunlight) is free, abundant and widely distributed, available to every country and person in the world.
At over 165,000 TW the solar resource dwarfs the world’s current power usage of 16 TW or even our projected future usage of 60 TW.

The many advantages of photovoltaics lend itself to being the ultimate energy source. However, it required the semiconductor
revolution and advances in manufacturing before photovoltaics could begin to reveal its full potential.

For the last two decades, photovoltaics has been the fastest growing industry for its size. Continuing at the present growth rate of 40%
for the next two decades will allow photovoltaics to be the worlds largest energy source. To keep growing at 40% a year, however, will
require that everyone learn a little bit about how photovoltaics works and for today’s engineering students to be conversant in the
principles and technologies that will make this growth possible. This site provides casual readers an opportunity to research this new
technology and students of physics, engineering, or energy policy the opportunity to deeply explore photovoltaics..

This site is an electronic book on PV. It can either be read from start to finish as a course on photovoltaics or used as a reference. This
project started out as an electronic textbook, distributed on CDs and so we have kept the ungainly title of PVCDROM. As the project
and the text itself has grown over the years we have added many enhancements unavailable in static textbooks. Be sure to read the
instruction page to make sure you have all the appropriate plugins installed and so that you will be aware of all of the features available
to learners.

Help us improve the PVCDROM. If you find any errors or have suggestions please click on the [+] Feedback link at the bottom
right of each page.

Stuart Bowden and Christiana Honsberg work at the Solar Power Labs at ASU (http://pv.asu.edu/)

This material is based in part upon work supported by the National Science Foundation under Grant Numbers 0935247 and 1041895.
Any opinions, findings, and conclusions or recommendation expressed in this material are those of the author(s) and do not necessarily
reflect the views of the National Science Foundation.

Instructions
Add new comment
10 comment(s)
Español
Login

The only reason to login is to leave a comment. The link for login is available at the bottom of each page. The comments are reviewed
about once a week. Spam is a huge problem so you need to confirm a valid email address.

Referencing the Website

For an academic work use one of the following in your reference format:
C.B.Honsberg and S.G.Bowden, “Photovoltaics Education Website,” www.pveducation.org, 2019.
or
C.B.Honsberg and S.G.Bowden, “Absorption Coefficient,” page on www.pveducation.org, 2019.
While I try to keep the weblinks the same, they do change around as content is added. A URL reference to a specific page will change
over the life of a textbook or paper.
Send me an email (sgbowden@asu.edu) when your work is about to be published. It’s not essential but I like to to know where the
work is used.

Use of images etc

You may use a small fraction of the website such as a few pictures in other websites or publications. Just make sure to reference the
site and send an email to sgbowden@asu.edu. Use the reference style above for academic work. Web pages are easier to keep up to
date so they can just link directly to the URL. Websites can also use the academic reference above if they wish. If your need goes
beyond “fair use”, let’s call it three pictures, please send me an email. As long as its not a wholesale copy of the site its probably ok.

Browser compatibility

The PVCDROM uses interactivity and animations to explain concepts in photovoltaics. We are gradually phasing out the animations in
Flash and Shockwave in preference for modern browsers where plugins are not needed and everything is done in HTML5. Most
modern browsers such as Firefox, Chrome, Internet Explorer (10 or later) and Safari support HTML5.

Live Calculation

Spread throughout the document are mini-calculators for the equations in the text. Enter numbers in the blue fields. Javascript is
required for the live calculations and is present on nearly all browsers.

Enter very large or very small numbers using the standard exponential notation. For example, 1 000 000 (106) is entered as 1e6.

Flash Player

Most of the animations are now in Flash. It is available on most computers, including linux, but not on the iPhone, iPad or most tablets.
Flash is no longer installed with Firefox and is also being phased out in Chrome. To install flash go to: https://get.adobe.com
/flashplayer/ We would like to convert all the Flash to HTML5 but it’s a laborious process.

Shockwave

Director Shockwave is no more. The plugin works on very few computers and phased out by Adobe in early 2019. The last remaining
animations were converted to GIFs that run automatically.
Java

Java was previously used for the interactive graphs but is no longer necessary.

Appearance

Press F11 on most browsers to make the pages fit the screen.

Introduction
Add new comment
Español
한국어
Русский
简体中⽂

Introduction Page - English

Introduction
Add new comment
1 comment(s)
Español
Русский
한국어
简体中⽂

Photovoltaics is the process of converting sunlight directly into electricity using solar cells. Today it is a rapidly growing and
increasingly important renewable alternative to conventional fossil fuel electricity generation, but compared to other electricity
generating technologies, it is a relative newcomer, with the first practical photovoltaic devices demonstrated in the 1950s. Research
and development of photovoltaics received its first major boost from the space industry in the 1960s which required a power supply
separate from "grid" power for satellite applications. These space solar cells were several thousand times more expensive than they are
today and the perceived need for an electricity generation method apart from grid power was still a decade away, but solar cells became
an interesting scientific variation to the rapidly expanding silicon transistor development with several potentially specialized niche
markets. It took the oil crisis in the 1970s to focus world attention on the desirability of alternate energy sources for terrestrial use,
which in turn promoted the investigation of photovoltaics as a means of generating terrestrial power. Although the oil crisis proved
short-lived and the financial incentive to develop solar cells abated, solar cells had entered the arena as a power generating technology.
Their application and advantage to the "remote" power supply area was quickly recognized and prompted the development of
terrestrial photovoltaics industry. Small scale transportable applications (such as calculators and watches) were utilised and remote
power applications began to benefit from photovoltaics.

In the 1980s research into silicon solar cells paid off and solar cells began to increase their efficiency. In 1985 silicon solar cells
achieved the milestone of 20% efficiency. Over the next decade, the photovoltaic industry experienced steady growth rates of between
15% and 20%, largely promoted by the remote power supply market. The year 1997 saw a growth rate of 38% and today solar cells are
recognized not only as a means for providing power and increased quality of life to those who do not have grid access, but they are
also a means of significantly diminishing the impact of environmental damage caused by conventional electricity generation in
advanced industrial countries.

The increasing market for, and profile of photovoltaics means that more applications than ever before are "photovoltaically powered".
These applications range from power stations of several megawatts to the ubiquitous solar calculators. PVCDROM aims to provide an
overview of terrestrial photovoltaics to furnish the non-specialist with basic information. It is hoped that having used PVCDROM you
will understand the principles of photovoltaic devices and system operation, you will be able to identify appropriate applications, and
you will be capable of undertaking photovoltaic system design. By gradually increasing the number of people who are familiar with
photovoltaic concepts and applications, we hope to increase the use of photovoltaics in appropriate applications.

Solar Energy
Add new comment
Español
 한국어
Русский

Solar energy in one form or another is the source of nearly all energy on the earth. Humans, like all other animals and plants, rely on
the sun for warmth and food. However, people also harness the sun's energy in many other different ways. For example, fossil fuels,
plant matter from a past geological age, is used for transportation and electricity generation and is essentially just stored solar energy
from millions of years ago. Similarly, biomass converts the sun's energy into a fuel, which can then be used for heat, transport or
electricity. Wind energy, used for hundred of years to provide mechanical energy or for transportation, uses air currents that are created
by solar heated air and the rotation of the earth. Today wind turbines convert wind power into electricity as well as its traditional uses.
Even hydroelectricity is derived from the sun. Hydropower depends on the evaporation of water by the sun, and its subsequent return
to the Earth as rain to provide water in dams. Photovoltaics (often abbreviated as PV) is a simple and elegant method of harnessing the
sun's energy. PV devices (solar cells) are unique in that they directly convert the incident solar radiation into electricity, with no noise,
pollution or moving parts, making them robust, reliable and long lasting. Solar cells are based on the same principles and materials
behind the communications and computer revolutions, and this CDROM covers the operation, use and applications of photovoltaic
devices and systems.

Solar powered light house at Montague Island, a National Parks and Wildlife sanctuary on the East coast of Australia. The small panel
on the left powers the light house. The large panel on the right powers the cottages which can partially be seen in the background. The
cottages contain facilities for the National Parks' caretaker and facilities for researchers on the island.

The Greenhouse Effect

Add new comment
3 comment(s)
Español
한국어
Русский

Although the bulk of photovoltaic devices today are used for purely practical and economic reasons, a potential benefit of
photovoltaics is that PV is one of the most environmentally benign of any electricity generating source. The environmental impact of
electricity generation, particularly the greenhouse effect, adds an important reason for examining photovoltaics. A brief overview of
the greenhouse effect is given below.

The Earth's temperature is a result of an equilibrium established between the incoming radiation from the sun and the energy radiated
into space by the Earth. The outgoing radiation emitted by the Earth is strongly affected by the presence and composition of the Earth's
atmosphere. If we had no atmosphere, as on the moon, the average temperature on the Earth's surface would be about -18°C. However,
a natural background level of 270 ppm carbon dioxide (CO2) in the atmosphere absorbs outgoing radiation, thereby keeping this
energy in the atmosphere and warming the Earth. The atmosphere causes the Earth's temperature to be about 15°C on average, 33°C
above the moon's. Carbon dioxide absorbs strongly in the 13-19 µm wavelength band and water vapour, another atmospheric gas,
absorbs strongly in the 4-7 µm wavelength band. Most outgoing radiation (70%) escapes in the "window" between 7-13 µm.
Human activities are increasingly releasing "anthropogenic gases" into the atmosphere, which absorb in the 7-13 µm wavelength range,
particularly carbon dioxide, methane, ozone, nitrous oxides and chlorofluorocarbons (CFC's). These gases prevent the normal escape
of energy and potentially will lead to an increase in terrestrial temperature. Present evidence suggests "effective" CO2 levels will
double by 2030, causing global warming of 1~4°C. This would lead to changes in wind patterns and rainfall and as a result may cause
the interior of continents to dry out and cause the Earth's oceans to rise. Further increases in the release of anthropogenic gases would,
of course, cause more severe effects.

Correlation of the rise in atmospheric carbon dioxide concentration (blue line) with the rise in average temperature (red line)1.

The graph above only goes to the mid-90s since that was the time we started writing the PVCDROM. At the time there was
considerable discussion about whether the warming was a trend or a statistical fluctuation. There was some hope that the average
temperatures would go down again to the level of the statistical average. In the intervening years the temperature of the earth has
continued to rise as shown in the graph below:

Average surface temperature of the earth. Temperatures are continuing their upward rise.NASA20102

Clearly, human activities have now reached a scale where they are impacting on the planet's environment and its attractiveness to
humans. The side-effects could be devastating and technologies with low environmental impact and no "greenhouse gas" emissions are
likely to be of increasing importance over the coming decades. Since the energy sector is the major producer of "greenhouse gases" via
the combustion of fossil fuels, technologies such as photovoltaics, which can substitute for fossil fuels, must increasingly be used 3.

1. R. A. Kerr, “Sun’s Role in Warming Is Discounted”, Science, vol. 268, pp. 28 - 29, 1995.
2. J. Hansen, “Global temperature change”, Proceedings of the National Academy of Sciences, vol. 103, pp. 14288 - 14293,
2006.
 3. A. W. Blakers, Green, M. A., Leo, T., Outhred, H., and Robins, B., The Role of Photovoltaics in Reducing Greenhouse Gas
Emissions. Canberra: Australian Government Publishing Service, 1991.

Properties of Sunlight
Add new comment
Español
한국어

Properties of sunlight page.

Basics of Light
Add new comment
4 comment(s)
Español
한국어

As its name implies, photovoltaics is the direct conversion of light (photo) to electricity (-voltaic). It is explained by the great scientific
advance of the 20th century, namely quantum physics. In this section we discuss the nature of light and the interaction with matter.

Properties of Light
Add new comment
2 comment(s)
Español
한국어

The light that we see everyday is only a fraction of the total energy emitted by the sun incident
on the earth. Sunlight is a form of "electromagnetic radiation" and the visible light that we see is
a small subset of the electromagnetic spectrum shown at the right.

The electromagnetic spectrum describes light as a wave which has a particular wavelength. The
description of light as a wave first gained acceptance in the early 1800's when experiments by
Thomas Young, François Arago, and Augustin Jean Fresnel showed interference effects in light
beams, indicating that light is made of waves. By the late 1860's light was viewed as part of the
electromagnetic spectrum. However, in the late 1800's a problem with the wave-based view of
light became apparent when experiments measuring the spectrum of wavelengths from heated
objects could not be explained using the wave-based equations of light. This discrepancy was
resolved by the works of 1 in 1900, and 2 in 1905. Planck proposed that the total energy of light
is made up of indistinguishable energy elements, or a quanta of energy. Einstein, while
examining the photoelectric effect (the release of electrons from certain metals and
semiconductors when struck by light), correctly distinguished the values of these quantum
energy elements. For their work in this area Planck and Einstein won the Nobel prize for physics
in 1918 and 1921, respectively and based on this work, light may be viewed as consisting of
"packets" or particles of energy, called photons..

Today, quantum-mechanics explains both the observations of the wave nature and the particle
nature of light. In quantum mechanics, a photon, like all other quantum-mechanical particles
such as electrons, protons etc, is most accurately pictured as a "wave-packet". A wave packet is
defined as a collection of waves which may interact in such a way that the wave-packet may
either appear spatially localized (in a similar fashion as a square wave which results from the
addition of an infinite number of sine waves), or may alternately appear simply as a wave. In the
cases where the wave-packet is spatially localized, it acts as a particle. Therefore, depending on
the situation, a photon may appear as either a wave or as a particle and this concept is called
"wave-particle duality".
The electromagnetic spectrum.
A complete physical description of the properties of light requires a quantum-mechanical
analysis of light, since light is a type of quantum-mechanical particle called a photon. For photovoltaic applications, this level of detail
is seldom required and therefore only a few sentences on the quantum nature of light are given here. However, in some situations
(fortunately, rarely encountered in PV systems), light may behave in a manner which seems to defy common sense, based on the
simple explanations given here. The term "common sense" refers to our own observations and cannot be relied on to observe the
quantum-mechanical effects because these occur under conditions outside the range of human observation. For further information on
the modern interpretation of light please refer to 3. A wave-packet, or photon is pictured as used in PVCDROM below.

High energy photon for blue light.

Lower energy photon for red light.

Low energy photon for infrared light. (should be invisible)

There are several key characteristics of the incident solar energy which are critical in determining how the incident sunlight interacts
with a photovoltaic converter or any other object. The important characteristics of the incident solar energy are:

the spectral content of the incident light;

the radiant power density from the sun;
the angle at which the incident solar radiation strikes a photovoltaic module; and
the radiant energy from the sun throughout a year or day for a particular surface.

By the end of this chapter you should be familiar with the above four concepts..

1. M. Planck, “Distribution of energy in the normal spectrum”, Verhandlungen der Deutschen Physikalischen Gesellschaft, vol.
2, pp. 237-245, 1900.
2. A. Einstein, “Generation and transformation of light”, Annalen der Physik, vol. 17, 1905.
3. R. P. Feynman, QED : The Strange Theory of Light and Matter. 1985.

Energy of Photon
Add new comment
1 comment(s)
Español
한국어

A photon is characterized by either a wavelength, denoted by λ or equivalently an energy, denoted by E. There is an inverse
relationship between the energy of a photon (E) and the wavelength of the light (λ) given by the equation:

Photon Energy
Edit
Delete
Clone
 Read more about Photon Energy

where h is Planck's constant and c is the speed of light. The value of these and other commonly used constants are given in the
constants page.

h = 6.626 × 10 -34 joule·s

c = 2.998 × 108 m/s

By multiplying to get a single expression, hc = 1.99 × 10-25 joules-m

The above inverse relationship means that light consisting of high energy photons (such as "blue" light) has a short wavelength. Light
consisting of low energy photons (such as "red" light) has a long wavelength.

When dealing with "particles" such as photons or electrons, a commonly used unit of energy is the electron-volt (eV) rather than the
joule (J). An electron volt is the energy required to raise an electron through 1 volt, thus a photon with an energy of 1 eV = 1.602 ×
10-19 J.

Therefore, we can rewrite the above constant for hc in terms of eV:

hc = (1.99 × 10-25 joules-m) × (1ev/1.602 × 10-19 joules) = 1.24 × 10-6 eV-m

Further, we need to have the units be in µm (the units for λ):

hc = (1.24 × 10-6 eV-m) × (106 µm/ m) = 1.24 eV-µm

By expressing the equation for photon energy in terms of eV and µm we arrive at a commonly used expression which relates the
energy and wavelength of a photon, as shown in the following equation:

Photon Energy : Electron-Volt

Edit
Delete
Clone

Read more about Photon Energy : Electron-Volt

The exact value of 1 × 106(hc/q) is 1.2398 but the approximation 1.24 is sufficient for most purposes.

To find the energy of a photon at a particular wavelength, click on the map above.

Photon Energy
Edit
Delete
Clone

Wavelength - Photon Energy Calculator

Wavelength (µm)
Photon Energy (eV)
Photon Energy (joules)

Español
Photon Flux
Add new comment
7 comment(s)
Español
한국어

The photon flux is defined as the number of photons per second per unit area:

The photon flux is important in determining the number of electrons which are
generated, and hence the current produced from a solar cell. As the photon flux does
not give information about the energy (or wavelength) of the photons, the energy or
wavelength of the photons in the light source must also be specified. At a given
wavelength, the combination of the photon wavelength or energy and the photon flux
at that wavelength can be used to calculate the power density for photons at the
particular wavelength. The power density is calculated by multiplying the photon flux
by the energy of a single photon. Since the photon flux gives the number of photons
striking a surface in a given time, multiplying by the energy of the photons comprising
the photon flux gives the energy striking a surface per unit time, which is equivalent to
a power density. To determine the power density in units of W/m², the energy of the
photons must be in Joules. The equation is:

Power Density
Edit
Delete
Clone

Read more about Power Density

where Φ is the photon flux and q is the value of the electronic charge 1.6 ·10-19

Power Density
Edit
Delete
Clone

Photon Flux - Power Density Calculator

Photon Flux m-2s-1 Photon Energy eV

Power Density W/m2

One implication of the above equations is that the photon flux of high energy (or short wavelength) photons needed to give a certain
radiant power density will be lower than the photon flux of low energy (or long wavelength) photons required to give the same radiant
power density. In the animation, the radiant power density incident on the surface is the same for both the blue and red light, but fewer
blue photons are needed since each one has more energy.

Spectral Irradiance
Add new comment
5 comment(s)
Español
한국어

The spectral irradiance as a function of photon wavelength (or energy), denoted by F, is the most common way of characterising a light
source. It gives the power density at a particular wavelength. The units of spectral irradiance are in Wm-2µm-1. The Wm-2 term is the
power density at the wavelength λ(µm). Therefore, the m-2 refers to the surface area of the light emitter and the µm-1 refers to the
wavelength of interest.

In the analysis of solar cells, the photon flux is often needed as well as the spectral irradiance. The spectral irradiance can be
determined from the photon flux by converting the photon flux at a given wavelength to W/m2 as shown in the section on Photon Flux.
The result is then divided by the given wavelength, as shown in the equation below.

Photon Flux
Edit
Delete
Clone

Read more about Photon Flux

where in SI units:
F(λ) is the spectral irradiance in Wm-2μm-1;
Φ is the photon flux in # photons m-2sec-1;
E and λ are the energy and wavelength of the photon in joules and meters respectively; and

The spectral irradiance is more commonly expressed in terms of wavelength so that:

Photon Flux : Wavelength

Edit
Delete
Clone

Read more about Photon Flux : Wavelength

where:
F(λ) is the spectral irradiance in Wm-2µm-1;
Φ is the photon flux in # photons m-2sec-1;
E and λ are the energy and wavelength of the photon in eV and µm respectively; and
q is a constant of 1.6 · 10-19
The spectral irradiance of artificial light sources (left axis) compared to the spectral irradiance from the sun (right axis).

Radiant Power Density

Add new comment
1 comment(s)
Español
한국어

The total power density emitted from a light source can be calculated by integrating the spectral irradiance over all wavelengths or
energies

where:

H is the total power density emitted from the light source in W m-2;
F(λ) is the spectral irradiance in units of Wm-2μm-1; and
dλ is the wavelength.

However, a closed form equation for the spectral irradiance for a light source often does not exist. Instead the measured spectral
irradiance must be multiplied by a wavelength range over which it was measured, and then calculated over all wavelengths. The
following equation can be used to calculate the total power density emitted from a light source.

H is the total power density emitted from the light source in W m-2;
F(λ) is the spectral irradiance in units of Wm-2μm-1; and
Δλ is the wavelength.
Calculating the total power density from a source requires integrating over the spectrum by calculating the area of each element and
then summing them together.

Measured spectra are typically not smooth as they contain emission and absorption lines. The wavelength spacing is usually not
uniform to allow for more data points in the rapidly changing parts of the spectrum. The spectral width is calculated from the mid-
points between two the adjacent wavelengths.

Power in each segment is then:

Summing all the segments gives the total power H as in the equation above.

Blackbody Radiation
Add new comment
2 comment(s)
Español
한국어

Many commonly encountered light sources, including the sun and incandescent light bulbs, are closely modelled as "blackbody"
emitters. A blackbody absorbs all radiation incident on its surface and emits radiation based on its temperature. Blackbodies derive
their name from the fact that, if they do not emit radiation in the visible range, they appear black due to the complete absorption of all
wavelengths. The blackbody sources which are of interest to photovoltaics, emit light in the visible region. The spectral irradiance
from a blackbody is given by Planck's1 radiation law, shown in the following equation:

Planck's Radiation Law

Edit
Delete
Clone

Read more about Planck's Radiation Law

where:
λ is the wavelength of light;
T is the temperature of the blackbody (K);
F is the spectral irradiance in Wm-2µm-1; and
h,c and k are constants.

Getting the correct result requires care with the units. The simplest is to use SI units so that c is in m/s, h is in joule·seconds, T is in
kelvin, k is in joule/kelvin, and λ is in meters. This will give units of spectral irradiance in Wm-3. Dividing by 106 gives the
conventional units of spectral irradiance in Wm-2µm-1. The notation of F(λ) denotes that the spectral irradiance changes with
wavelength.

The total power density from a blackbody is determined by integrating the spectral irradiance over all wavelengths which gives:

Blackbody Power Density

Edit
Delete
Clone

Read more about Blackbody Power Density

where σ is the Stefan-Boltzmann constant and T is the temperature of the blackbody in kelvin.

An additional important parameter of a blackbody source is the wavelength where the spectral irradiance is the highest, or, in other
words the wavelength where most of the power is emitted. The peak wavelength of the spectral irradiance is determined by
differentiating the spectral irradiance and solving the derivative when it equals 0. The result is known as Wien's Law and is shown in
the following equation:

Wien's Law
Edit
Delete
Clone

Read more about Wien's Law

where λp is the wavelength where the peak spectral irradiance is emitted and
T is the temperature of the blackbody (K).

Peak Wavelength, Radiation Intensity Calculator

Edit
Delete
Clone

Black Body Temperature - Peak Wavelength, Radiation Intensity Calculator

Blackbody Temperature (K)

Peak Wavelength (µm )0.9667
Integrated Radiation Intensity (W/m2)4592700

Drag the slider at the bottom of the graph to see the change in the blackbody radiation spectrum as the temperature is increased from
1000 to 6000 K. MATLAB/Octave Code.
 Blackbody Temperature: 2999.2 K

Moving the slider to higher temperature causes a substantial increase in the emission and the peak to shift to shorter wavelengths.

The above equations and animation show that as the temperature of a blackbody increases, the spectral distribution and power of light
emitted change. For example, near room temperature, a blackbody emitter (such as a human body or light bulb which is turned off) will
emit low power radiation at wavelengths predominantly greater than 1µm, well outside the visual range of human observation. If the
blackbody is heated to 3000 K, it will glow red because the spectrum of emitted light shifts to higher energies and into the visible
spectrum. If the temperature of the filament is further increased to 6000 K, radiation is emitted at wavelengths across the visible
spectrum from red to violet and the light appears white. The graphs below compare the spectral irradiance of a blackbody at these three
temperatures. The room temperature case of 300K (the black dotted line) has essentially no power emitted in the visible and near
infrared portions of the spectrum shown on the graph. Because of the huge variation in both emitted power and the range of
wavelengths over which the power is emitted, the log graph below demonstrates more clearly the variation in the emitted blackbody
spectrum as a function of temperature.

Spectral intensity of light emitted from a black body on a log-log scale. At room temperature the emission is very low and centered
around 10 µm.
 Blackbody temperature: 1000.0 K

1. M. Planck, “Distribution of energy in the spectrum”, Annalen der Physik, vol. 4, pp. 553-563, 1901.

Solar Radiation
Add new comment
Español
한국어

We have so far described light sources in a general sense. In this section we describe the most common light source being that of the
sun. Using the terminology of the previous section we note that the sun can be described as a blackbody with a temperature of 6000 K.

The Sun
Add new comment
2 comment(s)
Español
한국어

The sun is a hot sphere of gas whose internal temperatures reach over 20 million kelvin due to nuclear fusion reactions at the sun's core
which convert hydrogen to helium. The radiation from the inner core is not visible since it is strongly absorbed by a layer of hydrogen
atoms closer to the sun's surface. Heat is transferred through this layer by convection1.
The surface of the sun, called the photosphere, is at a temperature of about 6000K and closely approximates a blackbody (see graph).
For simplicity, the 6000 K spectrum is commonly used in detailed balance calculations but temperatures of 5762 ± 50 K 2 and 5730 ±
90 K3 have also been proposed as a more accurate fit to the sun's spectrum. Astronomers use 5778 K when classifying the sun as a
star. For consistency in this site we use the approximation of 5800 K.

Sun Radiation
Edit
Delete
Clone

Sun Radiation Calculator

T= K
λpeak= µm H=
W/m2

Español

Using the equation above, and a temperature of 5800 K, gives a surface luminosity of Hsun = 64 x 106 W/m2. The total power emitted
by the sun is calculated by multiplying the emitted power density by the surface area of the sun. The sun has a radius of 695 x 106 m 4
giving a surface area of 6.07 x 1018 m2. Thus the total power output of the sun is 64 x 10 times 6.09 x 1018 m2, which is equal to 3.9 x
1026 watts. Clearly an enormous amount of power when we consider that the entire world’s energy usage is only 16 TW.
Current image of the sun (updated every few hours) from SOHO.

The total power emitted from the sun is composed not of a single wavelength, but is composed of many wavelengths and therefore
appears white or yellow to the human eye. These different wavelengths can be seen by passing light through a prism, or water droplets
in the case of a rainbow. Different wavelengths show up as different colours, but not all the wavelengths can be seen since some are
"invisible" to the human eye.

1. S. M. Hanasoge, Duvall, T. L., and Sreenivasan, K. R., “From the Cover: Anomalously weak solar convection”, Proceedings
of the National Academy of Sciences, vol. 109, no. 30, pp. 11928 - 11932, 2012.
2. C. E. Backus, Solar Cells. New York: IEEE, 1976, p. 512.
3. J. E. Parrott, “Choice of an equivalent black body solar temperature”, Solar Energy, vol. 51, pp. 195 - 195, 1993.
4. M. Emilio, Kuhn, J. R., Bush, R. I., and Scholl, I. F., “MEASURING THE SOLAR RADIUS FROM SPACE DURING THE
2003 AND 2006 MERCURY TRANSITS”, The Astrophysical Journal, vol. 750, no. 2, p. 135, 2012.

Solar Radiation in Space

Add new comment
4 comment(s)
Español
한국어

Only a fraction of the total power emitted by the sun impinges on an object in space which is some distance from the sun. The
solar irradiance (H0 in W/m2) is the power density incident on an object due to illumination from the sun. At the sun's surface, the
power density is that of a blackbody at about 6000K and the total power from the sun is this value multiplied by the sun's surface area.
However, at some distance from the sun, the total power from the sun is now spread out over a much larger surface area and therefore
the solar irradiance on an object in space decreases as the object moves further away from the sun.
The solar irradiance on an object some distance D from the sun is found by dividing the total power emitted from the sun by the
surface area over which the sunlight falls. The total solar radiation emitted by the sun is given by σT4 multiplied by the surface area of
the sun (4πR2sun) where Rsun is the radius of the sun. The surface area over which the power from the sun falls will be 4πD2. Where D
is the distance of the object from the sun. Therefore, the solar radiation intensity, H0 in (W/m2), incident on an object is:

Solar Radiation
Edit
 Delete
Clone

Read more about Solar Radiation

where:
Hsun is the power density at the sun's surface (in W/m2) as determined by Stefan-Boltzmann's blackbody equation;
Rsun is the radius of the sun in meters as shown in the figure below; and
D is the distance from the sun in meters as shown in the figure below.

At a distance, D, from the sun the same amount of power is spread over a much wider area so the solar radiation power intensity is
reduced. In the calculation below the radiant solar intensity at the sun’s surface, Hsun is 64 × 106 W/m2 and the radius of the sun, Rsun,
is 695 × 106 m.

Solar Radiation
Edit
Delete
Clone

Distance from the Sun - Radiation Intensity Calculator

Distance from the Sun, D x109m

Sunlight Intensity, H0 W/m2

The table below gives standardised values for the radiation at each of the planets but by entering the distance you can obtain an
approximation. The distance to the sun varies for each planet since the orbits are elliptical not linear. For instance, the earth sun
distance is defined by the International Astronomical Union as 149,597,870,700 (149.5978707 × 109) meters1 but the actual distance
to the sun varies from 152.10 × 109 m to 147.10 × 109 m.

Mean Solar
Planet Distance (x 109 m)
Irradiance (W/m2)
Mercury 57 9116.4
Venus 108 2611.0

Earth 150 1366.1

Mars 227 588.6

Jupiter 778 50.5

Saturn 1426 15.04

Uranus 2868 3.72

Neptune 4497 1.51

 Pluto 5806 0.878

Further details on the planets are at: https://nssdc.gsfc.nasa.gov/planetary/factsheet/

1. I. Astronomic Union, “Measuring the Universe”, 2012. .

Solar Radiation Outside the Earth's Atmosphere

Add new comment
2 comment(s)
Español
한국어

The solar radiation outside the earth's atmosphere is calculated using the radiant power density (Hsun) at the sun's surface (5.961 x 107
W/m2), the radius of the sun (Rsun), and the distance between the earth and the sun. The calculated solar irradiance at the Earth's
atmosphere is about 1.36 kW/m2. The geometrical constants used in the calculation of the solar irradiance incident on the Earth are
shown in the figure below.

Geometrical constants for finding the Earth's solar irradiance. The diameter of the Earth is not needed but is included for the sake of
completeness.

The actual power density varies slightly since the Earth-Sun distance changes as the Earth moves in its elliptical orbit around the sun,
and because the sun's emitted power is not constant. The power variation due to the elliptical orbit is about 3.4%, with the largest solar
irradiance in January and the smallest solar irradiance in July. An equation 1 which describes the variation throughout the year just
outside the earth's atmosphere is:

where:
H is the radiant power density outside the Earth's atmosphere (in W/m2);
Hconstant is the value of the solar constant, 1.353 kW/m2; and
n is the day of the year.

These variations are typically small and for photovoltaic applications the solar irradiance can be considered constant. The value of the
solar constant and its spectrum have been defined as a standard value called air mass zero (AM0) and takes a value of 1.366 kW/m2.
The spectral irradiance is given in the Appendix.
 Day of the Year - Sunlight Intensity Calculator

Day of the Year

Sunlight Intensity W/m2

1. G. D. Rai, “Solar Energy Utilisation”, Khanna Publishers, 1980, p. 44.

Solar Radiation at the Earth's Surface

Add new comment
Español
한국어

While the solar radiation incident on the Earth's atmosphere is relatively constant, the radiation at the Earth's surface varies widely due
to:

atmospheric effects, including absorption and scattering;

local variations in the atmosphere, such as water vapour, clouds, and pollution;
latitude of the location; and
the season of the year and the time of day.

The above effects have several impacts on the solar radiation received at the Earth's surface. These changes include variations in the
overall power received, the spectral content of the light and the angle from which light is incident on a surface. In addition, a key
change is that the variability of the solar radiation at a particular location increases dramatically. The variability is due to both local
effects such as clouds and seasonal variations, as well as other effects such as the length of the day at a particular latitude. Desert
regions tend to have lower variations due to local atmospheric phenomena such as clouds. Equatorial regions have low variability
between seasons.

Solar radiation at the Earth's surface varies from the solar radiation incident on the Earth's atmosphere. Cloud cover, air pollution,
latitude of a location, and the time of the year can all cause variations in solar radiance at the Earth's surface.

The amount of energy reaching the surface of the Earth every hour is greater than the amount of energy used by the Earth's population
over an entire year.

PV Lighthouse hosts Altermatt's lectures on the solar spectrum.

Atmospheric Effects
Add new comment
1 comment(s)
Español
한국어

Atmospheric effects have several impacts on the solar radiation at the Earth's surface. The major effects for photovoltaic applications
are:

a reduction in the power of the solar radiation due to absorption, scattering and reflection in the atmosphere;
a change in the spectral content of the solar radiation due to greater absorption or scattering of some wavelengths;
the introduction of a diffuse or indirect component into the solar radiation; and
local variations in the atmosphere (such as water vapor, clouds and pollution) which have additional effects on the incident
power, spectrum and directionality.

These effects are summarized in the figure below.

Typical clear sky absorption and scattering of incident sunlight 1.

Absorption in the Atmosphere

As solar radiation passes through the atmosphere, gasses, dust and aerosols absorb the incident photons. Specific gasses, notably ozone
(O3), carbon dioxide (CO2), and water vapor (H2O), have very high absorption of photons that have energies close to the bond energies
of these atmospheric gases. This absorption yields deep troughs in the spectral radiation curve. For example, much of the far infrared
light above 2 µm is absorbed by water vapor and carbon dioxide. Similarly, most of the ultraviolet light below 0.3 µm is absorbed by
ozone (but not enough to completely prevent sunburn!).

While the absorption by specific gasses in the atmosphere change the spectral content of the terrestrial solar radiation, they have a
relatively minor impact on the overall power. Instead, the major factor reducing the power from solar radiation is the absorption and
scattering of light due to air molecules and dust. This absorption process does not produce the deep troughs in the spectral irradiance,
but rather causes a power reduction dependent on the path length through the atmosphere. When the sun is overhead, the absorption
due to these atmospheric elements causes a relatively uniform reduction across the visible spectrum, so the incident light appears
white. However, for longer path lengths, higher energy (lower wavelength) light is more effectively absorbed and scattered. Hence in
the morning and evening the sun appears much redder and has a lower intensity than in the middle of the day.

Show Blackbody Show AM0 Show AM1.5

Show Visible Wavelengths Show Notes

A comparison of solar radiation outside the Earth's atmosphere with the amount of solar radiation reaching the Earth itself. The human
eye has evolved to the point where sensitivity is greatest at the most intense wavelengths 2.

The standard spectra given above and described in more detail in the appendices give a typical spectra for a sunlight. Computers
models allow for more detailed models of the solar spectra for a particular location and time of day. The Simple Model of the
Atmospheric Radiative Transfer of Sunshine, or SMARTS,3 is used to generate the standard solar spectra. The solar spectrum
calculator at PV lighthouse also gives the solar spectrum as a function of location and time of the day. It uses a slightly simpler
algorithm from Bird4

Direct and Diffuse Radiation Due to Scattering of Incident Light

Light is absorbed as it passes through the atmosphere and at the same time it is subject to scattering. One of the mechanisms for light
scattering in the atmosphere is known as Rayleigh scattering which is caused by molecules in the atmosphere. Rayleigh scattering is
particularly effective for short wavelength light (that is blue light) since it has a λ-4 dependence. In addition to Rayleigh scattering,
aerosols and dust particles contribute to the scattering of incident light known as Mie scattering.

Scattered light is undirected, and so it appears to be coming from any region of the sky. This light is called "diffuse" light. Since diffuse
light is primarily "blue" light, the light that comes from regions of the sky other than where the sun is, appears blue. In the absence of
scattering in the atmosphere, the sky would appear black, and the sun would appear as a disk light source. On a clear day, about 10% of
the total incident solar radiation is diffuse.

Effect of clouds and other local variations in the atmosphere

The final effect of the atmosphere on incident solar radiation is due to local variations in the atmosphere. Depending on the type of
cloud cover, the incident power is severely reduced. An example of heavy cloud cover is shown below.

Relative output current from a photovoltaic array on a sunny and a cloudy winter's day in Melbourne with an array tilt angle of 60° 5.

1. C. Hu and White, R. M., Solar Cells: From Basic to Advanced Systems. New York: McGraw-Hill, 1983.
2. R. Sekuler and Blake, R., Perception. New York: Alfred A. Knopf Inc, 1985.
3. C. Gueymard, SMARTS2: a simple model of the atmospheric radiative transfer of sunshine: algorithms and performance
assessment. Florida Solar Energy Center Cocoa, FL, 1995.
4. R. E. Bird and Riordan, C., “Simple Solar Spectral Model for Direct and Diffuse Irradiance on Horizontal and Tilted Planes at
the Earth's Surface for Cloudless Atmospheres”, Journal of Climate and Applied Meteorology, vol. 25, no. 1, pp. 87 - 97, 1986.
5. M. Mack, “Solar Power for Telecommunications”, The Telecommunication Journal of Australia, vol. 29, pp. 20-44, 1979.

Air Mass
Add new comment
Español
한국어

The Air Mass is the path length which light takes through the atmosphere normalized to the shortest possible path length (that is, when
the sun is directly overhead). The Air Mass quantifies the reduction in the power of light as it passes through the atmosphere and is
absorbed by air and dust. The Air Mass is defined as:

Air Mass Zenith

Edit
Delete
Clone

Read more about Air Mass Zenith

where θ is the angle from the vertical (zenith angle). When the sun is directly overhead, the Air Mass is 1.

Air Mass Zenith

Edit
Delete
Clone
 Sun Position - Air
Mass Calculator

Angle from vertical, θ

°

Air mass, AM:

(units)

The air mass represents the proportion of atmosphere that the light must pass through before striking the Earth relative to
its overhead path length, and is equal to Y/X.

A more detailed model showing the effect of airmass on the solar spectrum is available at the PV Lighthouse Solar Spectrum
Calculator.

An easy method to determine the air mass is from the shadow of a vertical pole.

Air mass is the length of the hypotenuse divided by the object height h, and from Pythagoras's theorem we get:

Air Mass Shadow

Edit
Delete
Clone

Read more about Air Mass Shadow

Air Mass Shadow

Edit
Delete
Clone
 Object & It's Shadow - Air Mass Calculator

Enter the object height, h (distance units) Enter the shadow length. s (distance units)

Air Mass, AM:1.4142 (units)

The above calculation for air mass assumes that the atmosphere is a flat horizontal layer, but because of the curvature of the
atmosphere, the air mass is not quite equal to the atmospheric path length when the sun is close to the horizon. At sunrise, the angle of
the sun from the vertical position is 90° and the air mass is infinite, whereas the path length clearly is not. An equation which
incorporates the curvature of the earth is1:

Air Mass Curvature

Edit
Delete
Clone

Read more about Air Mass Curvature

Standardised Solar Spectrum and Solar Irradiation

The efficiency of a solar cell is sensitive to variations in both the power and the spectrum of the incident light. To facilitate an accurate
comparison between solar cells measured at different times and locations, a standard spectrum and power density has been defined for
both radiation outside the Earth's atmosphere and at the Earth's surface.

The standard spectrum at the Earth's surface is called AM1.5G, (the G stands for global and includes both direct and diffuse radiation)
or AM1.5D (which includes direct radiation only). The intensity of AM1.5D radiation can be approximated by reducing the AM0
spectrum by 28% (18% due to absorption and 10% to scattering). The global spectrum is 10% higher than the direct spectrum. These
calculations give approximately 970 W/m2 for AM1.5G. However, the standard AM1.5G spectrum has been normalized to give
1kW/m2 due to the convenience of the round number and the fact that there are inherently variations in incident solar radiation. The
standard spectrum is listed in the Appendix page.

The standard spectrum outside the Earth's atmosphere is called AM0, because at no stage does the light pass through the atmosphere.
This spectrum is typically used to predict the expected performance of cells in space.

Intensity Calculations Based on the Air Mass

The intensity of the direct component of sunlight throughout each day can be determined as a function of air mass from the
experimentally determined equation 2:

Intensity Sea Level

Edit
Delete
Clone

Read more about Intensity Sea Level

where ID is the intensity on a plane perpendicular to the sun's rays in units of kW/m2 and AM is the air mass. The value of 1.353
kW/m2 is the solar constant and the number 0.7 arises from the fact that about 70% of the radiation incident on the atmosphere is
transmitted to the Earth. The extra power term of 0.678 is an empirical fit to the observed data and takes into account the non-
uniformities in the atmospheric layers.

Sunlight intensity increases with the height above sea level. The spectral content of sunlight also changes making the sky 'bluer' on
high mountains. Much of the southwest of the United States is two kilometers above sea level, adding significantly to solar isolation. A
simple empirical fit to observed data 3 and accurate to a few kilometers above sea level is given by:
Intensity
Edit
Delete
Clone

Read more about Intensity

where a = 0.14 and h is the location height above sea level in kilometers.

Even on a clear day, the diffuse radiation is still about 10% of the direct component. Thus on a clear day the global irradiance on a
module perpendicular to the sun's rays is:

Global Irradiance
Edit
Delete
Clone

Read more about Global Irradiance

Radiation Intensity
Edit
Delete
Clone

Air Mass - Radiation Intensity Calculator

Air Mass, AM (units)

Direct beam intensity, ID (kW/m2) Estimate of global irradiance, IG (kW/m2)

1. F. Kasten and Young, A. T., “Revised optical air mass tables and approximation formula”, Applied Optics, vol. 28, pp.
4735–4738, 1989.
2. A. B. Meinel and Meinel, M. P., Applied Solar Energy. Addison Wesley Publishing Co., 1976.
3. E. G. Laue, “The measurement of solar spectral irradiance at different terrestrial elevations”, Solar Energy, vol. 13, pp. 43 -
50, IN1-IN4, 51-57, 1970.

Motion of the Sun

Add new comment
3 comment(s)
Español
한국어

The apparent motion of the sun, caused by the rotation of the Earth about its axis, changes the angle at which the direct component of
light will strike the Earth. From a fixed location on Earth, the sun appears to move throughout the sky. The position of the sun depends
on the location of a point on Earth, the time of day and the time of year. This apparent motion of the sun is shown in the figure below..
Path of the sun in the southern hemisphere.

This apparent motion of the sun has a major impact on the amount of power received by a solar collector. When the sun's rays are
perpendicular to the absorbing surface, the power density on the surface is equal to the incident power density. However, as the angle
between the sun and the absorbing surface changes, the intensity on the surface is reduced. When the module is parallel to the sun's
rays (and the angle to the module normal = 90°) the intensity of light essentially falls to zero. For intermediate angles, the relative
power density is cos(θ) where θ is the angle between the sun's rays and the module normal.

Drag the slider to examine the impact of changing the angle between the absorbing surface and the incident light.

The angle between the sun and a fixed location on Earth depends on the particular location (the longitude of the location), the time of
year and the time of day. In addition, the time at which the sun rises and sets depends on the longitude of the location. Therefore,
complete modeling of the sun's angle to a fixed position on Earth requires the latitude, longitude, day of the year, and time of day. This
is discussed in the following pages.

Solar Time
Add new comment
6 comment(s)
Español
한국어
Until the late 19th century most people used local solar time so that noon was when the sun was directly overhead, and each town had
its own definition. Transport was slow, so it did not matter that the time in a town miles away varied by a few minutes. The advent of
railways necessitated more accurate time keeping and time zones were introduced to keep an entire region on the same time1. Time
zones follow political boundaries so that local time may be up to 2 hours different from solar time.

Local Solar Time (LST) and Local Time (LT)

Twelve noon local solar time (LST) is defined as when the sun is highest in the sky. Local time (LT) usually varies from LST because
of the eccentricity of the Earth's orbit, and because of human adjustments such as time zones and daylight saving.

Local Standard Time Meridian (LSTM)

The Local Standard Time Meridian (LSTM) is a reference meridian used for a particular time zone and is similar to the Prime
Meridian, which is used for Greenwich Mean Time. The LSTM is illustrated below.

The (LSTM) is calculated according to the equation:

where ΔTUTC is the difference of the Local Time (LT) from Universal Coordinated Time (UTC) in hours. ΔTUTC is also equal to the
time zone. 15°= 360°/24 hours. For instance, Sydney Australia is UTC +10 so the Local Standard Time Meridian is 150 °E. Phoenix,
USA is UTC -7 so the LSTM is 105 °W

Equation of Time (EoT)

The equation of time (EoT) (in minutes) is an empirical equation that corrects for the eccentricity of the Earth's orbit and the Earth's
axial tilt. An approximation 2 accurate to within ½ minute is:

where

$$B=\frac{360}{365}(d-81)$$

in degrees and d is the number of days since the start of the year. The time correction EoT is plotted in the figure below.
Time Correction Factor (TC)

The net Time Correction Factor (in minutes) accounts for the variation of the Local Solar Time (LST) within a given time zone due to
the longitude variations within the time zone and also incorporates the EoT above.

The factor of 4 minutes comes from the fact that the Earth rotates 1° every 4 minutes.

Local Solar Time (LST)

The Local Solar Time (LST) can be found by using the previous two corrections to adjust the local time (LT).

Hour Angle (HRA)

The Hour Angle converts the local solar time (LST) into the number of degrees which the sun moves across the sky. By definition, the
Hour Angle is 0° at solar noon. Since the Earth rotates 15° per hour, each hour away from solar noon corresponds to an angular motion
of the sun in the sky of 15°. In the morning the hour angle is negative, in the afternoon the hour angle is positive.

Local Time to Local Solar Time Calculator

The conversion of LT to LST is complicated by the world’s time zones and there is not a simple conversion from a location to its time
zone. The calculator below uses Google to determine the time zone from its location. It will accept any description such as an address,
city name, airport code and then show its location on a map. It also determines the most up to date time zone for that location. The
most commonly used location is assumed so that simply “Paris” will be Paris, France whereas “Paris, Texas” will be Paris in the US
state of Texas.
Enter a location above to get its longitude and time zone. Once the time zone is determined, the latitude is not needed when calculating
the local solar time. The longitude and time zone can also be entered directly without having to use the lookup. Standard time is used
and there is no correction for Daylight Savings.

Local Solar Time

Input Parameters

longitude timezone Day of year, Local Time - Standard

(hours) Local Time (min)

Results

Equation of Time X minutes.

Local Standard Time Meridian: X °
Time Correction: X minutes
Hour Angle X °
Local Solar Time X (HH:MM)
latitude X ° (not needed)

1. E. Zerubavel, “The Standardization of Time: A Sociohistorical Perspective”, American Journal of Sociology, vol. 88, pp. 1-23,
1982.
2. R. M. Milne, “Note on the Equation, of Time”, The Mathematical Gazette, vol. 10, no. 155, pp. 372 - 375, 1921.

Declination Angle
Add new comment
1 comment(s)
Español
한국어

The declination angle, denoted by δ, varies seasonally due to the tilt of the Earth on its axis of rotation and the rotation of the Earth
around the sun. If the Earth were not tilted on its axis of rotation, the declination would always be 0°. However, the Earth is tilted by
23.45° and the declination angle varies plus or minus this amount. Only at the spring and fall equinoxes is the declination angle equal
to 0°. The rotation of the Earth around the sun and the change in the declination angle is shown in the animation below.
Animation showing how the tilt angle changes from the summer solstice in the northern hemisphere (or winter in the southern
hemisphere) to the northern hemisphere winter solstice (summer in the south).

The declination of the sun is the angle between the equator and a line drawn from the centre of the Earth to the centre of the sun. The
seasonal variation of the declination angle is shown in the animation below.

Despite the fact that the Earth revolves around the sun, it is simpler to think of the sun revolving around a stationary Earth. This
requires a coordinate transformation. Under this alternative coordinate system, the sun moves around the Earth.

The declination angle can be calculated by the equation 1:

Declination Angle
Edit
Delete
Clone

Read more about Declination Angle

where d is the day of the year with Jan 1 as d = 1

The declination is zero at the equinoxes (March 22 and September 22), positive during the northern hemisphere summer and negative
during the northern hemisphere winter. The declination reaches a maximum of 23.45° on June 22 (summer solstice in the northern
hemisphere) and a minimum of -23.45° on December 21-22 (winter solstice in the northern hemisphere). In the equation above, the
+10 comes from the fact that the winter solstice occurs before the start of the year. The equation also assumes that the suns orbit is a
perfect circle and the factor of 360/365 converts the day number to a position in the orbit.

Alternate Formulations

The declination angle can also be defined in other ways. The equations:
Declination Angle (alternate)
Edit
Delete
Clone

Read more about Declination Angle (alternate)

are also often used in the literature. They are referred to the equinoxes instead so use sin instead of cos.

Finally, there are many algorithms for more accurate determinations of declination angle to account for the elliptic and yearly
movement of the earth's orbit. These are only needed for concentrators that require more accurate tracking of the sun.

For instance the SPA algorithm 2(http://www.psa.es/sdg/sunpos.htm) uses:

dOmega=2.1429-0.0010394594*dElapsedJulianDays;

dMeanLongitude = 4.8950630+ 0.017202791698*dElapsedJulianDays;

// Radians dMeanAnomaly = 6.2400600+ 0.0172019699*dElapsedJulianDays;

dEclipticLongitude = dMeanLongitude + 0.03341607*sin( dMeanAnomaly ) + 0.00034894*sin( 2*dMeanAnomaly )-0.0001134

-0.0000203*sin(dOmega);

dEclipticObliquity = 0.4090928 - 6.2140e-9*dElapsedJulianDays +0.0000396*cos(dOmega);

dSin_EclipticLongitude= sin( dEclipticLongitude );

dY = cos( dEclipticObliquity ) * dSin_EclipticLongitude;

dX = cos( dEclipticLongitude );

dRightAscension = atan2( dY,dX ); if( dRightAscension < 0.0 ) dRightAscension = dRightAscension + twopi;

dDeclination = asin( sin( dEclipticObliquity )*dSin_EclipticLongitude );

dElapsedJulian days is the number of days since January 1, 2000 and dDeclination is the resulting declination.

The various ways of calculating the declination angle are graphed below using declination.py python code.
<

1. P. I. Cooper, “The absorption of radiation in solar stills”, Solar Energy, vol. 12, pp. 333 - 346, 1969.
2. M. Blanco-Muriel, Alarcón-Padilla, D. C., López-Moratalla, T., and Lara-Coira, M. Í., “Computing the solar vector”, Solar
Energy, vol. 70, no. 5, pp. 431 - 441, 2001.

Elevation Angle
Add new comment
1 comment(s)
Español
한국어

The elevation angle (used interchangeably with altitude angle) is the angular height of the sun in the sky measured from the horizontal.
Confusingly, both altitude and elevation are also used to describe the height in meters above sea level. The elevation is 0° at sunrise
and 90° when the sun is directly overhead (which occurs for example at the equator on the spring and fall equinoxes).

The elevation angle varies throughout the day. It also depends on the latitude of a particular location and the day of the year.
An important parameter in the design of photovoltaic systems is the maximum elevation angle, that is, the maximum height of the sun
in the sky at a particular time of year. This maximum elevation angle occurs at solar noon and depends on the latitude and declination
angle as shown in the figure below.

The maximum elevation angle at solar noon (α) is a function of latitude and the declination angle (δ).

From the previous figure, a formula for the elevation angle at solar noon can be determined according to the formula:

When the equation above gives a number greater than 90° then subtract the result from 180°. It means the sun at solar noon is coming
from the south as is typical the northern hemisphere.

where:
φ is the latitude of the location of interest (+ve for the northern hemisphere and -ve for the southern hemisphere).
δ is the declination angle, which depends on the day of the year.

At the Tropic of Cancer on the summer solstice, the sun is directly overhead and the elevation angle is 90°. In summer at latitudes
between the equator and the Tropic of Cancer, the elevation angle at solar noon is greater than 90°, implying that the sunlight is coming
from the north rather than from the south as in most of the northern hemisphere. Similarly, at latitudes between the equator and the
Tropic of Capricorn, during some periods of the year, sunlight is incident from the south, rather than from the north.

While the maximum elevation angle is used even in very simple PV system design, more accurate PV system simulation requires the
knowledge of how the elevation angle varies throughout the day. These equations are given in the following page.
The elevation, α, can be found using the following formula:

Elevation Angle
Edit
Delete
Clone

Read more about Elevation Angle

where HRA is the hour angle

Zenith Angle

The zenith angle is the angle between the sun and the vertical. The zenith angle is similar to the elevation angle but it is measured from
the vertical rather than from the horizontal, thus making the zenith angle = 90° - elevation.

Zenith Angle
Edit
Delete
Clone

Read more about Zenith Angle

Sunrise and Sunset

To calculate the sunrise and sunset time the elevation is set to zero and the elevation equation above is rearranged to give:

Sunrise Time
Edit
Delete
Clone

Read more about Sunrise Time

and sunset:

Sunset Time
 Edit
Delete
Clone

Read more about Sunset Time

these equations can be simplified as:

Sunrise Time (Simplified)

Edit
Delete
Clone

Read more about Sunrise Time (Simplified)

Sunset Time (simplified)

Edit
Delete
Clone

Read more about Sunset Time (simplified)

where TC is the time correction.

Azimuth Angle
Add new comment
4 comment(s)
Español
한국어

The azimuth angle is the compass direction from which the sunlight is coming. At solar noon, the sun is always directly south in the
northern hemisphere and directly north in the southern hemisphere. The azimuth angle varies throughout the day as shown in the
animation below. At the equinoxes, the sun rises directly east and sets directly west regardless of the latitude, thus making the azimuth
angles 90° at sunrise and 270° at sunset. In general however, the azimuth angle varies with the latitude and time of year and the full
equations to calculate the sun's position throughout the day are given on the following page.
The azimuth angle is like a compass direction with North = 0° and South = 180°. Other authors use a variety of slightly different
definitions (i.e., angles of ± 180° and South = 0°).

The azimuth is calculated from the above parameters:

Azimuth
Edit
Delete
Clone

$$Azimuth=cos^{-1}[\frac{sin\delta cos\varphi-cos\delta sin \varphi cos(HRA)}{cos\alpha}]$$

Read more about Azimuth

where α is the elevation, Φ is the latitude, and δ is the declination.

The above equation only gives the correct azimuth in the solar morning so that:

Azimuth = Azi, for LST <12 or HRA < 0

Azimuth = 360° - Azi, for LST > 12 or HRA >0

The Sun's Position

Add new comment
4 comment(s)
Español
한국어

The azimuth angle and the elevation angle at solar noon are the two key angles which are used to orient photovoltaic modules.
However, to calculate the sun's position throughout the day, both the elevation angle and the azimuth angle must be calculated
throughout the day. These angles are calculated using "solar time". In conventional time keeping, regions of the Earth are divided into
certain time zones. However, in these time zones, noon does not necessarily correspond to the time when the sun is highest in the sky.
Similarly, sunrise is defined as the stage when the sun rises in one part of the time zone. However, due to the distance covered in a
single time zone, the time at which the sun actually clears the horizon in one part of the time zone may be quite different to the
"defined" sunrise (or what is officially recognized as the time of sunrise). Such conventions are necessary otherwise a house one block
away from another would actually be different in time by several seconds. Solar time, on the other hand is unique to each particular
longitude. Consequently, to calculate the sun's position, first the local solar time is found and then the elevation and azimuth angles are
calculated.

Local Solar Time (LST) and Local Time (LT)

Twelve noon local solar time (LST) is defined as when the sun is highest in the sky. Local time (LT) usually varies from LST because
of the eccentricity of the Earth's orbit, and because of human adjustments such as time zones and daylight saving.

Local Standard Time Meridian (LSTM)

The Local Standard Time Meridian (LSTM) is a reference meridian used for a particular time zone and is similar to the Prime
Meridian, which is used for Greenwich Mean Time. The LSTM is illustrated below.

The (LSTM) is calculated according to the equation:

where ΔTGMT is the difference of the Local Time (LT) from Greenwich Mean Time (GMT) in hours.

Equation of Time (EoT)

The equation of time (EoT) (in minutes) is an empirical equation that corrects for the eccentricity of the Earth's orbit and the Earth's
axial tilt.

where

in degrees and d is the number of days since the start of the year. The time correction EoT is plotted in the figure below.
Sundials include the equation time as a table to provide the correct time throughout the year. The other time corrections are either
incorporated into the design of the sundial or given as a constant offset.

Time Correction Factor (TC)

The net Time Correction Factor (in minutes) accounts for the variation of the Local Solar Time (LST) within a given time zone due to
the longitude variations within the time zone and also incorporates the EoT above.

The factor of 4 minutes comes from the fact that the Earth rotates 1° every 4 minutes.

Local Solar Time (LST)

The Local Solar Time (LST) can be found by using the previous two corrections to adjust the local time (LT).
Hour Angle (HRA)

The Hour Angle converts the local solar time (LST) into the number of degrees which the sun moves across the sky. By definition, the
Hour Angle is 0° at solar noon. Since the Earth rotates 15° per hour, each hour away from solar noon corresponds to an angular motion
of the sun in the sky of 15°. In the morning the hour angle is negative, in the afternoon the hour angle is positive.

Declination

The declination angle has been previously given as:

Where d is the number of days since the start of the year.

Elevation and Azimuth

The elevation, α, is described on page: https://www.pveducation.org/pvcdrom/properties-of-sunlight/elevation-angle

Elevation Angle
Edit
Delete
Clone

Read more about Elevation Angle

Azimuth
Edit
Delete
Clone

$$Azimuth=cos^{-1}[\frac{sin\delta cos\varphi-cos\delta sin \varphi cos(HRA)}{cos\alpha}]$$

Read more about Azimuth

where: φ is the latitude.

Sun's position throughout the day

A calculator to find the elevation and azimuth throughout the year at any location is given on the following page.

Sun Position Calculator

Add new comment
Español
한국어

Using the equations on the previous page, the position of the sun in the sky can be determined from the observer's location and the time
of day. In the top blue squares, enter the observer's location and time of day. An alternate calculator for the sun's path is also available
at the PV Lighthouse Solar Path Calculator

Time and Date

The time is given as 24 hour time and the minutes are entered separately. Thus for 5:15 pm, enter 17 in the hour box and 15 in the
minute box.

Longitude, Latitude and Time Zone (GMT)

Longitude, latitude and time zone of locations throughout the world are available at www.timeanddate.com. Minutes of longitude and
latitude are entered as fractions, so 17° 30' becomes 17.5. Enter locations with a west longitude as negative. For daylight saving
(summer time), subtract 1 hour from the given values. Generally speaking, locations east of Greenwich (UK) are positive and locations
west of Greenwich are negative.
There are other services that will determine your approximate latitude and longitude from your IP address. For example
ip2location.com. For a given latitude and longitude http://timezonedb.com/ will calculate the GMT offset.

The graph on the right shows the position of the sun's azimuth and elevation angles projected onto a two-dimensional plane. An
elevation angle of 90° corresponds to the stage when the sun is directly overhead and appears in the centre of the graph. An elevation
angle of 0° corresponds to the point when the sun is on the horizon, and appears on the outer edge of the graph. The azimuth angles are
marked around the graph's edge, so an azimuth angle of 0° is at the top of the graph. The graph is best understood by trying a number
of times and locations and seeing where the azimuth and elevations are plotted(expand to window).

Explanation of polar plots

The most convenient way to plot the sun's position throughout the day is on a polar plot. Click on the animation for an explanation of
these polar plots.
Sun's Position to High Accuracy
Add new comment
1 comment(s)
Español
한국어

The algorithms presented so far are accurate to within about 1° and are sufficient for most terrestrial photovoltaic applications. For flat
plate modules the siting is only accurate to a few degrees and the errors introduced by the simple algorithms are negligible when
compared to the unknown factors at the location such as atmosphere effects. For concentrator modules, where the modules track the
sun and focus the light, the simple equations introduce an unacceptable degree of error. As the concentration increases so does the need
for sun tracking accuracy. For systems with concentration ratios of 1000:1 the sun must be tracked to within 3.5 minutes (0.06°) of arc
1. One possibility for tracking the sun is to use a look up table based on the Astronomical Almanac 2 or the computer implementation,
the Multiyear Interactive Computer Almanac (MICA) available from the United States Naval Observatory. However, such a system
would be unwieldy for the microcontrollers used in tracking systems. There are numerous algorithms developed for sun tracking with a
trade off between accuracy and complexity. Before the advent of modern computers, the emphasis was on simplicity but now even
small microcontrollers can handle quite complex mathematical formula.

Blanco-Muriel et al. 3 from the Plataforma Solar de Almerýa (PSA) review the accuracy of all the algorithms. Further they develop a
simplified algorithm that is accurate to within 0.5 minutes of arc for the year 1999-2015. The PSA algorithm has been specially
optimised in C++ code for microcontrollers and is available at http://www.psa.es/sdg/sunpos.htm. The code has been converted to
work with the PVCDROM and is presented below. The PSA algorithm is also implemented the PV Lighthouse Solar Path Calculator.

A further refinement is available was reported by NREL4 with an online implementation available at: http://www.nrel.gov/midc/solpos
/spa.html

PSA algorithm for High Accuracy Tracking of the Sun

The PSA algorithm uses Universal Time (UT) to remove the uncertainty caused by local time zones. The location is entered using
longitude and latitude with the minutes and seconds converted to fractions of a degree. The azimuth angle is measured from true north
not magnetic north and the zenith angle is measured from the vertical. The elevation angle is measured from the horizontal.

The default values of midday on 1st January 2003, with a longitude and latitude of 0° (a location off East Africa) give an azimuth of
178°, denoting that the sun is almost due south. The zenith angle shows the sun is high in the sky but 23° from being directly overhead.
 PSA Algorithm Sun's Position Calculator

Year Month Day

The date displayed is UTC, i.e. at Greenwich. Hour

Minute Second Longitude =
Latitude =

Azimuth degrees Zenith

degrees Elevation
degrees

1. L. L. Vant-Hull and Hildebrandt, A. F., “Solar thermal power system based on optical transmission”, Solar Energy, vol. 18,
pp. 31 - 39, 1976.
2. U. S. Naval Observatory, “The Astronomical Almanac”. 2020.
3. M. Blanco-Muriel, Alarcón-Padilla, D. C., López-Moratalla, T., and Lara-Coira, M. Í., “Computing the solar vector”, Solar
Energy, vol. 70, pp. 431 - 441, 2001.
4. I. Reda and Andreas, A., “Solar Position Algorithm for Solar Radiation Applications”, 2003.

Solar Radiation on a Tilted Surface

Add new comment
8 comment(s)
Español
한국어

The power incident on a PV module depends not only on the power contained in the sunlight, but also on the angle between the module
and the sun. When the absorbing surface and the sunlight are perpendicular to each other, the power density on the surface is equal to
that of the sunlight (in other words, the power density will always be at its maximum when the PV module is perpendicular to the sun).
However, as the angle between the sun and a fixed surface is continually changing, the power density on a fixed PV module is less than
that of the incident sunlight.

The amount of solar radiation incident on a tilted module surface is the component of the incident solar radiation which is
perpendicular to the module surface. The following figure shows how to calculate the radiation incident on a tilted surface (Smodule)
given either the solar radiation measured on horizontal surface (Shoriz) or the solar radiation measured perpendicular to the sun
(Sincident).

Tilting the module to the incoming light reduces the module output.
The animation shows the calculation of the various insolations. In each case the length of the vector gives the relative intensity of the
radiation.

The equations relating Smodule, Shoriz and Sincident are:

where
α is the elevation angle; and
β is the tilt angle of the module measured from the horizontal.

The elevation angle has been previously given as:

where φ is the latitude; and

δ is the declination angle previously given as:

where d is the day of the year. Note that from simple math (284+d) is equivalent to (d-81) which was used before. Two equations are
used interchangeably in literature.

From these equations a relationship between Smodule and Shoriz can be determined as:

The following active equations show the calculation of the incident and horizontal solar radiation and that on the module. Enter only
one of Smodule, Shoriz and Sincident and the program will calculate the others.
 Components of Radiation on Tilted Surface Calculator

Array Tilt, β = degrees. Latitude, φ = degrees. Hemisphere: North South Day Number, d =

Declination, δ= degrees Sun Angle, α = degrees.

Shoriz = Sincident =
Smodule =
Smodule = Sincident = Shoriz = Smodule =
Shoriz = Sincident =

The tilt angle has a major impact on the solar radiation incident on a surface. For a fixed tilt angle, the maximum power over the course
of a year is obtained when the tilt angle is equal to the latitude of the location. However, steeper tilt angles are optimized for large
winter loads, while lower title angles use a greater fraction of light in the summer. The simulation below calculates the maximum
number of solar insolation as a function of latitude and module angle.

Latitude: 0°

-90 90

Array Tilt: 45°

0 90

The effect of latitude and module tilt on the solar radiation received throughout the year in W.h.m-2.day-1 without cloud. On the x-axis,
day is the number of days since January 1. The Module Power is the solar radiation striking a tilted module. The module tilt angle is
measured from the horizontal. The Incident Power is the solar radiation perpendicular to the sun's rays and is what would be received
by a module that perfectly tracks the sun. Power on Horizontal is the solar radiation striking the ground and is what would be received
for a module lying flat on the ground. These values should be regarded as maximum possible values at the particular location as they
do not include the effects of cloud cover. The module is assumed to be facing south in the northern hemisphere and north in the
southern hemisphere. For some angles, the light is incident from the rear of the module and in these cases the module power drops to 0.

As can be seen from the above animation, for a module tilt of 0°, the Module Power and Power on Horizontal are equal since the
module is lying flat on the ground. At a module tilt of 80°, the module is almost vertical. The Module Power is less than the Incident
Power except when the module is perpendicular to the sun's rays and the values are equal. The module is orientated to the equator so it
faces north in the Southern Hemisphere and south in the Northern Hemisphere. As module moves from the Northern to Southern
Hemisphere (latitude = 0°), the module is turned to face in the opposite direction and so the Module Power curve flips. When the light
is incident from the rear of the module the Module Power drops to zero . Try setting the latitude to your location and then varying the
module tilt to see the effect on the amount of power received throughout the year.

Arbitrary Orientation and Tilt

Add new comment
3 comment(s)
Español
한국어

For a module at an arbitrary tilt and orientation the equation becomes a little more complicated:

Solar Intensity
Edit
Delete
Clone

Read more about Solar Intensity

α is the sun elevation angle and Θ is the sun azimuth angle. β is the module tilt angle. A module lying flat on the ground has β =0° and
a vertical module has a β =90°. Ψ is the azimuth angle that the module faces. The vast majority of modules are aligned to face towards
the equator. A module in the southern hemisphere will be facing north with Ψ = 0° and a module in the northern hemisphere will
typically face directly south with Ψ = 180°. Smodule and Sincident are respectively the light intensities on the module and of the
incoming light in W/m², the Sincident being a direct only component.

A module that directly faces the sun so that the incoming rays are perpendicular to the module surface has the module tilt equal to the
sun's zenith angle (90 - α = β), and the module azimuth angle equal to the sun's azimuth angle (Ψ = Θ).

The following calculations combine the calculation of sun's position with the Airmass formula and then calculates the intensity of light
incident on a module with arbitrary tilt and orientation.

Light Intensity
Edit
Delete
Clone

Full Light Intensity Calculator

Year Month Day

The time is UTS, not local time.

Hour Minute Second

Longitude = Latitude =
Module Azimuth= ° Module Tilt=
°

Sun's Azimuth ° Zenith ° Elevation ° (using PSA

formula) Airmass units (from Airmass formula) Sincident
kW/m² (from Airmass Intensity formula) reduction in
light intensity using equation above Smodule kW/m² light
intensity on the module
Using Vectors to Calculate Solar Direction

As the number of tilts and orientations become more complicated it is often easier to convert the solar directions of azimuth and
elevation to vectors. An example is where there is a tilted module on a building that is also at an arbitrary tilt and orientation. The
simplicity of using vector comes from the fact that the reduction in intensity of light on a tilted surface is simply the dot product
between the incident ray and the normal to the module.

Light striking a surface at an angle is spread out over a larger area. The reduction in intensity is the dot product of the unit vectors S
and N

Solar Intensity Vector

Edit
Delete
Clone

Read more about Solar Intensity Vector

where Smodule and Sincident are as defined before and S is the unit vector point towards the sun and N is the unit vector normal to the
surface of the module. γ is the angle between the two vectors

Solar Radiation Data

Add new comment
2 comment(s)
Español

In the design of the photovoltaic system it is essential to know the amount of sunlight available in a given place at a given time.

Calculation of Solar Insolation

Add new comment
5 comment(s)
Español
한국어

Based on the equation of the sun's position in the sky throughout the year, the maximum amount of solar insolation on a surface at a
particular tilt angle can be calculated as a function of latitude and day of the year. These calculations are also essential in using
experimental data from sunshine hour recorders. The following animations calculate the daily solar irradiance, the solar insolation and
the number of hours during the day which the sun is shining. They do not include local weather effects and so these theoretical graphs
are not used in system sizing or prediction of operation. A description of each graph is given in the caption underneath.
 latitude= 0 degrees

-90 90

Day number of the year, Day= 1 days

1 365

The graph shows the intensity of direct radiation in W/m² throughout the day. It is the amount of power that would be received by a
tracking concentrator in the absence of cloud. The time is the local solar time. Set the latitude to your location and then adjust the day
slider to see how much radiation there is for each day of the year.

latitude2 =

-90 90

arrayTilt =
 0 80

The average daily solar insolation as a function of latitude. The three curves are the incident solar insolation, the horizontal solar
insolation and the solar insolation on a titled surface as defined in the page Module Tilt. The daily insolation is numerically equal to the
number of sunhours in a day. The module is assumed to face the equator so that it faces South in the northern hemisphere in North in
the southern hemisphere. As the latitude is adjusted through zero going across the equator, the module faces in the opposite direction.
the graph changes suddenly at the equator since the module is now facing in the opposite direction.

Latitude: 0°

-90 90

The number of hours the sun is shining each day, that is the number of hours between sunrise and sunset each day. In latitudes above
67° the sun shines for 24 hours during part of the year. Surprisingly, when averaged over the year, the sun shines an average of 12
hours per day everywhere in the world. In the northern latitudes the average intensity is lower than at the southern latitudes.

The equations to generate the above plots are given below. These equations are calculated in solar time, and not in local time. The
correction between local solar time and local time is given in the page The Sun's Position.

The number of sun hours is simply the time between sunrise:

and sunset:

The direct component of the solar radiation is determined from the air mass:

The airmass can be determined from the Air Mass formula:

Measurement of Solar Radiation

 Add new comment
Español
한국어

In PV system design it is essential to know the amount of sunlight available at a particular

location at a given time. The two common methods which characterise solar radiation are
the solar radiance (or radiation) and solar insolation. The solar radiance is an instantaneous
power density in units of kW/m2. The solar radiance varies throughout the day from 0
kW/m2 at night to a maximum of about 1 kW/m2. The solar radiance is strongly dependant
on location and local weather. Solar radiance measurements consist of global and/or direct
radiation measurements taken periodically throughout the day. The measurements are
taken using either a pyranometer (measuring global radiation) and/or a pyrheliometer
(measuring direct radiation). In well established locations, this data has been collected for
more than twenty years.

An alternative method of measuring solar radiation, which is less accurate but also less
expensive, is using a sunshine recorder. These sunshine recorders (also known as
Campbell-Stokes recorders), measure the number of hours in the day during which the
sunshine is above a certain level (typically 200 mW/cm2). Data collected in this way can
be used to determine the solar insolation by comparing the measured number of sunshine
hours to those based on calculations and including several correction factors.

A final method to estimate solar insolation is cloud cover data taken from existing satellite
images.

While solar irradiance is most commonly measured, a more common form of radiation
data used in system design is the solar insolation. The solar insolation is the total amount The photograph at left shows equipment
of solar energy received at a particular location during a specified time period, often in for solar irradiance measurements.
units of kWh/(m2 day). While the units of solar insolation and solar irradiance are both a (Photograph from David Pearsons) via
power density (for solar insolation the "hours" in the numerator are a time measurement as NREL information exchange.
is the "day" in the denominator), solar insolation is quite different than the solar irradiance
as the solar insolation is the instantaneous solar irradiance averaged over a given time
period. Solar insolation data is commonly used for simple PV system design while solar radiance is used in more complicated PV
system performance which calculates the system performance at each point in the day. Solar insolation can also be expressed in units
of MJ/m2 per year and other units and conversions are given in the units page.

Solar radiation for a particular location can be given in several ways including:

Typical mean year data for a particular location

Average daily, monthly or yearly solar insolation for a given location
Global isoflux contours either for a full year, a quarter year or a particular month
Sunshine hours data
Solar Insolation Based on Satellite Cloud-Cover Data
Calculations of Solar Radiation

Analysis of Solar Irradiance Data Sets

Add new comment
1 comment(s)
Español
한국어

The most accurate measurements of solar radiation are obtained by a pyrometer placed at a location for a number of years, usually on
the order of a decade or more, measuring the direct radiation every few minutes. However, the volume of data generated by this
technique makes it impractical (and unnecessary) to provide the full data set for each location for PV system design. Instead, the data
can be presented in several other formats.

The most conceptually straight forward method of reducing the data set is to average the data over the measuring period. This form of
data is called average daily, monthly or yearly radiation data. Although this data is useful for basic system design, the day-to-day
variation in the solar radiation is lost. The loss of the day-to-day variation is critical since the design and performance of a system with,
for example, 5 kWh/day nearly every day is quite different than one with 8 kWh/day on some days followed by several cloudy days
with 2 kWh/day.

Comparison of TMY and average solar radiation data.

The most common format for solar radiation data is TMY data (or TMY2 data used by the National Renewable Energy Laboratories in
the USA) which includes daily variability in the data. TMY data sets are described in the following page. However, average solar
radiation data, particularly for each month of the year is also extensively used in rough estimates on the amount of PV panels required.

An additional useful, although less common data which can be determined from the full radiation data sets, is the probability of having
a certain number of cloudy days which occur in a row, whereby the definition of a cloudy day is usually a day where less than 50% of
the theoretically expected radiation is received. For example, at a certain location, 4 cloudy days in a row may occur once a year and 5
cloudy days in a row may occur once every 5 years. This information is particularly useful in estimating storage sufficient
requirements. However, this information is less commonly tabulated and, if used, must be determined from the original data sets.

Typical Meteorological Year Data (TMY)

Add new comment
Español
한국어

The most common data for describing the local solar climate is through what is called Typical Meteorological Year data (TMY). To
determine TMY data, various meteorological measurements are made at hourly intervals over a number of years to build up a picture
of the local climate. A simple average of the yearly data underestimates the amount of variability, so the month that is most
representative of the location is selected. For each month, the average radiation over the whole measurement period is determined,
together with the average radiation in each month during the measurement period. The data for the month that has the average radiation
most closely equal to the monthly average over the whole measurement period is then chosen as the TMY data for that month. This
process is then repeated for each month in the year. The months are added together to give a full year of hourly samples.

There is no strict standard for TMY data so the user must adjust the data to suit the application. Considerable care must be taken with
sample periods. An example of a raw TMY data file is given below for January 1st in Melbourne, Australia. The comments on the
right hand side of the file describe the measurements taken and the data format.

TMY2 and TMY3

Due to the inconsistencies in TMY data, the data sets were updated for the US in 1994 (http://rredc.nrel.gov/solar/old_data/nsrdb
/1961-1990/tmy2/) and again in 2008 (http://rredc.nrel.gov/solar/old_data/nsrdb/1991-2005/tmy3/)
 1 1 1 0 0130 0 99007 MELBOURNE TMY: 75 71 75 73 76 78

1 1 2 0 0127 10 98040 78 72 78 70 69 68

1 1 3 0 0118 7 94040

1 1 4 0 0109 3 90041 -----------

1 1 5 0 0100 0 86001 PARAMETERS:

1 1 6 33106117 3 96141 -----------

1 1 7110260133 7106140 MONTH

1 1 8186311150 10115140 DAY

1 1 9257342183 7133140 HOUR

1 110317351217 3151140 HORIZONTAL GLOBAL IRRADIATION, MJ*100/HOUR

1 111356353250 0167000 SUN TRACKING BEAM IRRADIATION, MJ*100/HOUR

1 112381357272 14166120 AMBIENT TEMPERATURE, DEG.C*10

1 113387362293 27165120 WIND SPEED, M/S*10

1 114365359315 41164120 WET BULB TEMPERATURE, DEG.C*10

1 115324355314 43164130 WIND DIRECTION, COMPASS POINTS

1 116267348314 45163130 CLOUD COVER, OCTAS

1 117198336313 46162140 --------

1 118120277307 38162140 FORMAT :

1 119 44147300 29160150 --------

1 120 2 12294 21159150 1X,3I2,5I3,I2,I1

1 121 0 0274 26154150

1 122 0 0253 31148160

1 123 0 0233 36143160

1 124 0 0235 36142160

The description at the side of the above data is formatted into a more readable format, and is shown below:

Tracking Wind Wet

Global
Beam Temp Speed Bulb Wind Cloud
Month Day Hour Irradiation Irradiation (x10°C) (x10 Temp Direction Cover
x100MJ/h/m2
x100MJ/h/m2 m/s) (x10°C)
1 1 1 0 0 130 0 99 00 7

1 1 2 0 0 127 10 98 04 0
1 1 3 0 0 118 7 94 04 0

1 1 4 0 0 109 3 90 04 1
1 1 5 0 0 100 0 86 00 1

1 1 6 33 106 117 3 96 14 1

1 1 7 110 260 133 7 106 14 0

1 1 8 186 311 150 10 115 14 0

1 1 9 257 342 183 7 133 14 0

 1 1 10 317 351 217 3 151 14 0

1 1 11 356 353 250 0 167 00 0

1 1 12 381 357 272 14 166 12 0

1 1 13 387 362 293 27 165 12 0

1 1 14 365 359 315 41 164 12 0
1 1 15 324 355 314 43 164 13 0

1 1 16 267 348 314 45 163 13 0

1 1 17 198 336 313 46 162 14 0
1 1 18 120 277 307 38 162 14 0

1 1 19 44 147 300 29 160 15 0

1 1 20 2 12 294 21 159 15 0

1 1 21 0 0 274 26 154 15 0
1 1 22 0 0 253 31 148 16 0

1 1 23 0 0 233 36 143 16 0
1 1 24 0 0 235 36 142 16 0

TMY data is used for a wide variety of meteorological applications and therefore a large amount of data is usually irrelevant for
photovoltaic applications. Of the parameters given, usually only the time and irradiation figures are used. However, more advanced
models also use the temperature and wind speed.

Month is the month of the year with January = 1 and December = 12.
Day is the number of days within a month. Not all months are of equal value.
Hour is the hour of the day in 24 hour time, so the table above covers one full day. The data is usually an average for the hour
and covers ½ an hour before the sample to ½ an hour after the sample. Thus the first row on the table is for 1st January and
covers 12:30am to 1:30am.
The Global Irradiation is the amount of energy striking a horizontal surface during the hour. To convert from the above units of
x100MJ/hr/m2 to the typical photovoltaic units of kW/m2, divide by 360. The greatest irradiance is at midday and falls to 0 at
night.
The Direct Beam Irradiation gives the irradiation striking a plate perpendicular to the sun's rays but does not include diffuse
radiation. A tracking concentrator only sees the direct beam irradiation
The Temperature and Wind Speed are averaged over the hour. Note the x10 multiplication factor.
The Wet Bulb Temperature is the temperature shown by a thermometer bulb surrounded by a damp wick. The WBT varies
from the dry bulb temperature according to the amount of evaporation and thus the humidity. This detail is generally not used for
PV applications.
The Wind Direction shows the compass points from which the wind is coming. North is 00, East is 04 etc. Again, this detail is
generally not used for PV applications.
The Cloud Cover is a visual estimate of the proportion of cloud covering the sky. In this case it is in eights so 0 is no cloud and
8 is full cloud cover. And, once again this detail is generally not used for PV applications.

One additional unit often required in PV applications is the amount of diffuse radiation. The diffuse irradiation (Dh) falling on a
horizontal surface can be calculated from the global irradiation (Gh), the direct beam tracking radiation (It) and the elevation (el) from:

The diffuse radiation is not uniformly spread across the sky. For instance, the area just around the sun (circumsolar) is considerably
brighter than the rest of the sky. A commonly used model is the Perez model 1 which divides the sky into a circumsolar component,
horizon component,and the rest of the sky. The size of each of these regions and the relative intensities of the regions are adjusted to
closely model the measured radiation.

1. R. Perez, Ineichen, P., Seals, R., Michalsky, J., and Stewart, R., “Modeling daylight availability and irradiance components
from direct and global irradiance”, Solar Energy, vol. 44, pp. 271 - 289, 1990.
Making Use of TMY Data
Add new comment
1 comment(s)
Español
한국어

The amount of insolation acquired by the module during a certain time interval is one of the main preconditions for designing a PV
system. It can be estimated by calculating the position of the Sun in the sky as well as the Air Mass value as shown on the page
Calculation of Solar Insolation. Although this type of calculations gives a feeling of how much irradiation is available for a particular
location and module orientation and how it is varying throughout the year, its output can't be used for the real system design because it
doesn't include local weather variation which can dramatically change the radiation intensity. Another way of doing this is to use
empirical radiation data collected by meteo station located at the place of interest. The obtained data sets including radiation and
meteorological information for a typical year are available for the public and can be found in the form of xls spreadsheets.

The total amount of radiation received by the PV module, G, is composed of direct (beam), B, and diffuse, D, components.

In TMY file one can find both beam and diffuse components of radiation. Beam irradiance implies that it is received by the surface
perpendicular to the Sun's rays. The beam component is actually calculated from diffuse and global horizontal irradiances, which are
directly measured. For the systems with 2-axis trackers, adjusting themselves to face the sun in the same fashion as sunflowers do, the
beam component comes directly from TMY (where it is called Direct Normal Irradiance, DNI). Unfortunately, the vast majority of
PV systems are fixed in place and don't rotate. So they receive only a portion of direct sunlight which can be calculated having in mind
the system location, module tilt and orientation 1:

where:

δ is the Declination Angle,

φ is the latitude of the location,
β is module tilt,
ψ is module azimuth (orientation measured from South to West),
and HRA is hour angle, discussed on the page Solar Time.

The calculation of diffuse component is more straight forward. From a simple model, assuming isotropic radiation from the whole sky
dome, it follows that the module tilted by the angle β will receive just the proportional part of Diffuse Horizontal Irradiation, DHI,
provided by TMY.
Although the simple model works pretty well for the desert locations it will introduce errors for the places with significant cloud
coverage. For those parts of the Earth, the special models of the sky should be used to achieve higher accuracy. 2

TMY is measured on an hourly bases. The data is accumulated during the 60 min time intervals and is stored in a standardized form as
shown on the following figure:

In the example above with a time stamp of 11:00 am, the data is collected over the period of 10 am to 11 am. The approximate average
position of the sun over that period is 10:30 am or 30 minutes before the time stamp.

Combining the equations for the sun's position with those for the TMY data gives the hourly global irradiation on an arbitrary oriented
and tilted surface, which can further be averaged or integrated through the required time interval. The contour plot, such as the one
below, shows the irradiation on the modules versus module tilt and azimuth and can be used to calculate the yearly output of the
module. For example, modules integrated into a vertical building wall facing East will generate around half the energy of modules that
are South facing and tilted by the angle equal to the latitude.
The contour diagram of the global daily irradiation averaged through the year for Phoenix, AZ plotted for different module tilts and
orientations. Note, that the number on the color bar corresponds to the peak sun hours, e.g. in Phoenix, if the module is facing South
and tilted 40o, it will get about 6 peak sun hours of insolation during the day throughout the year. Click here to see the full size image.
The actual data can also be obtained here.

Contour plots as well as data files for other locations throughout the US can be found below. It's interesting to note that the maximum
incoming irradiation point is shifted from South to either East or West depending on the location. The reason for this behavior is the
cloud coverage, which can appear mainly in the afternoon or in the morning during one of the seasons, causing the shift of the
optimum orientation to the East or to the West correspondingly.

Wilmington, DE Miami, FL Austin, TX Seattle, WA Honolulu, HI

Image Image Image Image Image
Data Data Data Data Data

1. A. Luque and Hegedus, S., “Handbook of Photovoltaic Science and Engineering”, p. 1117, 2003.
2. R. Perez, Ineichen, P., Seals, R., Michalsky, J., and Stewart, R., “Modeling daylight availability and irradiance components
from direct and global irradiance”, Solar Energy, vol. 44, pp. 271 - 289, 1990.

Average Solar Radiation

Add new comment
2 comment(s)
Español
한국어

Although TMY data is commonly used for PV system simulation, the average daily solar radiation at a location in a given month is
often sufficient for a basic system analysis. This data may be presented either as measured on the horizontal or measured with the
measuring surface perpendicular to the solar radiation (corresponding to a PV system which tracks the sun). In either case, an
additional angular dependence to account for the tilt of the module will need to be incorporated in order to determine the amount of
solar radiation available to a PV module.
Peak Sun Hours

The average daily solar insolation in units of kWh/m2 per day is sometimes referred to as "peak sun hours". The term "peak sun hours"
refers to the solar insolation which a particular location would receive if the sun were shining at its maximum value for a certain
number of hours. Since the peak solar radiation is 1 kW/m2, the number of peak sun hours is numerically identical to the average daily
solar insolation. For example, a location that receives 8 kWh/m2 per day can be said to have received 8 hours of sun per day at 1
kW/m2. Being able to calculate the peak sun hours is useful because PV modules are often rated at an input rating of 1kW/m2.

Isoflux Contour Plots

Add new comment
Español
한국어

Worldwide solar radiation measurements may be compiled to give an indication of solar insolation across the world. Due to the lack of
detail in these plots, they are not used for system design.
Average quarterly global isoflux contour plots for each quarter in the year. The units are in MJ/m2 and give the solar insolation falling
on a horizontal surface per day.

Sunshine Hour Data

Add new comment
Español
한국어

Although measurements of the solar radiance provide the most accurate information of solar radiation, these measurements can be
difficult to obtain. A simpler method of measuring sunshine hours is to use a card system whereby sunlight is focused into a recording
card. If the sunshine is greater than about 200W/m2 a mark is burned onto the recording chart. The number of sunshine hours is
determined as the number of hours during which the sun is shining. When the number of sunshine hours is combined with a knowledge
of local atmospheric conditions, sunshine data can be used to estimate the available monthly average solar radiation according to the
equation:

where:
a and b are constants determined by the location and depend on atmospheric conditions
HO is the average solar radiation for that month on a clear day (which can be calculated)
n is the measured number of sunshine hours data
N is the number of sunshine hours calculated as shown in the page Calculation of Solar Insolation.

The calculations for HO and N are shown in later pages. The constant a varies around 0.25, with values just below 0.2 used for some
locations and values up to 0.4 used in others. The constant b takes values in the range of 0.4 to 0.6.

For Australia, studies by Telecom Australia (now Telstra) indicate that values of a = 0.24 and b = 0.48 are accurate to within 10%.

Values for several of the locations in India have been tabulated by G.D. Rai and are given below.

Indian City a b
Ahmedabad 0.28 0.48

Bangalore 0.18 0.64

Baroda 0.28 0.48

Bhavnagar 0.28 0.47

Bhopal 0.27 0.50

Calcutta 0.28 0.42

Goa 0.30 0.48

Jodhpur 0.33 0.46

Madras 0.30 0.44

Mangalore 0.27 0.43

Shillong 0.22 0.57

Srinagar 0.35 0.40

Trivandrum 0.37 0.39

Vishakhapatnam 0.28 0.47

Nagpur 0.28 0.50

New Delhi 0.25 0.57

Poona 0.31 0.43

Roorkee 0.25 0.56

Cloud Cover Data

Add new comment
Español
한국어

An additional valuable source of solar radiation data is the determination of solar radiation from satellite images. These images provide
information about the level of cloud cover in a particular region. Information in relation to cloud cover levels is used to provide
estimates of the solar insolation at a specific location. While such cloud cover data represents an important resource to determine the
radiation at a broader level, this data is not presently generally used in detailed system design for a specific location.

Isoflux contours based on cloud cover data and presented by courtesy of Solarex. The units on the map are in kWh/m2/day and
represent the minimum case values at a module tilt angle equal to the latitude. Click on the image for a much larger version.
The solar insolation for a module at tilt equal to the latitude. The map is essentially the same as the one above but has been updated
with more detail. The data in this graph is used to calculate the output of a flat plate module without tracking.
Solar insolation available to a two dimension tracking module with concentration. Such systems only respond to direct normal
insolation (DNI) but receive a higher irradiance in sunny locations due to tracking. A concentrator tracking system with the same
efficiency as a flat plate module would produce more energy over the year in the southwest but less energy in the north east and
Florida. However, concentrators typically have a higher efficiency than flat plate. The best concentrator modules approach 30%
efficiency while the best flat plate modules approach 20%.
The presented here for Europe uses a slightly different scale of the total irradiance over the year. To convert to the daily average used
in the NREL plots, divide by 365 so that the global irradiance scale on the graph goes from 1.6 kWh/m2/day (dark blue) to 6
kWh/m2/day (dark red). More information on the solar irradiance in Europe is found at: http://sunbird.jrc.it/pvgis/

Satellite Derived Irradiance Estimates

by Nick Engerer at Solcast

Previously, pages have covered the basis of solar geometry, and the general properties of solar radiance, including its measurement at
the surface. However, given the relatively limited number of surface radiance measurement stations, compared to the total surface area
of the Earth, other methods for estimating solar radiance over time have been developed through the advent of modern weather satellite
technologies.

The modern methods for estimating solar radiance from geostationary weather satellites consists of four major steps. First, is the
detection of cloud cover. Second is the characterization of that cloud cover in terms of its impact on solar radiance. Third is modeling
the available solar radiance under clear skies. And fourth is the final estimate of the amount of solar irradiance reaching the Earth’s
surface after it passes through the clouds (if any are present).

Step 1: Detecting Cloud Cover with Albedo

The core technology used for detecting clouds from space, is the imagery produced by satellites at 36,000 km (22,000 mi) above the
Earth’s surface in geostationary orbits (meaning they are stationary relative to the Earth's surface). The latest generation satellites
produce full-disk scans of the Earth every 10-15 minutes at resolutions as fine as 500 m. Imagers are recorded in several wavelengths
of visible light, as well as many infrared wavelengths.

The above raw image from the satellite includes both land and ocean surfaces, as well as cloud features. Upon closer inspection, there
are also regions of very bright land surfaces such as sands and also snow cover. In order to separate the land surface information and
the cloud cover features, an estimate of the ‘albedo’ or appearance of the background when no clouds are present, needs to be
calculated.
 A raw visible satellite image over Europe, as provided by a Meteosat Second Generation satellite.

Albedo estimates are created by scanning satellite images from the recent history, which do not have any cloud cover present, to store
in memory, what the surface should look like without any clouds. A deviation from this expected appearance will then reveal the
location of a cloud. In the figure below you can review an example albedo calculation. Importantly, you’ll note the presence of snow
cover in various locations on the map, as well as some areas of bright sands in Northern Africa. Without an appropriate albedo
calculation, these features could easily be miscategorised as cloud cover.

The albedo calculation for the same time-step as the above image. Solcast uses data from approximately the past 30 days to
compute the albedo estimate.

Step 2: Characterizing Cloud Cover via Cloud Opacity

Once the albedo calculation has been applied to the raw visible imagery to, the next step is make an estimate of the thickness of the
cloud cover that is present to solar radiance. The opaqueness of a given cloud cover feature to solar radiance is termed by Solcast to be
‘cloud opacity’. Cloud opacity is calculated in a variety of ways, some of which are proprietary and not documented publicly.
However, most, if not all, will use the difference between the raw visible imagery and calculated albedo as the principle input to
estimating cloud opacity. These and the formulae for estimating the relationships between cloud opacity and solar radiance are then
calibrated using surface solar measurements from pyranometers located in the satellite coverage region. An example of the cloud
opacity estimate from our original satellite image in the first image, is presented below.

Cloud opacity calculation. Note how the land surface features have been completely removed, with only the cloud features
remaining. Cloud opacity is then calculated from a value of 0 - 100, using the legend at right..

Step 3: Estimating Solar Radiance with a Clear-Sky Model

There is one additional step to estimating the solar radiance that reaches the Earth’s surface, and that is to apply the estimated cloud
opacity value to an expectation of how much solar radiance would be available under clear sky conditions. This is best thought of as
the amount of sunlight that would reach the Earth’s surface if there were no clouds in the way. The scientific literature has a rich trove
models for computing clear-sky radiance, with a wide variety of models which use varying data inputs such as the REST2 model1.
Generally speaking these models rely on a handful of input variables to be computed:

Solar geometry. The more overhead the sun is, the more solar radiation a given patch of the surface gets.
Earth/Sun distance. There is a few percent variation in sunlight reaching the earth as its elliptical orbit takes it further and closer
to the sun.
Aerosols/Turbidity: Air is not 100% transparent, and water vapor and other aerosols attenuate some of the solar radiation that
would reach the surface if there weren't an atmosphere in the way.

Step 4: Satellite Derived Solar Radiance

Once the clear-sky radiance has been accurately estimated through a clear-sky radiance model, a simple conversion is applied using the
estimated cloud opacity values from the prior steps to generate a spatially continuous estimate of the availability of solar radiance at
each location covered by the geostationary weather satellite imagery.

This calculation can be completed as follows:

$$ E=E_{clearsky} \times \frac{100 - \alpha}{100} $$

Where E is the solar radiance (global horizontal) and α is the cloud opacity (from 0 - 100).

The end result, which is an estimate of the solar radiance over the entire region of the original satellite image is presented below.

1. C. A. Gueymard, “REST2: High-performance solar radiation model for cloudless-sky irradiance, illuminance, and
 The final solar radiance (global horizontal) estimate produced from the original satellite image. Units in W/m2.

photosynthetically active radiation – Validation with a benchmark dataset”, Solar Energy, vol. 82, no. 3, pp. 272 - 285, 2008.

PN Junctions
Add new comment
2 comment(s)
Español

PN Junctions

Basics
Add new comment
Español

The wide range of semiconductor applications comes from the ability to easily change their conductivity. The addition of even very
small amounts of impurities, known as dopants, can change their material conductivity over orders of magnitude even though the
impurity concentration might be very small with concentrations of the order of parts per billion. The conductivity may also be changed
by applying a voltage or current in one part of the materials to cause a large change in the conductivity in another part of the device.
Solar cells are usually based around PN junction devices and are just large diodes that have been optimized to absorb light. As such
solar cells belong to the family of of bipolar junction devices, which also includes diodes and bipolar junction transistors (BJTs).

Semiconductors typically use the CGS (centimeter, gram, second) unit system. However, the thickness of a solar cell will be measured
in microns. For modeling and calculations, the easiest approach is to convert all dimensions to cm so that a wafer of thickness 160 μm
becomes 0.016 cm.

Introduction to Semiconductors
Add new comment
Español
한국어

Section Goals
 1. Understand semiconductor function within the context of PV.
2. Learn how to optimize semiconductor performance in PV.
3. Understand why silicon is the most commonly used semiconductor material for PV applications.

Solar cells have always been aligned closely with other electronic devices. The following pages cover the basic aspects of
semiconductor materials and the physical mechanisms which are at the center of photovoltaic devices. These physical mechanisms are
used to explain the operation of a p-n junction, which forms the basis not only for the great majority of solar cells, but also most other
electronic devices such as lasers and bipolar junction transistors. Much of the theory of solid-state semiconductors was worked out
during the invention of the transistor in the late 40s and early 50s 1. While PV semiconductor materials are not limited to silicon, the
dominance of silicon in the PV market has led to our particular emphasis of that material for the PVCDROM. For the purposes of the
website, the material presented specifically focuses on silicon-based solar cells, unless otherwise noted.

An ingot of silicon, consisting of a single large crystal of silicon. Such an ingot is sliced into individual wafers and then used to make a
variety of semiconductor devices, including solar cells and computer chips.

1. W. Shockley, Electrons and holes in semiconductors with applications to transistor electronics. New York: van Nostrand,
1950.

Semiconductor Materials
Add new comment
2 comment(s)
Español
한국어

Overview

1. Semiconductor materials come from different groups in the periodic table, yet share certain similarities.
2. The properties of the semiconductor material are related to their atomic characteristics and change from group to group.
3. Researchers and designers take advantage of these differences to improve the design and choose the optimal material for a PV
application.

The atoms in a semiconductor are materials from either group IV of the periodic table, or from a combination of group III and group V
(called III-V semiconductors), or of combinations from group II and group VI (called II-VI semiconductors). Because different
semiconductors are made up of elements from different groups in the periodic table, properties vary between semiconductors. Silicon,
which is a group IV, is the most commonly used semiconductor material as it forms the basis for integrated circuit (IC) chips and is the
most mature technology and most solar cells are also silicon based. A full periodic table is given in the page Periodic Table. Several of
the material properties of silicon are given in the page Silicon Material Parameters
A section from the periodic table. More common semiconductor materials are shown in blue. A semiconductor can be either of a single
element, such as Si or Ge, a compound, such as GaAs, InP or CdTe, or an alloy, such as SixGe(1-x) or AlxGa(1-x)As, where x is the
fraction of the particular element and ranges from 0 to 1.

Semiconductor Structure
Add new comment
2 comment(s)
Español
한국어

Overview

1. Semiconductors are made up of atoms bonded together to form a uniform structure

2. Each silicon atom has four valence electrons which are shared, forming covalent bonds with the four surrounding Si atoms.
3. Understanding how these atoms are arranged is vital in understanding the material properties of different semiconductors, and
how best to engineer them.

Semiconductors, such as Silicon (Si) are made up of individual atoms bonded together in a regular, periodic structure to form an
arrangement whereby each atom is surrounded by 8 electrons. An individual atom consists of a nucleus made up of a core of protons
(positively charged particles) and neutrons (particles having no charge) surrounded by electrons. The number of electrons and protons
is equal, such that the atom is overall electrically neutral. The electrons surrounding each atom in a semiconductor are part of a
covalent bond. A covalent bond consists of two atoms "sharing" a pair of electrons. Each atom forms 4 covalent bonds with the 4
surrounding atoms. Therefore, between each atom and its 4 surrounding atoms, 8 electrons are being shared. The structure of a
semiconductor is shown in the figure below.

Schematic representation of covalent bonds in a silicon crystal lattice. Each line connecting the atoms represents an electron being
shared between the two. Two electrons being shared are what form the covalent bond.

Conduction in Semiconductors
Add new comment
Español
한국어

Overview

1. Semiconductors act as insulators at low temperatures and conductors at higher temperatures.

2. Conduction occurs at higher temperature because the electrons surrounding the semiconductor atoms can break away from their
covalent bond and move freely about the lattice
3. The conductive property of semiconductors forms the basis for understanding how we can use these materials in electrical
devices.

The bond structure of a semiconductor determines the material properties of a semiconductor. One key effect are the energy levels
which the electrons can occupy and how they move about the crystal lattice. The electrons in the covalent bond formed between each
of the atoms in the lattice structure are held in place by this bond and hence they are localized to the region surrounding the atom.
These bonded electrons cannot move or change energy, and thus are not considered "free" and cannot participate in current flow,
absorption, or other physical processes of interest in solar cells. However, only at absolute zero are all electrons in this "stuck,"
bonded arrangement. At elevated temperatures, especially at the temperatures where solar cells operate, electrons can gain enough
energy to escape from their bonds. When this happens, the electrons are free to move about the crystal lattice and participate in
conduction. At room temperature, a semiconductor has enough free electrons to allow it to conduct current. At or close to absolute zero
a semiconductor behaves like an insulator.

When an electron gains enough energy to participate in conduction (is "free"), it is at a high energy state. When the electron is bound,
and thus cannot participate in conduction, the electron is at a low energy state. Therefore, the presence of the bond between the two
atoms introduces two distinct energy states for the electrons. The electron cannot attain energy values intermediate to these two levels;
it is either at a low energy position in the bond, or it has gained enough energy to break free and therefore has a certain minimum
energy. This minimum energy is called the "band gap" of a semiconductor. The number and energy of these free electrons, those
electrons participating in conduction, is basic to the operation of electronic devices.

The space left behind by the electrons allows a covalent bond to move from one electron to another, thus appearing to be a positive
charge moving through the crystal lattice. This empty space is commonly called a "hole", and is similar to an electron, but with a
positive charge.

Animation showing formation of "free" electrons and holes when an electron can escape its bond.

The most important parameters of a semiconductor material for solar cell operation are:

the band gap;

the number of free carriers (electrons or holes) available for conduction; and
the "generation" and recombination of free carriers (electrons or holes) in response to light shining on the material.

More detail on these properties is given in the following pages.

Band Gap
Add new comment
2 comment(s)
 Español
한국어

Overview

1. The band gap is the minimum amount of energy required for an electron to break free of its bound state.
2. When the band gap energy is met, the electron is excited into a free state, and can therefore participate in conduction.
3. The band gap determines how much energy is needed from the sun for conduction, as well as how much energy is generated.
4. A hole is created where the electron was formerly bound. This hole also participates in conduction.

The band gap of a semiconductor is the minimum energy required to excite an electron that is stuck in its bound state into a free state
where it can participate in conduction. The band structure of a semiconductor gives the energy of the electrons on the y-axis and is
called a "band diagram". The lower energy level of a semiconductor is called the "valence band" (EV) and the energy level at which an
electron can be considered free is called the "conduction band" (EC). The band gap (EG) is the gap in energy between the bound state
and the free state, between the valence band and conduction band. Therefore, the band gap is the minimum change in energy required
to excite the electron so that it can participate in conduction.

Schematic of the energy bands for electrons in a solid.

Once the electron becomes excited into the conduction band, it is free to move about the semiconductor and participate in conduction.
However, the excitation of an electron to the conduction band will also allow an additional conduction process to take place. The
excitation of an electron to the conduction band leaves behind an empty space for an electron. An electron from a neighboring atom
can move into this empty space. When this electron moves, it leaves behind another space. The continual movement of the space for an
electron, called a "hole", can be illustrated as the movement of a positively charged particle through the crystal structure.
Consequently, the excitation of an electron into the conduction band results in not only an electron in the conduction band but also a
hole in the valence band. Thus, both the electron and hole can participate in conduction and are called "carriers".

The concept of a moving "hole" is analogous to that of a bubble in a liquid. Although it is actually the liquid that moves, it is easier to
describe the motion of the bubble going in the opposite direction.

Intrinsic Carrier Concentration

Add new comment
Español
한국어

Overview

1. Intrinsic carriers are the electrons and holes that participate in conduction.
2. The concentration of these carriers is contingent upon the temperature and band gap of the material, thus affecting a material's
conductivity.
3. Knowledge of intrinsic carrier concentration is linked to our understanding of solar cell efficiency, and how to maximize it.

The thermal excitation of a carrier from the valence band to the conduction band creates free carriers in both bands. The concentration
of these carriers is called the intrinsic carrier concentration, denoted by ni. Semiconductor material which has not had impurities added
to it in order to change the carrier concentrations is called intrinsic material. The intrinsic carrier concentration is the number of
electrons in the conduction band or the number of holes in the valence band in intrinsic material. This number of carriers depends on
the band gap of the material and on the temperature of the material. A large band gap will make it more difficult for a carrier to be
thermally excited across the band gap, and therefore the intrinsic carrier concentration is lower in higher band gap materials.
Alternatively, increasing the temperature makes it more likely that an electron will be excited into the conduction band, which will
increase the intrinsic carrier concentration. This translates directly to solar cell efficiency.

Intrinsic carrier concentration in a semiconductor at two temperatures. In both cases, the number of electrons and the number of holes
is equal. Undoped silicon (intrinsic) is rarely used in the electronics industry it is almost always doped for device fabrication.

Intrinsic Carrier Concentration of Silicon as a Function of Temperature

The exact value of the intrinsic carrier concentration in silicon has been extensively studied due to its importance in modeling. At 300
K the generally accepted value for the intrinsic carrier concentration of silicon, ni, is 9.65 x 109 cm-3 as measured by Altermatt1, which
is an update to the previously accepted value given by Sproul2. A formula for the intrinsic carrier concentration in silicon as a function
of temperature is given by Misiakos3:

The small difference in the values of Altermatt and Misiakos is within the bounds of experimental error. While the intrinsic carrier
concentration is normally quoted at 300 K, solar cells are usually measured at 25 °C where the intrinsic carrier concentration is 8.3 x
109 cm-3. The above equation is implemented in the mini-calculator below:

Temperature - Intrinsic Carrier Concentration Calculator

Temperature, T = K
Intrinsic Carrier Concentration, ni= cm-3

1. P. P. Altermatt, Schenk, A., Geelhaar, F., and Heiser, G., “Reassessment of the intrinsic carrier density in crystalline silicon in
view of band-gap narrowing”, Journal of Applied Physics, vol. 93, no. 3, p. 1598, 2003.
2. A. B. Sproul and Green, M. A., “Improved value for the silicon intrinsic carrier concentration from 275 to 375 K”, Journal of
Applied Physics, vol. 70, pp. 846-854, 1991.
3. K. Misiakos and Tsamakis, D., “Accurate measurements of the silicon intrinsic carrier density from 78 to 340 K”, Journal of
Applied Physics, vol. 74, no. 5, p. 3293, 1993.
Generation
Add new comment
1 comment(s)
Español

The absorption of light and the generation of an electron hole pair is fundamental to the operation of a solar cell. In this section the
process whereby the energy of a photon is initially converted to electrical energy through the creation of an electron hole pair.

Doping
Add new comment
2 comment(s)
Español

Overview

1. Doping is a technique used to vary the number of electrons and holes in semiconductors.
2. Doping creates N-type material when semiconductor materials from group IV are doped with group V atoms. P-type materials
are created when semiconductor materials from group IV are doped with group III atoms.
3. N-type materials increase the conductivity of a semiconductor by increasing the number of available electrons; P-type materials
increase conductivity by increasing the number of holes present.

It is possible to shift the balance of electrons and holes in a silicon crystal lattice by "doping" it with other atoms.

Atoms with one more valence electron than silicon are used to produce "n-type" semiconductor material. These n-type materials are
group V elements in the periodic table, and thus their atoms have 5 valence electrons that can form covalent bonds with the 4 valence
electrons that silicon atoms have. Because only 4 valence electrons are needed from each atom (silicon and n-type) to form the
covalent bonds around the silicon atoms, the extra valence electron present (because n-type materials have 5 valence electrons) when
the two atoms bond is free to participate in conduction. Therefore, more electrons are added to the conduction band and hence
increases the number of electrons present.

Atoms with one less valence electron result in "p-type" material. These p-type materials are group III elements in the periodic table.
Therefore, p-type material has only 3 valence electrons with which to interact with silicon atoms. The net result is a hole, as not
enough electrons are present to form the 4 covalent bonds surrounding the atoms. In p-type material, the number of electrons trapped in
bonds is higher, thus effectively increasing the number of holes. In doped material, there is always more of one type of carrier than the
other and the type of carrier with the higher concentration is called a "majority carrier", while the lower concentration carrier is called a
"minority carrier."

Schematic of a silicon crystal lattice doped with impurities to produce n-type and p-type semiconductor material.

The following table summarizes the properties of semiconductor types in silicon.

N-type (negative) P-type (positive)

Dopant Group V (e.g. Phosphorous) Group III (e.g. Boron)

Bonds Excess Electrons Missing Electrons (Holes)
 Majority Carriers Electrons Hole
Minority Carriers Holes Electrons

The animations below represent p-type and n-type silicon. In a typical semiconductor there might be 1017cm-3 majority carriers and
106cm-3 minority carriers. Expressed in a different form, the ratio of minority to majority carriers is less than one person to the entire
population of the planet. Minority carriers are created either thermally or by incident photons.

N-type semiconductor. These are called "n-type" since the majority carriers are Negatively charged electrons.

P-type semiconductor. These are called "p-type" since the majority carriers are Positively charged holes.

Equilibrium Carrier Concentration

Add new comment
Español

Overview

1. Semiconductors contain majority and minority carriers. The more abundant charge carriers are the majority carriers; the less
abundant are the minority carriers.
2. The equilibrium carrier concentration can be increased through doping.
3. The total number of carriers in the conduction and valence band is called the equilibrium carrier concentration.
4. The product of minority and majority charge carriers is a constant.
The number of carriers in the conduction and valence band with no externally applied bias is called the equilibrium carrier
concentration. For majority carriers, the equilibrium carrier concentration is equal to the intrinsic carrier concentration plus the number
of free carriers added by doping the semiconductor. Under most conditions, the doping of the semiconductor is several orders of
magnitude greater than the intrinsic carrier concentration, such that the number of majority carriers is approximately equal to the
doping.

At equilibrium, the product of the majority and minority carrier concentration is a constant, and this is mathematically expressed by the
Law of Mass Action.

where ni is the intrinsic carrier concentration and n0 and p0 are the electron and hole equilibrium carrier concentrations.

Using the Law of Mass Action above, the majority and minority carrier concentrations are given as:

where ND is the concentration of donor atoms and NA is the concentration of acceptor atoms.

The above equations show that the number of minority carriers decreases as the doping level increases. For example, in n-type
material, some of the extra electrons added by doping the material will occupy the empty spots (i.e., holes) in the valence band, thus
lowering the number of holes.

Equilibrium carrier concentration for low and high doping, showing that as the doping increases, the minority carrier concentration
decreases.
Absorption of Light
Add new comment
3 comment(s)
Español

Overview

1. When the energy of a photon is equal to or greater than the band gap of the material, the photon is absorbed by the material and
excites an electron into the conduction band.
2. Both a minority and majority carrier are generated when a photon is absorbed.
3. The generation of charge carriers by photons is the basis of the photovoltaic production of energy.

Photons incident on the surface of a semiconductor will be either reflected from the top surface, will be absorbed in the material or,
failing either of the above two processes, will be transmitted through the material. For photovoltaic devices, reflection and transmission
are typically considered loss mechanisms as photons which are not absorbed do not generate power. If the photon is absorbed it has the
possibility of exciting an electron from the valence band to the conduction band. A key factor in determining if a photon is absorbed or
transmitted is the energy of the photon. Therefore, only if the photon has enough energy will the electron be excited into the
conduction band from the valence band. Photons falling onto a semiconductor material can be divided into three groups based on their
energy compared to that of the semiconductor band gap:

Eph < EG Photons with energy Eph less than the band gap energy EG interact only weakly with the semiconductor, passing
through it as if it were transparent.
Eph = EG have just enough energy to create an electron hole pair and are efficiently absorbed.
Eph > EG Photons with energy much greater than the band gap are strongly absorbed. However, for photovoltaic applications, the
photon energy greater than the band gap is wasted as electrons quickly thermalize back down to the conduction band edges.

The effect of the three classes of photons on the semiconductor is shown in the two animations below.

The creation of electron-hole pairs when illuminated with light Eph = hf, where Eph > EG.

The absorption of photons creates both a majority and a minority carrier. In many photovoltaic applications, the number of light-
generated carriers are of orders of magnitude less than the number of majority carriers already present in the solar cell due to doping.
Consequently, the number of majority carriers in an illuminated semiconductor does not alter significantly. However, the opposite is
true for the number of minority carriers. The number of photo-generated minority carriers outweighs the number of minority carriers
existing in the doped solar cell in the dark (because in doping the minority carrier concentration is so small), and therefore the number
of minority carriers in an illuminated solar cell can be approximated by the number of light generated carriers.

Absorption Coefficient
Add new comment
Español

Overview
 1. Different semiconductor materials have different absorption coefficients.
2. Materials with higher absorption coefficients more readily absorb photons, which excite electrons into the conduction band.
3. Knowing the absorption coefficients of materials aids engineers in determining which material to use in their solar cell designs.

The absorption coefficient determines how far into a material light of a particular wavelength can penetrate before it is absorbed. In a
material with a low absorption coefficient, light is only poorly absorbed, and if the material is thin enough, it will appear transparent to
that wavelength. The absorption coefficient depends on the material and also on the wavelength of light which is being absorbed.
Semiconductor materials have a sharp edge in their absorption coefficient, since light which has energy below the band gap does not
have sufficient energy to excite an electron into the conduction band from the valence band. Consequently, this light is not absorbed.
The absorption coefficient for several semiconductor materials is shown below.

The absorption coefficient, α, in a variety of semiconductor materials at 300K as a function of the vacuum wavelength of light.

The above graph shows that even for those photons which have an energy above the band gap, the absorption coefficient is not
constant, but still depends strongly on wavelength. The probability of absorbing a photon depends on the likelihood of having a
photon and an electron interact in such a way as to move from one energy band to another. For photons which have an energy very
close to that of the band gap, the absorption is relatively low since only those electrons directly at the valence band edge can interact
with the photon to cause absorption. As the photon energy increases, not just the electrons already having energy close to that of the
band gap can interact with the photon. Therefore, a larger number of electrons can interact with the photon and result in the photon
being absorbed.

The absorption coefficient, α, is related to the extinction coefficient, k, by the following formula:

where λ is the wavelength. If λ is in nm, multiply by 107 to get the absorption coefficient in the units of cm-1.

Additional optical properties of silicon are given in the page Optical Properties of Silicon.
Absorption Depth
Add new comment
Español

Overview

1. The absorption depth is given by the inverse of the absorption coefficient, and describes how deeply light penetrates into a
semiconductor before being absorbed.
2. Higher energy light is of a shorter wavelength and has a shorter absorption depth than lower energy light, which is not as readily
absorbed, and has a greater absorption depth.
3. Absorption depth affects aspects of solar cell design, such as the thickness of the semiconductor material.

The relationship between absorption coefficient and wavelength makes it so that different wavelengths penetrate different distances
into a semiconductor before most of the light is absorbed. The absorption depth is given by the inverse of the absorption coefficient, or
α-1. The absorption depth is a useful parameter which gives the distance into the material at which the light drops to about 36% of its
original intensity, or alternately has dropped by a factor of 1/e. Since high energy light (short wavelength), such as blue light, has a
large absorption coefficient, it is absorbed in a short distance (for silicon solar cells within a few microns) of the surface, while red
light (lower energy, longer wavelength) is absorbed less strongly. Even after a few hundred microns, not all red light is absorbed in
silicon. The variation in the absorption depth for "blue" and "red" photons is shown below.

The blue photons are absorbed very close to the surface but most of the red photons are absorbed deep in the device.

Generation Rate
Add new comment
3 comment(s)
Español

Overview

1. The generation of an electron-hole pair can be calculated at any location within the solar cell, at any wavelength of light, or for
the entire standard solar spectrum.
2. Generation is the greatest at the surface of the material, where the majority of the light is absorbed.
3. Because the light used in PV applications contains many different wavelengths, many different generation rates must be taken
into account when designing a solar cell.

The generation rate gives the number of electrons generated at each point in the device due to the absorption of photons. Generation is
an important parameter in solar cell operation.
Neglecting reflection, the amount of light which is absorbed by a material depends on the absorption coefficient (α in cm-1) and the
thickness of the absorbing material. The intensity of light at any point in the device can be calculated according to the equation:

where α is the absorption coefficient typically in cm-1;

x is the distance into the material at which the light intensity is being calculated; and
I0 is the light intensity at the top surface.

The above equation can be used to calculate the number of electron-hole pairs being generated in a solar cell. Assuming that the loss in
light intensity (i.e., the absorption of photons) directly causes the generation of an electron-hole pair, then the generation G in a thin
slice of material is determined by finding the change in light intensity across this slice. Consequently, differentiating the above
equation will give the generation at any point in the device. Hence:

where N0 = photon flux at the surface (photons/unit-area/sec.);

α = absorption coefficient; and
x = distance into the material.

The above equations show that the light intensity exponentially decreases throughout the material and further that the generation is
highest at the surface of the material.

For photovoltaic applications, the incident light consists of a combination of many different wavelengths, and therefore the generation
rate at each wavelength is different. The generation rate at different wavelengths in silicon is shown below.

Wavelength of light, λ= 800 nm, with an absorption coefficient α = 850 cm-1

300 1100

Changing the slider in the graph above changes the wavelength of the incoming light. The changing absorption coefficient causes the
light to be absorbed at different depths. The generation rate has been normalized.

To calculate the generation for a collection of different wavelengths, the net generation is the sum of the generation for each
wavelength. The generation as a function of distance for a standard solar spectrum (AM 1.5) incident on a piece of silicon is shown
below. The y-axis scale is logarithmic showing that there is an enormously greater generation of electron-hole pairs near the front
surface of the cell, while further into the solar cell the generation rate becomes nearly constant.
Generation rate of electron-hole pairs in a piece of silicon as a function of distance into the cell. The cell front surface is at 0 µ m and is
where most of the high energy blue light is absorbed.

Types of Recombination
Add new comment
Español

Overview

1. Eventually, electrons lose energy and stabilize back to the valence band, recombining with a hole.
2. There are three types of recombination; Radiative, Shockley-Read-Hall, and Auger.
3. Auger and Shockley-Read-Hall recombination dominate in silicon-based solar cells.
4. Among other factors, recombination is associated with the lifetime of the material, and thus of the solar cell.

Any electron which exists in the conduction band is in a meta-stable state and will eventually stabilize to a lower energy position in the
valence band. When this occurs, it must move into an empty valence band state. Therefore, when the electron stabilizes back down into
the valence band, it also effectively removes a hole. This process is called recombination. There are three basic types of recombination
in the bulk of a single-crystal semiconductor. These are:

Radiative recombination;
Auger recombination1; and
Shockley-Read-Hall recombination.2,3

These are described in the animation and text below.

Radiative (Band-to-Band) Recombination

Radiative recombination is the recombination mechanism that dominates in direct bandgap semiconductors. The light produced from a
light emitting diode (LED) is the most obvious example of radiative recombination in a semiconductor device. Concentrator and space
solar cells are typically made from direct bandgap materials (GaAs etc) and radiative recombination dominates. However, most
terrestrial solar cells are made from silicon, which is an indirect bandgap semiconductor and radiative recombination is extremely low
and usually neglected. The key characteristics of radiative recombination are:

In radiative recombination, an electron from the conduction band directly combines with a hole in the valence band and releases
a photon; and
The emitted photon has an energy similar to the band gap and is therefore only weakly absorbed such that it can exit the piece of
semiconductor.

Recombination Through Defect Levels

Recombination through defects, also called Shockley-Read-Hall or SRH recombination, does not occur in perfectly pure, undefected
material. SRH recombination is a two-step process. The two steps involved in SRH recombination are:

An electron (or hole) is trapped by an energy state in the forbidden region which is introduced through defects in the crystal
lattice. These defects can either be unintentionally introduced or deliberately added to the material, for example in doping the
material; and
If a hole (or an electron) moves up to the same energy state before the electron is thermally re-emitted into the conduction band,
then it recombines.

The rate at which a carrier moves into the energy level in the forbidden gap depends on the distance of the introduced energy level
from either of the band edges. Therefore, if an energy is introduced close to either band edge, recombination is less likely as the
electron is likely to be re-emitted to the conduction band edge rather than recombine with a hole which moves into the same energy
state from the valence band. For this reason, energy levels near mid-gap are very effective for recombination.

Auger Recombination

Auger Recombination involves three carriers. An electron and a hole recombine, but rather than emitting the energy as heat or as a
photon, the energy is given to a third carrier, an electron in the conduction band. This electron then thermalizes back down to the
conduction band edge.

Auger recombination is most important at high carrier concentrations caused by heavy doping or high level injection under
concentrated sunlight. In silicon-based solar cells (the most popular), Auger recombination limits the lifetime and ultimate efficiency.
The more heavily doped the material is, the shorter the Auger recombination lifetime.

The magnitude of the various recombination mechanisms is available at the PV Lighthouse Recombination Calculator

1. P. Auger, “Sur les rayons β secondaires produits dans un gaz par des rayons X”, C.R.A.S., vol. 177, pp. 169-171, 1923.
2. W. Shockley and Read, W. T., “Statistics of the Recombinations of Holes and Electrons”, Physical Review, vol. 87, p. 835,
1952.
3. R. N. Hall, “Electron-Hole Recombination in Germanium”, Phys. Rev., vol. 87, p. 387, 1952.

Lifetime
Add new comment
1 comment(s)
Español

Overview
 1. The lifetime of a semiconductor is contingent upon the recombination rate, which is dependent upon the concentration of
minority carriers.
2. The lifetime of the material takes into account the different types of recombination.
3. Lifetime is an indicator of the efficiency of a solar cell, and thus is a key consideration in choosing materials for solar cells.

If the number of minority carriers is increased above that at equilibrium by some transient external excitation (such as incident sun),
the excess minority carriers will decay back to that equilibrium carrier concentration due to and through the process of recombination.
A critical parameter in a solar cell is the rate at which recombination occurs. Such a process, known as the "recombination rate"
depends on the number of excess minority carriers. If for example, there are no excess minority carriers, then the recombination rate
must be zero. Two parameters that are integral to recombination rate are the minority carrier lifetime and the minority carrier diffusion
length. The first will be discussed here.

The minority carrier lifetime of a material, denoted by τn or τp, is the average time which a carrier can spend in an excited state after
electron-hole generation before it recombines. It is often just referred to as the "lifetime" and has nothing to do with the stability of the
material. Stating that "a silicon wafer has a long lifetime" usually means minority carriers generated in the bulk of the wafer by light or
other means will persist for a long time before recombining. Depending on the structure, solar cells made from wafers with long
minority carrier lifetimes will usually be more efficient than cells made from wafers with short minority carrier lifetimes. The terms
"long lifetime" and "high lifetime" are used interchangeably.

In low level injected material (where the number of minority carriers is less than the doping), the lifetime is related to the
recombination rate by:

where τ is the minority carrier lifetime, Δn is the excess minority carriers concentration and R is the recombination rate.

Auger lifetime is a function of the carrier concentration and is given by:

Where the auger coefficient, C, for silicon is typically given as: 1.66 × 10-30cm6/s 1, 2

1. R. A. Sinton and Swanson, R. M., “Recombination in highly injected silicon”, Electron Devices, IEEE Transactions on, vol.
34, pp. 1380 - 1389, 1987.
2. P. P. Altermatt, Sinton, R. A., and Heiser, G., “Improvements in numerical modelling of highly injected crystalline silicon
solar cells”, Solar Energy Materials and Solar Cells, vol. 65, pp. 149-155(7), 2001.

Diffusion Length
Add new comment
Español

Overview

1. Diffusion length is the average length a carrier moves between generation and recombination.
2. Semiconductor materials that are heavily doped have greater recombination rates and consequently, have shorter diffusion
lengths.
3. Higher diffusion lengths are indicative of materials with longer lifetimes and are, therefore, an important quality to consider with
semiconductor materials.

The second related parameter to recombination rate, the "minority carrier diffusion length," is the average distance a carrier can move
from point of generation until it recombines. As we shall see in the next chapter, the diffusion length is closely related to the collection
probability.
The minority carrier lifetime and the diffusion length depend strongly on the type and magnitude of recombination processes in the
semiconductor. For many types of silicon solar cells, SRH recombination is the dominant recombination mechanism. The
recombination rate will depend on the number of defects present in the material so that as doping the semiconductor increases the
defects in the solar cell. Doping will also increase the rate of SRH recombination. In addition, since Auger recombination is more
likely in heavily doped and excited material, the recombination process is itself enhanced as the doping increases. The method used to
fabricate the semiconductor wafer and the processing also have a major impact on the diffusion length.

Color chart of a high-efficiency multicrystalline PERL solar cell. The percentage refers to the fraction of light-generated carrier and the
variations between one region and another are due to variations in the diffusion length in the solar cell caused by the grain boundaries
in the multicrystalline material.

In silicon, the lifetime can be as high as 1 msec. For a single crystalline silicon solar cell, the diffusion length is typically 100-300 µm.
These two parameters give an indication of material quality and suitability for solar cell use.

The diffusion length is related to the carrier lifetime by the diffusivity according to the following formula:

where:
L is the diffusion length;
D is the diffusivity and
τ is the lifetime in seconds.

The diffusivity is a measure of how quickly a group of particles fill a space and it is described in more detail on the diffusion page The
diffusivity is typically given in cm2/s and the lifetime is in seconds so the above formula will give the diffusion length in cm. However,
a diffusion length in a silicon solar cell will typically be quoted in microns (µm). Multiply the result by 10,000 (104) to convert from
cm to µm

The following calculator provides a way of converting between lifetime and diffusion length using more familiar units. The diffusivity
can be found in the appendices.

Lifetime - Diffusion Length Calculator

Minority carrier lifetime, τb= µsec Minority

Carrier Diffusivity, D = cm²/s.
Minority carrier diffusion, L= µm.
Conversion from minority carrier diffusion length to lifetime

Conversion from minority carrier diffusion length to lifetime

Minority carrier diffusion, L= µm Minority

Carrier Diffusivity, D = cm²/s.
Minority carrier lifetime, τb= µsec .

Surface Recombination
Add new comment
Español

Overview

1. Areas of defect, such as at the surface of solar cells where the lattice is disrupted, recombination is very high.
2. Surface recombination is high in solar cells, but can be limited.
3. Understanding the impacts and the ways to limit surface recombination leads to better and more robust solar cell designs.

Any defects or impurities within or at the surface of the semiconductor promote recombination. Since the surface of the solar cell
represents a severe disruption of the crystal lattice, the surfaces of the solar cell are a site of particularly high recombination. The high
recombination rate in the vicinity of a surface depletes this region of minority carriers. As explained in the Diffusion Page, a localized
region of low carrier concentration causes carriers to flow into this region from the surrounding, higher concentration regions.
Therefore, the surface recombination rate is limited by the rate at which minority carriers move towards the surface. A parameter
called the "surface recombination velocity", in units of cm/sec, is used to specify the recombination at a surface. In a surface with no
recombination, the movement of carriers towards the surface is zero, and hence the surface recombination velocity is zero. In a surface
with infinitely fast recombination, the movement of carriers towards this surface is limited by the maximum velocity they can attain,
and for most semiconductors is on the order of 1 x 107 cm/sec.

The dangling bonds at the surface of semiconductor cause a high local recombination rate.

The defects at a semiconductor surface are caused by the interruption to the periodicity of the crystal lattice, which causes dangling
bonds at the semiconductor surface. The reduction of the number of dangling bonds, and hence surface recombination, is achieved by
growing a layer on top of the semiconductor surface which ties up some of these dangling bonds. This reduction of dangling bonds is
known as surface passivation.

The answer to a question "Why does limiting surface recombination lead to longer cell lifetimes?" is the following:

The lifetime of the material is contingent upon the concentration of minority carriers. Limiting surfacing recombination can lessen the
rate at which minority carriers are depleted. If the rate of minority carrier depletion can be limited, the lifetime of the material can be
extended.
Movement of Carriers in Semiconductors
Add new comment
Español

Overview

1. Carriers move freely about the semiconductor lattice in a random direction at a certain velocity determined by the temperature
and the mass of the carrier.
2. Carriers will continue in that direction until they collide with another semiconductor lattice atom.
3. There is no net overall movement of carriers in any direction.

Electrons in the conduction band and holes in the valence band are considered "free" carriers in the sense that they can move
throughout the semiconductor lattice that makes up the crystal structure of the material. A simple, but in most cases adequate
description of carrier movement views each carrier as moving in a random direction at a certain velocity.

The carrier moves in this random direction for a distance called the scattering length before colliding with a lattice atom. Once the
collision takes place, the carrier moves away in a different random direction.

A model of carrier movement is shown in the animation below.

The velocity of the carriers is determined by the temperature of the lattice. The thermal velocity is an average carrier velocity. Carriers
have a thermal velocity that is normally distributed around this average thermal velocity. Therefore, some carriers having a greater
velocity and some lower.

Barring certain circumstances which will be discussed in the following sections, there is no net movement of carriers in any
direction. Each direction of carrier movement is equally likely, therefore the motion of a carrier in one direction will eventually be
balanced by the movement of the carrier in the opposite direction. In the following animation, a carrier moves a distance equal to the
scattering distance in a random direction before it collides with a lattice atom (for clarity the lattice atoms are not shown). After
scattering off the lattice atoms, the carrier again moves in a random direction. The following animation has 5000 scattering events.
Although carriers in a semiconductor are in constant random motion, there is no net motion of carriers unless there is a
concentration gradient or an electric field, to be discussed in the next sections.

Diffusion
Add new comment
2 comment(s)
Español

Overview

1. When light is incident on a solar cell, carriers get generated at that surface, but not in the bulk of the solar cell. This creates a
carrier concentration gradient within the semiconductor
2. When a carrier concentration gradient exists in the semiconductor, through random motion, carriers will have a net movement
from areas of high carrier concentration to areas of low concentration in the process of diffusion.
3. With time, these carriers will diffuse throughout the cell until the concentration is uniform.

The constant random motion of carriers can lead to a net movement of carriers if one particular region has a higher concentration of
carriers than another region (a concentration gradient between the high carrier-concentration region and the low carrier-concentration
region). The net movement of carriers is therefore from areas of high concentration to low. If we let random movement do it's thing,
over time, the carriers will become evenly spread across the space through random motion alone. This can be seen in the
animation below.

The rate at which diffusion occurs depends on the velocity at which carriers move and on the distance between scattering events. It is
termed diffusivity and measured in cm2s-1. Values for silicon, the most used semiconductor material for solar cells, are given in the
appendix. Since raising the temperature will increase the thermal velocity of the carriers, diffusion occurs faster at higher temperatures.
A single particle in a box will eventually be found at any random location in the box.

Start Stop Reset

One major effect of diffusion is that, with time, it evens out the carrier concentrations in a device, such as those induced by generation
and recombination, without an external force being applied to the device. This is shown in the animation below in which one region of
the device has a high concentration of electrons and the other has a high concentration of holes. Due purely to the random movement
of carriers, the two concentrations will become uniform throughout the material, over time.

The holes (blue) and the electrons (red) move from areas of high concentration to low concentration w/in a semiconductor towards an
even distribution. The holes (colored blue) have a lower diffusivity than the electrons (colored red), and so take longer to fill the full
space.

Diffusion Equation Derivation

It is known from the molecular physics that the flux of diffusing particles is proportional to the concentration gradient.

One-dimensional diffusion equations

for electrons (n) and holes (p) can be written as follows:

where:

Jn and Jp = the diffusion current densities

q = electron charge

Dn and Dp = diffusion coefficients for electrons and holes

n and p = electron and hole concentrations

Equation of diffusion for carriers in the bulk of semiconductor

With time (t1, t2, t3), an initial pulse of electrons will diffuse.
Spreading of a pulse of electrons by diffusion. Arbitrary part of n(x) is divided into the segments of length equal to a mean free path for
the electrons. Concentration is supposed to be constant for every segment.

Rate of diffusion for electrons in semiconductors =

Electron flux density = number of electrons passing xo per unit time per unit area

Consider small segments of width l to the left and right of x0 and approximate the electron concentrations n1 and n2 in these segments
as uniform.

Therefore, the electron flux density from left to right =

The difference in electron concentration between the two points (n1 and n2) needs to be written in terms that we understand. We can
define both n1 and n2 using an excerpt of the above graph.
We assume l is very small, and therefore can use the slope at xo in order to determine the electron concentration (n) at xo ±

Substitute:
Simplify:

Cross out terms:

Combine terms:

Simplify:

Substitute back into original equation:

Combine l terms:

In order to approximate the electron concentration as x changes, assume that x is very small.
To do this, we take the limit as x → 0

Move constants in front of limit:

Take limit:

*this derivation can also be used for holes!

Redefine the constants and equation for electrons:

For ELECTRONS:

Dn is the electron diffusion coefficient with units cm2/s. The minus sign arises from the fact that the vector of the concentration
gradient is directed toward the increase of the concentration, while the particles diffuse to the area with lower concentration.
Direction of the concentration gradient is opposite to the direction of the carrier motion with the result that the

formula for the carrier flux density should have a minus sign.

Now, redefine the constants and equation for holes:

Substitute:

For HOLES:

Diffusion current density = carrier flux density multiplied by carrier charge (denoted by q)

For Electrons:
Simplify:

For holes:

Overview:

Continuity Equation

Continuity equations give the rate of carriers buildup in the bulk of semiconductor.
 ,

where U is carrier recombination rate, G - generation rate.

Currents entering and leaving a volume AΔx.

Consider the length dx of semiconductor and movement of holes through it. The net increase in hole concentration per unit time is the
difference between the flux of holes entering and leaving the volume AΔx plus generation rate and minus recombination rate.

In the derivative form

Finally, plugging in the diffusion equations one can get

Drift
Add new comment
Español

Overview

1. Carrier transport when an electric field is imposed on the semiconductor.

2. Electrons move in the net direction opposite of the electric field. Holes move in the net direction of the electric field.

As noted in the movement of carriers page, in the absence of an electric field, carriers move a certain distance at a constant velocity in
a random direction. However, in the presence of an electric field, superimposed on this random direction, and in the presence of
thermal velocity, carriers move in a net direction. There is an acceleration in the direction of the electric field if the carrier is a hole
or opposite to the electric field if the carrier is an electron. The acceleration in a given direction causes a net motion of carriers over a
certain net distance, as shown in the animation below. The direction of the carrier is obtained as a vector addition between its direction
and the electric field. The net carrier movement in the presence of an electric field is characterized by mobility, which varies between
different semiconductor materials. Values for silicon, the most commonly used semiconductor material for PV applications, are given
in the appendix.
The animation shows how the presence of an electric field will introduce a net distance into the movement of a carrier. In this
animation, the carrier is a hole so that it moves in the same direction as the electric field.

Transport due to the movement of carriers due to the presence of an electric field is called "drift transport". Drift transport is the type of
transport that occurs not only in semiconductor material, but also in metals. The following animation shows the motion of carriers in
random direction with, and without an electric field. The carrier in this case is an electron. Since the electron has a negative charge, it
will tend to move in the direction opposite to the electric field. Notice that in the majority of cases, the electron moves in the direction
opposite to the electric field. In some cases, for example if the electron follows a sequence of moves in the direction of the electric
field, the net motion may actually be in the direction of the electric field for a short distance.

The animation above shows a randomly moving electron with and without an electric field. The application of an electric field causes a
net movement of an electron towards the right. A positively charged hole would move to the left.

In the next animation, an intrinsic semiconductor with an equal number of electrons and holes is pictured. Without the electric field the
electrons and holes move around randomly in the semiconductor. When the field is turned on the electrons and holes drift in opposite
directions.
For clarity, the effect of the electric field is greatly exaggerated. In a typical semiconductor the electric field has only a very minor
effect on the random movement of the carriers.

Drift Equation. Conductivity and Mobility.

One-dimensional drift equation is given by the following formula.

where Jx is the current density in the x-direction, Ex - electric field applied in the x-direction, q - electron charge, n and p - electron and
hole concentrations, µn and µp - electron and hole mobilities.

To derive the drift equation let's consider the bulk of semiconductor.

If the electric field Ex is applied in the x-direction each electron experiences a net force which leads to additional acceleration in the
direction opposite to the direction of the field.

The net acceleration in the case of steady state current flow is balanced by the decelerations of the collision processes. If N(t) is the
number of electrons that have not undergone a collision by time t, then the rate of decrease N(t) is proportional to the number left
unscattered at t.

Where τ represents the mean time between scattering events.

The probability that an electron has a collision in dt is , then the differential change in px due to collisions in dt is

where n is the electron concentration.

And average momentum per electron is

The net drift speed is equal to

The current density is the number of electrons crossing the unit area per unit time

Where is the conductivity of a semiconductor and is the mobility of carriers.

Rearranging gives
Finally considering both hole and electron conduction

Formation of a PN-Junction
Add new comment
3 comment(s)
Español

Overview

1. Joining n-type material with p-type material causes excess electrons in the n-type material to diffuse to the p-type side and
excess holes from the p-type material to diffuse to the n-type side.
2. Movement of electrons to the p-type side exposes positive ion cores in the n-type side while movement of holes to the n-type
side exposes negative ion cores in the p-type side, resulting in an electron field at the junction and forming the depletion region.
3. A voltage results from the electric field formed at the junction.

P-n junctions are formed by joining n-type and p-type semiconductor materials, as shown below. Since the n-type region has a high
electron concentration and the p-type a high hole concentration, electrons diffuse from the n-type side to the p-type side. Similarly,
holes flow by diffusion from the p-type side to the n-type side. If the electrons and holes were not charged, this diffusion process would
continue until the concentration of electrons and holes on the two sides were the same, as happens if two gasses come into contact with
each other. However, in a p-n junction, when the electrons and holes move to the other side of the junction, they leave behind exposed
charges on dopant atom sites, which are fixed in the crystal lattice and are unable to move. On the n-type side, positive ion cores are
exposed. On the p-type side, negative ion cores are exposed. An electric field E forms between the positive ion cores in the n-type
material and negative ion cores in the p-type material. This region is called the "depletion region" since the electric field quickly
sweeps free carriers out, hence the region is depleted of free carriers. A "built-in" potential Vbi is formed at the junction due to E. The
animation below shows the formation of the E at the junction between n and p-type material.

P-N Junction Diodes

Add new comment
1 comment(s)
Español
Overview

1. P-N junction diode is integral for all electronic devices to operate, aggregating all forms of carrier transport, generation, and
recombination.
2. Majority carriers can diffuse across the P-N junction depletion region, even though the electric field impedes their crossing.
Minority carriers that reach the junction are swept across the depletion region due to drift.
3. At equilibrium, the net current (diffusion and drift current) is zero for both electrons and holes because the diffusion current is
equal and opposite to the drift current for both carriers.

P-n junction diodes form the basis not only of solar cells, but of many other electronic devices such as LEDs, lasers, photodiodes and
bipolar junction transistors (BJTs). A p-n junction aggregates the recombination, generation, diffusion and drift effects described in the
previous pages into a single device.

Carrier Movement in Equilibrium

A p-n junction with no external inputs represents an equilibrium between carrier generation, recombination, diffusion and drift in the
presence of the electric field in the depletion region. Despite the presence of the electric field, which creates an impediment to the
diffusion of carriers across the electric field, some carriers still cross the junction by diffusion. In the animation below, most majority
carriers which enter the depletion region move back towards the region from which they originated. However, statistically some
carriers will have a high velocity and travel in a sufficient net direction such that they cross the junction. Once a majority carrier
crosses the junction, it becomes a minority carrier. It will continue to diffuse away from the junction and can travel a distance on
average equal to the diffusion length before it recombines. The current caused by the diffusion of carriers across the junction is called a
diffusion current. In the animation below, watch the carriers in the depletion region and wait for carriers which cross the junction.
Remember that in an actual p-n junction the number and velocity of the carriers is much greater and that the number of carriers
crossing the junction are much larger.

Minority carriers which reach the edge of the diffusion region are swept across it by the electric field in the depletion region. This
current is called the drift current. In equilibrium the drift current is limited by the number of minority carriers which are thermally
generated within a diffusion length of the junction.

In equilibrium, the net current from the device is zero. The electron drift current and the electron diffusion current exactly balance out
(if they did not there would be a net buildup of electrons on either one side or the other of the device). Similarly, the hole drift current
and the hole diffusion current also balance each other out.

Bias of PN Junctions
 Add new comment
2 comment(s)
Español

Overview

1. Forward bias occurs when a voltage is applied across the solar cell such that the electric field formed by the P-N junction is
decreased. It eases carrier diffusion across the depletion region, and leads to increased diffusion current.
2. In the presence of an external circuit that continually provides majority carriers, recombination increases which constantly
depletes the influx of carriers into the solar cell. This increases diffusion and ultimately increases current across the depletion
region.
3. Reverse bias occurs when a voltage is applied across the solar cell such that the electric field formed by the P-N junction is
increased. Diffusion current decreases.

Semiconductor devices have three modes of operation:

1. Thermal Equilibrium

At thermal equilibrium there are no external inputs such as light or applied voltage. The currents balance each other out so there is no
net current within the device.

2. Steady State

Under steady state there are external inputs such as light or applied voltage, but the conditions do not change with time. Devices
typically operate in steady state and are either in forward or reverse bias.

3. Transient

If the applied voltage changes rapidly, there will be a short delay before the solar cell responds. As solar cells are not used for high
speed operation there are few extra transient effects that need to be taken into account.

Diodes under Forward Bias

Forward bias refers to the application of voltage across the device such that the electric field at the junction is reduced. By applying a
positive voltage to the p-type material and a negative voltage to the n-type material, an electric field with opposite direction to that in
the depletion region is applied across the device. Since the resistivity of the depletion region is much higher than that in the remainder
of the device (due to the limited number of carriers in the depletion region), nearly all of the applied electric field is dropped across the
depletion region. The net electric field is the difference between the existing field in the depletion region and the applied field (for
realistic devices, the built-in field is always larger than the applied field), thus reducing the net electric field in the depletion region.
Reducing the electric field disturbs the equilibrium existing at the junction, reducing the barrier to the diffusion of carriers from one
side of the junction to the other and increasing the diffusion current. While the diffusion current increases, the drift current remains
essentially unchanged since it depends on the number of carriers generated within a diffusion length of the depletion region or in the
depletion region itself. Since the depletion region is only reduced in width by a minor amount, the number of minority carriers swept
across the junction is essentially unchanged.
Carrier Injection and Forward Bias Current Flow

The increased diffusion from one side of the junction to the other causes minority carrier injection at the edge of the depletion region.
These carriers move away from the junction due to diffusion and will eventually recombine with a majority carrier. The majority
carrier is supplied from the external circuit and hence a net current flows under forward bias. In the absence of recombination, the
minority carrier concentration would reach a new, higher equilibrium concentration and the diffusion of carriers from one side of the
junction to the other would cease, much the same as when two different gasses are introduced. Initially, gas molecules have a net
movement from the high carrier concentration to the low carrier concentration region, but when a uniform concentration is reached,
there is no longer a net gas molecule movement. In a semiconductor however, the injected minority carriers recombine and thus more
carriers can diffuse across the junction. Consequently, the diffusion current which flows in forward bias is a recombination current. The
higher the rate of recombination events, the greater the current which flows across the junction.

The "dark saturation current" (I0) is an extremely important parameter which differentiates one diode from another. I0 is a measure of
the recombination in a device. A diode with a larger recombination will have a larger I0.

Reverse Bias

In reverse bias a voltage is applied across the device such that the electric field at the junction increases. The higher electric field in the
depletion region decreases the probability that carriers can diffuse from one side of the junction to the other, hence the diffusion current
decreases. As in forward bias, the drift current is limited by the number of minority carriers on either side of the p-n junction and is
relatively unchanged by the increased electric field. A small increase in the drift current is experienced due to the small increase in the
width of the depletion region, but this is essentially a second-order effect in silicon solar cells. In many thin film solar cells where the
depletion region is around half the thickness of the solar cell the change in depletion region width with voltage has a large impact on
cell operation.

Diode Equation
Add new comment

Overview

1.
2. I0 is directly related to recombination, and thus, inversely related to material quality.
3. Non-ideal diodes include an "n" term in the denominator of the exponent. N is the ideality factor, ranging from 1-2, that
increases with decreasing current.

Ideal Diodes

The diode equation gives an expression for the current through a diode as a function of voltage. The Ideal Diode Law, expressed as:

where:
I = the net current flowing through the diode;
I0 = "dark saturation current", the diode leakage current density in the absence of light;
V = applied voltage across the terminals of the diode;
q = absolute value of electron charge;
k = Boltzmann's constant; and
T = absolute temperature (K).
The "dark saturation current" (I0) is an extremely important parameter which differentiates one diode from another. I0 is a measure of
the recombination in a device. A diode with a larger recombination will have a larger I0. An excellent discussion of the recombination
parameter is in 1

Note that:

I0 increases as T increases; and

I0 decreases as material quality increases.

At 300K, kT/q = 25.85 mV, the "thermal voltage".

Non-Ideal Diodes

For actual diodes, the expression becomes:

where:
n = ideality factor, a number between 1 and 2 which typically increases as the current decreases.

The diode equation is plotted on the interactive graph below. Change the saturation current and watch the changing of IV curve. Note
that although you can simply vary the temperature and ideality factor the resulting IV curves are misleading. In the simulation it is
implied that the input parameters are independent but they are not. In real devices, the saturation current is strongly dependent on the
device temperature. Similarly, mechanisms that change the ideality factor also impact the saturation current. Temperature effects are
discussed in more detail on the Effect of Temperature page.

Dark saturation current density, I0 = 1.0e-12 A/cm2

10-14 10-10

ideality factor, n = 1.1 units

1 2

Changing the dark saturation current changes the turn on voltage of the diode. The ideality factor changes the shape of the diode. The
graph is misleading for ideality factor. It implies that increasing the ideality factor would increase the turn on voltage. In reality this
is not the case as any physical effect that increases the ideality factor would substantially increase the dark saturation current, I0, so that
a device with a high ideality factor would typically have a lower turn on voltage.

The diode law is illustrated for silicon on the following picture. Increasing the temperature makes the diode to "turn ON" at lower
voltages.

Temperature: 300 K, 27°C

200 K 400 K

The diode law for silicon - current changes with voltage and temperature. For a given current, the curve shifts by approximately 2
mV/°C. The light blue curve shows the effect on the IV curve if I0 does not change with temperature. In reality, I0 changes rapidly with
temperature resulting in the dark blue curve.

1. A. Cuevas, “The Recombination Parameter J0”, Energy Procedia, vol. 55, pp. 53 - 62, 2014.

Diode Equations for PV

Add new comment
1 comment(s)
Español

Ideal Diode Equation Derivation

Add new comment
Español

The ideal diode equation is one of the most basic equations in semiconductors and working through the derivation provides a solid
background to the understanding of many semiconductors such as photovoltaic devices. The objective of this section is to take the
concepts introduced earlier in this chapter and mathematically derive the current-voltage characteristics seen externally. The operation
of actual solar cells is typically treated as a modification to the basic ideal diode equation described here.

The derivation of the ideal diode equation is covered in many textbooks. The treatment here is particularly applicable to photovoltaics
and uses the concepts introduced earlier in this chapter. For simplicity we also assume that one-dimensional derivation but the concepts
can be extended to two and three-dimensional notation and devices.

The objective is to determine the current as a function of voltage and the basic steps are:

Solve for properties in depletion region

Solve for carrier concentrations and currents in quasi-neutral regions
 Find total current

At the end of the section there are worked examples.

Semiconductors are analyzed under three conditions:

Thermal equilibrium - where the device is in the dark an no voltage is applied to the terminals. This step is important as it
describes the basic band diagram.
Steady state - where a constant voltage or light intensity is applied to the device.
Transient - where the devices are changing with time and so, for example, parasitic capacitances are considered. As solar cells
are steady state devices we generally don't consider transient analysis under normal operation. It is sometimes used during
characterization.

The ideal diode model is a one dimensional model. The diode itself is three dimensional but the n-type and p-type regions are assumed
to be infinite sheets so the properties are only changing in one dimension. The one dimensional model greatly simplifies the equations.

Basic Equations
Add new comment
Español

Poisson's Equation

Poisson's Equation
Edit
Delete
Clone

Read more about Poisson's Equation

where E is the electric field, ρ is the charge density and ε is the material permittivity. This equation gives the basic relationship
between charge and electric field strength. In semiconductors we divide the charge up into four components: hole density, p, electron
density, n, acceptor atom density, NA and donor atom density, ND.

For the ideal diode derivation NA is assumed constant in the p-region and zero in the n-region. Similarly, ND is assumed constant in the
n-region and zero in the p-region.
Transport Equations

Transport Equations
Edit
Delete
Clone

Read more about Transport Equations

where Jn is the electron current density, μn is the electron mobility and Dn is the electron diffusivity. Similarly, Jp is the hole current
density, μp is the hole mobility and Dp is the hole diffusivity. q is the electronic charge and E is the electric field. Note that in this
section we use E, and interchangeably for the electric field. Its a mistake and we should be consistent and stick to E.

The transport equations describe how carriers move, i.e. the flow of carriers or current. Its easier to use the current density, J, in A/cm2
rather than the absolute current, I, in A since we are not concerned with the area of the device. I = J × Area and for a 1 cm2 device J
and I are equal.

The first term in each equation is for drift and the second term is for diffusion.

Continuity Equations

The continuity equation keeps track of all the carriers in terms of movement, generation and recombination. They are sometimes
termed 'book keeping' equations since they make sure that every carrier is accounted for.

General Conditions

where U is the recombinaton rate and G is the generation rate.

Solar cells operate in steady state and we are not concerned with transients or switching times. Under thermal equilibrium and steady
state conditions the carrier concentrations do not change with time so that:

Rearranging the equations above leads to:

Summary

We now have the five basic equations to solve:

Poisson's Equation
Edit
Delete
 Clone

Read more about Poisson's Equation

Transport Equations
Edit
Delete
Clone

Read more about Transport Equations

The equations are readily solved using numerical approaches1, and there are many device simulators that perform this task. By making
a few approximations it is also possible to solve the equations closed form as outlined on the following pages.

1. P. A. Basore, “Numerical modeling of textured silicon solar cells using PC-1D”, Electron Devices, IEEE Transactions on, vol.
37, pp. 337 -343, 1990.

Applying the Basic Equations to a PN Junction

Add new comment
Español

Overview

1. Approximation that assumes that the electric field is confined to some region of the device.
2. The one-dimensional device is divided into three regions with or without an electric field.
3. In a P-N junction, 2 regions having no electric field are on the ends of the device and are termed quasi-neutral regions. The
depletion region has an electric field and is located where the junction is.

Depletion approximation

The equations on the previous page are difficult to solve in closed form unless several simplifying assumptions are made. In addition to
the assumption of a one-dimensional device, the most central simplifying assumption in determining a closed form solution to the
above equations is the depletion approximation. The depletion approximation assumes that the electric field in the device is confined to
some region of the device. According to this assumption, the device can then be broken up into regions that have an electric field and
those that do not. This is shown below for a pn junction, where Regions I and III do not have an electric field (called quasi-neutral
regions or QNR) and Region II has an electric field (which is called space-charge or depletion region).
Schematic showing the regions with and without the electric field according to the depletion approximation. Later in the derivation a
change in coordinate system is used where the origin is at the edge of the depletion region and leads into the quasi-neutral regions.

General Procedure using the depletion approximation:

Divide the device into regions with an electric field and without an electric field.

1. Solve for electrostatic properties in the depletion region (Region II on the diagram). This solution depends on the doping profile
assumed. Here we will restrict the calculations to constant doping and an abrupt junction.
2. Solve for the carrier concentration and current in the quasi-neutral regions (Regions I and III on the diagram) under steady-state
conditions. The steps in this are:
a. Determine the general solution for the particular device. The general solution will depend only on the types of
recombination and generation in the device.
b. Find the particular solution, which depends on the surfaces and the conditions at the edges of the depletion region.
3. Find the relationship between the currents on one side of the depletion region and the currents on the other side. This depends on
the recombination/generation mechanisms in the depletion region.

Solving for Depletion Region

Add new comment
2 comment(s)
Español

Overview

1. Major assumptions: depletion approximation, no free carriers in this region, dopant concentration is constant.
2. Based on these assumptions, can use Poisson's Equation to develop a solution for the depletion region.
3. Can solve for both the maximum electric field and the total depletion width.

As stated on the previous page we need to make certain assumption to solve the diode equations analytically.

Assumptions are:

1. Depletion approximation: the electric field is confined to the junction region and there is no electric field in the quasi-neutral
regions.
2. No free carriers (n(x), p(x) = 0 ) in depletion region.
3. We can assume no free carriers since the electric field sweeps them out of the depletion region quickly. No free carriers means
(1) transport equations drop out and (2) no recombination or generation, so the continuity equation becomes
.
This means that Jn is constant across the depletion region. Similarly, Jp is also constant across the depletion region.
4. Abrupt or step doping profile (NA+, ND+ are constant in their respective regions).
5. All dopants are ionised ( NA+= NA, ND+ = ND).
6. One-dimensional device.

Solution

The only equation left to solve is Poisson’s Equation, with n(x) and p(x) =0, abrupt doping profile and ionized dopant atoms. Poisson’s
equation then becomes:

or ,

where

ε0 is the permittivity in free space, and εs is the permittivity in the semiconductor and -xp and xn are the edges of
the depletion region in the p- and n-type side respectively, measured from the physical junction between the two materials.
The electric field then becomes

The integration constants C1 and C2 can be determined by using the depletion approximation, which states that the electric field must
go to zero at the boundary of the depletion regions. This gives:

and

The maximum electric field occurs at the junction between the p- and n-type material. Further, we know that the electric field lines
must be continuous across the interface, such the electric field in the p-type side and the n-type side must equal each other at the
interface or when x = 0. Putting x = 0 in the above equation for electric field and setting the two values of E equal to each other gives:
NAxp = NDxn. This equation makes physical sense since it states that the total charge on one side of the junction must be the same as
the total charge on the other. In other words, if the electric field is confined to the depletion region, then the net charge in Region II
must be zero, and hence the negative charge and the positive charge must be equal. NA xp A is the total negative charge, since NA is the
charge density and xpA is the volume of the depletion region (A is the cross-sectional area and xp is the depth). Similarly, ND xn A is
the positive charge. The cross sectional area (A) is the same and cancels out.

(a) Doping concentration in a pn junction. The dotted lines are the actual net charge density (the tails are exaggerated) and the solid
line represents the assumed charge density in the depletion approximation. (b) The electric field in a pn junction.

The graphs above are an illustration of Poisson's equation that we started with where the charge is slope of the electric field graph:

Possion's Equation (simple)

Edit
Delete
Clone

Read more about Possion's Equation (simple)

To find the voltage as a function of distance, we integrate the equation for the electric field.
We are usually interested in the potential difference across the junction and can arbitrarily set one side to zero. Here we define the
voltage on the p-type side as zero, such that at x= −xp, V=0. This gives the constant C3 as:

, which gives

We can find C4 by using the fact that the potential on the n-type side and p-type side are identical at the interface, such that:

or

Overall, V(x) is:

The total voltage is plotted below.

Plot of the voltage across a pn junction, assuming that the voltage on the p-type side is zero.

The maximum voltage across the junction is a x= xn, which is:

This voltage is also equal to the built-in voltage across the pn junction, V0, (which we can find from the difference in Fermi-levels
between the n and p-type material), giving

Using in the above equation and rearranging allows xp and xn to be determined. They are:

and

From these equations we can get the maximum electric field:

 ,

and the total width of the depletion region

Ideal Diode - width of the depletion region

Edit
Delete
Clone

Read more about Ideal Diode - width of the depletion region

and

Where is the built-in voltage and is calculated separately.

Solving for Quasi Neutral Regions

Add new comment
1 comment(s)
Español

Overview

1. Major assumptions: depletion approximation, this is the region of no electric field. Therefore, Poisson's equation equal to zero,
and drift is zero.
2. When solving for current, only consider diffusion, recombination, and generation. Diffusion has been previously solved for.
Recombination and generation are solved for using the continuity equations.
3. The general solution for the continuity equations gives the recombination (U) and generation (G) under certain situations within
the solar cell. Based on the region's boundary conditions, the specific particular solution is given.

In Regions I and III on the previous page, the depletion approximation states that there is no electric field. Therefore, Poisson’s
Equation becomes zero, and the drift term in the transport equation (q μ nnE or q μ p pE ) also becomes zero. Minority carriers in the
quasi-neutral regions move only by diffusion.

The transport equations then become:

(1a)

(1b)

The continuity equations remain:

(2a)

(2b)
 For electrons, differentiating (1a) above and substituting For holes, differentiating (1b) above and substituting into
into (2a) above gives: (2b) above gives:

, or , or

(3a) (3b)

Equations (3a) and (3b) are general under steady state conditions as long as the depletion approximation holds (which assumes low
injection) and as long as drift and diffusion are the only transport mechanisms. For steady state and low injection, these are the
equations that must be solved to determine the electrical characteristics, of the IV equation, of a device. The solution to these second
order differential equations involves firstly finding equations for U and G to determine the general solution to the differential equation,
and secondly determining boundary conditions to find the particular solution.

Dominance of the minority carrier currents

In either Region I or Region III, the total current consists of two current components, Jn (the current composed of electrons) and Jp (the
current composed of holes). However, we will solve only for the minority carrier current in each material, since the recombination of
minority carriers controls the current flow. Later, we will also calculate the majority carrier currents Jp (in p-type material) based on
the fact that the total current is constant.

Finding The Recombination Rate: Assumption of Low Injection

The general form of the recombination rate for SRH recombination is:

where n1, p1 are the numbers of carriers in recombination sites, n0, p0 are the electron and hole concentrations in equilibrium and τp0
and τn0 are the minority carrier lifetimes for holes and electrons.

In p-type material and under low injection (low bias) conditions, p >> n also p ≈ p0. Further, we will assume that n >> n1 and p >> p1
and that the lifetimes do not vary dramatically in n- and p-type material (i.e., that τn0p >> τp0n). The above equation then reduces to
the recombination rate for electrons in p-type material as:

This is the low injection form of the recombination rate and it is commonly used in closed form solutions of pn junctions as it greatly
simplifies the mathematics.

For holes in n-type material,

Finding the Generation Rate:

In general G will be given by the equation:

Generation Rate
Edit
Delete
Clone
 Read more about Generation Rate

where N0 = photon flux at the surface (photons/unit-area/sec.);

α = absorption coefficient; and
x = distance into the material.

Note that N0 may vary with time and may change with wavelength, such that in general N0 will be a function of time and wavelength
denoted by N0( λ,t) , which also makes the generation rate G a function of wavelength and time. However, here we are solving only for
steady state and ignore any time dependence. Wavelength dependence is often included by summing the result obtained at one
wavelength over all wavelengths of interest. The exponential form of the generation rate generally makes the differential equation to be
solved a nonhomogeneous second order differential equation. Therefore, approximation to the generation rate are often that the
generation is constant (valid when the dimensions of interest are small compared to α-1), that there is an impulse generation at the
surface, or that the generation is zero.

Finding the general solution

The general solution to the differential equation in (3a) and (3b) will depend on the equation for U and G. There are several common
general solutions. These are shown below, where is any function dependent only on x, and for the pn junction equations will
usually correspond to one of n(x), p(x), Δn(x) or Δp(x) . A and B are constants that need to be determined by the boundary conditions. C
and K are semiconductor or device constants. C for pn junction equations is usually Ln or Lp (the minority carrier diffusions length)
and K is often a constant generation term, G.

Differential equation of the form General solution When used

Bulk recombination, no
generation

Bulk recombination,
constant geneneration

Zero recombination and

generation

Zero recombination,
constant generation

Finding the particular solution (all equations will be for electrons in p-type material)

The particular solution depends on the conditions of each region at the edges or the regions. For many semiconductor devices, at least
one edge will be a pn junction, and hence Boundary Condition ?1 below applies to many semiconductor devices.

1. Boundary condition at the edge of a depletion region of a pn junction:

In p-type material: or

In n-type material: or
2. Possible boundary conditions at a semiconductor surface

The other boundary condition may depend on the surface of the device. The surface recombination velocity, Sr, determines the
conditions at the surface. Some common boundary conditions for surfaces are listed below for p-type material.

Location of surface Boundary Description Equation

 Surface is far away from the The minority carrier concentration must be finite as x ->
n(x->∞) = finite
junction (W>>Ln) ∞

Surface is within a few Surface recombination is “infinitely” fast (Sr = ∞) All

Δn (x=W) = 0
diffusion lengths of junction carriers that reach the surface recombine.

Surface is within a few

Surface recombination is finite.
diffusion lengths of junction

Surface next to a light Impulse light generation at surface with no surface

generation source recombination

Finding diffusion currents in Regions I and III

Once we have a form of n(x) and p(x) from the procedures described in 2(b) and 2(c), we can readily find the minority carrier currents
by using the general equations:

and

Finding Total Current

Add new comment
Español

Overview

1. Total current can only be solved, assuming it is constant over the distance of the solar cell, if there is not a contact to inject or
extract carriers and if the cell is operating under steady state conditions.
2. Current in region I and II is known, but the distance variable is different for the two regions. Therefore we must take this into
account before we solve for total current.
3. Total current = current at depletion region edges (current in region I and region II after accounting for the distance variable
difference).

To find the total current, we note that the TOTAL current in the device must be constant, independent of distance as long as there is not
a contact that can extract or inject carriers and as long as the device is under steady state conditions. This can be shown by:

Since each electron generates a hole and each recombining electron also uses up a hole, Un = Up and Gn= Gp so that the derivative of
JT is 0 and JT is a constant. Physically, the continuity equation is stating that the total number of electrons and holes cannot change in
the semiconductor (in steady state), and hence the total current also cannot change. Therefore if we find JT anywhere in the device, we
have found it everywhere in the device. It is most convenient to find the total current at the edges of the depletion regions. Since we
know the currents in Regions I and III, to calculate the total current, we need to do two things. (1) Account for the fact that the distance
variable x is not the same in Region I and Region III and (2) find the current at the depletion region edges.
Making the distance variable the same

In our solutions, the distance variable, x, in the above equations is usually not the same for the different regions of the device.
Typically, we define x' in Region I (here p-type region with an electron minority carrier current) as the distance from depletion region
edge and increasing further into Region I. The other distance variable x, is defined as zero at the other depletion region edge and is
increasing into Region III (here n-type material with a hole minority carrier current). Using these definitions, the transport equations
are:

and

Since x = - x' - W ,

and then the current becomes or

Current across the depletion region

Previously, we stated that the generation was zero and the number of free carriers was small, so the recombination could also be
neglected the depletion region. As previously stated, under these conditions, the change in current across the depletion region is zero
and we can find the total current just as the sum of the currents at the edges of Region I and III, as shown below:

A more accurate solution includes the change in Jn and Jp across the depletion region, and we find the total current by:

where ΔJp-dep is the change in Jp across the depletion region. We could solve for ΔJn-dep via the continuity

equation. The continuity equations, repeated below, give current dependence on recombination and generation.

and in the depletion region this becomes

Often, the recombination term is ignored and G is assumed to be a constant, such that

Example 1: General Solution for Wide Base P-N Junction

Add new comment
1 comment(s)

Overview

1. Situation where the solar cell surface is far enough away from the junction edges that recombination properties of carriers
injected into the quasi-neutral region under forward bias are not impacted.
2. The boundary conditions for the wide base diode particular solution are:
 (1) At the edge of the depletion region,

(2) minority concentration is finite (B=0)

3. Total Current is therefore:

A wide base junction is one in which the surface are far away enough from the junction edges such that they to not impact the
recombination properties of carriers injected into the QNR under forward bias.

Step 1: Solve for properties in depletion region

As in most devices, the solution for the electrostatic properties in the depletion region does not change, and so is not repeated.

Step 2: Solve for carrier concentrations and currents in quasi-neutral regions

The solution below is shown in detail only for the n-type material (in which there is a hole current).

Find U and G

We will set G equal to a constant and in the n-type material (in p-type material, ).

Find general solution

Using low injection recombination and constant generation gives the equation:

Note that ,

since

(where pn0 is a constant), so the derivative (and second derivative) of Δp(x) is the same as the derivative of p(x). In addition for
simplicity, we introduce a variable change using: .

The overall differential equation now becomes:

or

which has the general solution:

For electrons (p-type material), the differential equations and solutions are:

and
Particular solution for wide base diode

We need two boundary conditions these are:

(1) At the edge of the depletion region,

(2) The minority carrier concentration must be finite even as x tends to infinity. This can only be achieved if B = 0.

Since B = 0, the general solution for holes then becomes

at x = 0

Rearranging gives:

Plugging A back in gives:

or

The equation for electrons in p-type material, Δn(x') , can be similarly derived as:

This is plotted below for G=0.

Differentiating and plugging into equation for current gives:

Making the change from x to x' gives

Step 3: Finding total current

The change in the current across the depletion region is:

Assuming that there is no generation and recombination, then ΔJn = 0 and

This case is shown in the graph below.

If there is a constant generation across the depletion region, then , where xn is the depletion width in the p-type
material and xn +xp = W.

Jn at the edge of the depletion region in the p-type material is:

Jn at the edge of the depletion region in the n-type material is:

An analogous equations exists for Jp, and the total current is:

Typically, we write the equation in the form:

or
where

Example 2: General Solution for Narrow Base Diode

Add new comment

Overview

1. Situation where the quasi-neutral region in the solar cell is small, and therefore there is no recombination.
2. The boundary conditions for the narrow base diode particular solution are:

(1)

(2)

Step 1: Solve for properties in depletion region

As in most devices, the solution for the electrostatic properties in the depletion region does not change, and so is given here.

Step 2: Solve for carrier concentration and current in quasi-neutral regions

Find U and G

We will set G equal to a constant and U=0.

Find general solution

We still start out with the same equation derived from the continuity equations. However, in this case the recombination is zero, so the
equation becomes:

The general solution is:

Particular solution for narrow base diode with high recombination at edges

We need boundary conditions and these are:

At the edge of the depletion region

The excess minority carrier concentration Dn must be zero at x = W, or Dn(x = W)=0.

The first boundary condition gives :

The second boundary conditions gives :

which simplifies to

Substituting these equations into the general solutions gives the equation for the carrier concentration:

The current is found by differentiating the carrier concentration:

Simplifying this gives:

Step 3: Find total current

The change in the current across the depletion region is given by the general equation:

If there is a constant generation across the depletion region and no recombination, then

, where xn is the depletion width in the p-type material.

Jn at the edge of the depletion region in the p-type material is:

Jn at the edge of the depletion region in the n-type material is:

An analogous equations exists for Jp, and the total current is:

Summary
Add new comment

Surface Condition U (for n-type J0

material)
 Wide-base
Far away
diode

Narrow Base
Infinite Dp = 0 U=0
diode

General case Infinite Dp = 0

Chapter 3 Quiz
Questions 6
Attempts allowed Unlimited
Available Always
Pass rate 75 %
Backwards navigation Allowed
Start Quiz

Solar Cell Operation

Add new comment
Español

Solar Cell Operation module

Ideal Solar Cells

Add new comment
6 comment(s)
Español

Solar Cell Structure

Add new comment
2 comment(s)
Español

A solar cell is an electronic device which directly converts sunlight into electricity. Light shining on the solar cell produces both a
current and a voltage to generate electric power. This process requires firstly, a material in which the absorption of light raises an
electron to a higher energy state, and secondly, the movement of this higher energy electron from the solar cell into an external circuit.
The electron then dissipates its energy in the external circuit and returns to the solar cell. A variety of materials and processes can
potentially satisfy the requirements for photovoltaic energy conversion, but in practice nearly all photovoltaic energy conversion uses
semiconductor materials in the form of a p-n junction.
Cross section of a solar cell.

The basic steps in the operation of a solar cell are:

the generation of light-generated carriers;

the collection of the light-generated carries to generate a current;
the generation of a large voltage across the solar cell; and
the dissipation of power in the load and in parasitic resistances.

Light Generated Current

Add new comment
4 comment(s)
Español

The generation of current in a solar cell, known as the "light-generated current", involves two key processes. The first process is the
absorption of incident photons to create electron-hole pairs. Electron-hole pairs will be generated in the solar cell provided that the
incident photon has an energy greater than that of the band gap. However, electrons (in the p-type material), and holes (in the n-type
material) are meta-stable and will only exist, on average, for a length of time equal to the minority carrier lifetime before they
recombine. If the carrier recombines, then the light-generated electron-hole pair is lost and no current or power can be generated.

A second process, the collection of these carriers by the p-n junction, prevents this recombination by using a p-n junction to spatially
separate the electron and the hole. The carriers are separated by the action of the electric field existing at the p-n junction. If the light-
generated minority carrier reaches the p-n junction, it is swept across the junction by the electric field at the junction, where it is now a
majority carrier. If the emitter and base of the solar cell are connected together (i.e., if the solar cell is short-circuited), the light-
generated carriers flow through the external circuit. The ideal flow at short circuit is shown in the animation below.
The ideal short circuit flow of electrons and holes at a p-n junction. Minority carriers cannot cross a semiconductor-metal boundary
and to prevent recombination they must be collected by the junction if they are to contribute to current flow.

Collection Probability
Add new comment
Español

The "collection probability" describes the probability that a carrier generated by light absorption in a certain region of the device will
be collected by the p-n junction and therefore contribute to the light-generated current, but probability depends on the distance that a
light-generated carrier must travel compared to the diffusion length. Collection probability also depends on the surface properties of
the device. The collection probability of carriers generated in the depletion region is unity as the electron-hole pair are quickly swept
apart by the electric field and are collected. Away from the junction, the collection probability drops. If the carrier is generated more
than a diffusion length away from the junction, then the collection probability of this carrier is quite low. Similarly, if the carrier is
generated closer to a region such as a surface with higher recombination than the junction, then the carrier will recombine. The impact
of surface passivation and diffusion length on collection probability is illustrated below.
 Base base diffusion length, Lbase = 100.0 µm

10 1000

Rear surface recombination, Srear = 1000.0 cm/²

1 106

Emitter diffusion length, Lemitter = 1.0 µm

0.1 10

Rear surface recombination, Srear = 1000.0 cm/²

1 106

Calculation of the collection probability. The diffusion length in the emitter is in red and in the base is in blue. Ln denotes the minority
carrier diffusion length and SRV is the surface recombination velocity.

The collection probability in conjunction with the generation rate in the solar cell determine the light-generated current from the solar
cell. The light-generated current is the integration over the entire device thickness of the generation rate at a particular point in the
device, multiplied by the collection probability at that point. The equation for the light-generated current density (JL), with an arbitrary
generation rate (G(x))and collection probability (CP(x)), is shown below, as is the generation rate in silicon due to the AM1.5 solar
spectrum:

where:
q is the electronic charge;
W is the thickness of the device;
α(λ) is the absorption coefficient;
H0 is the number of photons at each wavelength.

The light-generated current depends on the generation of carriers and the collection probability of these carriers.
The generation profile in silicon due to the Am1.5 spectrum. Note that the carrier generation is the highest at the surface of the solar
cell, thus making photovoltaic devices very sensitive to surface properties.

A non-uniform collection probability will cause a spectral dependence in the light-generated current. For example, at the surfaces, the
collection probability is lower than in the bulk. Comparing the generation rates for blue, green and infrared light below, blue light is
nearly completely absorbed in the first few tenths of a micron in silicon. Therefore, if the collection probability at the front surface is
low, any blue light in the solar spectrum does not contribute to the light-generated current.

wavelength =

300 1100

The creation of electron-hole pairs for different wavelengths of light in silicon. Blue light of 0.45 µm has a high absorption coefficient
of 105 cm-1 and is therefore absorbed very close to the front surface. Red light at 0.8 µm and an absorption coefficient of 103 cm-1 is
absorbed deeper into the cell. Infrared light at 1.1 µm with an absorption coefficient of 103 cm-1 is barely absorbed since it is close to
the band gap of silicon.

Quantum Efficiency
 Add new comment
Español

The "quantum efficiency" (Q.E.) is the ratio of the number of carriers collected by the solar cell to the number of photons of a given
energy incident on the solar cell. The quantum efficiency may be given either as a function of wavelength or as energy. If all photons
of a certain wavelength are absorbed and the resulting minority carriers are collected, then the quantum efficiency at that particular
wavelength is unity. The quantum efficiency for photons with energy below the band gap is zero. A quantum efficiency curve for an
ideal solar cell is shown below.

The quantum efficiency of a silicon solar cell. Quantum efficiency is usually not measured much below 350 nm as the power from the
AM1.5 contained in such low wavelengths is low.

While quantum efficiency ideally has the square shape shown above, the quantum efficiency for most solar cells is reduced due to
recombination effects. The same mechanisms which affect the collection probability also affect the quantum efficiency. For example,
front surface passivation affects carriers generated near the surface, and since blue light is absorbed very close to the surface, high front
surface recombination will affect the "blue" portion of the quantum efficiency. Similarly, green light is absorbed in the bulk of a solar
cell and a low diffusion length will affect the collection probability from the solar cell bulk and reduce the quantum efficiency in the
green portion of the spectrum. The quantum efficiency can be viewed as the collection probability due the generation profile of a single
wavelength, integrated over the device thickness and normalized to the incident number of photons.

The "external" quantum efficiency of a silicon solar cell includes the effect of optical losses such as transmission and reflection.
However, it is often useful to look at the quantum efficiency of the light left after the reflected and transmitted light has been lost.
"Internal" quantum efficiency refers to the efficiency with which photons that are not reflected or transmitted out of the cell can
generate collectable carriers. By measuring the reflection and transmission of a device, the external quantum efficiency curve can be
corrected to obtain the internal quantum efficiency curve.

The animation below shows the effect on surface recombination and diffusion length on the internal quantum efficiency of a solar cell.
The emitter thickness is 1 µm, the base thickness is 300 µm, the emitter diffusivity is 4 cm2s-1 and the base diffusivity is 27 cm2s-1.
For base diffusion lengths greater than the device thickness of 300 µm the rear surface recombination velocity has a large effect on QE.
For low diffusion lengths recombination at the rear surface has no effect.
 Base diffusion length, Lbase = 100.0 µm

10 1000

Rear surface recombination, Srear = 100.0 cm/s

1.0 1000000

Emitter diffusion length, Lemitter = 1.0 µm

0.1 10

Front surface recombination Sfront = 100.0 cm/s

1.0 1000000

Internal quantum efficiency. Lp is the emitter diffusion length (µm), Sp is the front surface recombination velocity (cm/s), Ln is the base
diffusion length (µm), Sn is the rear surface recombination velocity (cm/s).

Spectral Response
Add new comment
4 comment(s)
Español

The spectral response is conceptually similar to the quantum efficiency. The quantum efficiency gives the number of electrons output
by the solar cell compared to the number of photons incident on the device, while the spectral response is the ratio of the current
generated by the solar cell to the power incident on the solar cell. A spectral response curve is shown below.
The spectral response of a silicon solar cell under glass. At short wavelengths below 400 nm the glass absorbs most of the light and the
cell response is very low. At intermediate wavelengths the cell approaches the ideal. At long wavelengths the response fall back to
zero. Silicon is an indirect band gap semiconductor so there is not a sharp cut off at the wavelength corresponding to the band gap (Eg
= 1.12 eV).

The ideal spectral response is limited at long wavelengths by the inability of the semiconductor to absorb photons with energies below
the band gap. This limit is the same as that encountered in quantum efficiency curves. However, unlike the square shape of QE curves,
the spectral response decreases at small photon wavelengths. At these wavelengths, each photon has a large energy, and hence the ratio
of photons to power is reduced. Any energy above the band gap energy is not utilized by the solar cell and instead goes to heating the
solar cell. The inability to fully utilize the incident energy at high energies, and the inability to absorb low energies of light represents a
significant power loss in solar cells consisting of a single p-n junction.

The spectral response and the quantum efficiency are both used in solar cell analysis and the choice depends on the application. The
spectral response uses the power of the light at each wavelength whereas the quantum efficiency uses the photon flux. Converting QE
to SR is done with the following formula:

Simplifying gives:

QE to SR in nm
Edit
Delete
Clone

Read more about QE to SR in nm

or

QE to SR in um
Edit
Delete
Clone
 Read more about QE to SR in um

The graph below is similar to the quantum efficiency graph shown on the previous page except that the spectral response is plotted
instead of the quantum efficiency.

Lbase_log =

10 1000

Srear_log =

1.0 1000000

Lemitter_log =

0.1 10

Sfront_log =

1.0 1000000

Internal spectral response. Lp is the emitter diffusion length (µm), Sp is the front surface recombination velocity (cm/s), Ln is the base
diffusion length (µm), Sn is the rear surface recombination velocity (cm/s).

Solar Cell Parameters

Add new comment
3 comment(s)
Español

The photovoltaic effect

Add new comment
2 comment(s)
 Español

The collection of light-generated carriers does not by itself give rise to power generation. In order to generate power, a voltage must be
generated as well as a current. Voltage is generated in a solar cell by a process known as the "photovoltaic effect". The collection of
light-generated carriers by the p-n junction causes a movement of electrons to the n-type side and holes to the p-type side of the
junction. Under short circuit conditions, there is no build up of charge, as the carriers exit the device as light-generated current.

However, if the light-generated carriers are prevented from leaving the solar cell, then the collection of light-generated carriers causes
an increase in the number of electrons on the n-type side of the p-n junction and a similar increase in holes in the p-type material. This
separation of charge creates an electric field at the junction which is in opposition to that already existing at the junction, thereby
reducing the net electric field. Since the electric field represents a barrier to the flow of the forward bias diffusion current, the reduction
of the electric field increases the diffusion current. A new equilibrium is reached in which a voltage exists across the p-n junction. The
current from the solar cell is the difference between IL and the forward bias current. Under open circuit conditions, the forward bias of
the junction increases to a point where the light-generated current is exactly balanced by the forward bias diffusion current, and the net
current is zero. The voltage required to cause these two currents to balance is called the "open-circuit voltage". The following
animation shows the carrier flows at short-circuit and open-circuit conditions.

Simulation of carrier flows in a solar cell under equilibrium, short-circuit current and open-circuit voltage conditions. Note the
different magnitudes of currents crossing the junction. In equilibrium (i.e. in the dark) both the diffusion and drift current are small.
Under short circuit conditions, the minority carrier concentration on either side of the junction is increased and the drift current, which
depends on the number of minority carriers, is increased. Under open circuit conditions, the light-generated carriers forward bias the
junction, thus increasing the diffusion current. Since the drift and diffusion current are in opposite direction, there is no net current
from the solar cell at open circuit.

IV Curve
Add new comment
Español

The IV curve of a solar cell is the superposition of the IV curve of the solar cell diode in the dark with the light-generated current.1 The
light has the effect of shifting the IV curve down into the fourth quadrant where power can be extracted from the diode. Illuminating a
cell adds to the normal "dark" currents in the diode so that the diode law becomes:
where IL = light generated current.

The effect of light on the current-voltage characteristics of a p-junction.

The equation for the IV curve in the first quadrant is:

Ideal Solar Cell

Edit
Delete
Clone

Read more about Ideal Solar Cell

The -1 term in the above equation can usually be neglected. The exponential term is usually >> 1 except for voltages below 100 mV.
Further, at low voltages, the light generated current IL dominates the I0 (...) term so the -1 term is not needed under illumination.

Ideal Solar Cell (simplified)

Edit
Delete
Clone

Read more about Ideal Solar Cell (simplified)

Plotting the above equation gives the IV curve below with the relevant points on the curve labeled and discussed in more detail on the
following pages. The power curve has a a maximum denoted as PMP where the solar cell should be operated to give the maximum
power output. It is also denoted as PMAX or maximum power point (MPP) and occurs at a voltage of VMP and a current of IMP.
 Light
Biased IV
Curve
Calculator

Input Parameters

Dark
Saturation
Current,

A Light
Generated
Current,

A Ideality
Factor,

Temperature,
T=
Current voltage (IV) cure of a solar cell. To get the maximum power output of a solar cell it needs to operate at the
maximum
K Voltage, V power point, PMP.
=

Results

Current, A =

Several important parameters which are used to characterize solar cells are discussed in the following pages. The short-circuit current
(ISC), the open-circuit voltage (VOC), the fill factor (FF) and the efficiency are all parameters determined from the IV curve.

Rearranging the equation above gives the voltage in terms of current:

$$ V = \frac{n k T}{q} ln \left(\frac{I_L – I}{I_0}\right) $$

When I > IL the number in side the ln( ) is negative and undefined. So what happens in reality? The solar cell goes into reverse bias
(negative voltage) and either the non-idealities in the solar cell limit the voltage or the supply limits the voltage. In either case, the solar
cell will dissipate power. If there is no limit on the supply then a solar cell close to ideal (very high RSHUNT in reverse bias) will be
destroyed almost instantly. Other cells will be destroyed due to heating. The problem of power dissipation in solar cells in reverse bias
is covered in the module chapter and in particular the use of bypass diodes.

1. F. A. Lindholm, Fossum, J. G., and Burgess, E. L., “Application of the superposition principle to solar-cell analysis”, IEEE
Transactions on Electron Devices, vol. 26, pp. 165–171, 1979.

Short-Circuit Current
Add new comment
1 comment(s)
Español

The short-circuit current is the current through the solar cell when the voltage across the solar cell is zero (i.e., when the solar cell is
short circuited). Usually written as ISC, the short-circuit current is shown on the IV curve below.

The short-circuit current is due to the generation and collection of light-generated carriers. For an ideal solar cell at most moderate
resistive loss mechanisms, the short-circuit current and the light-generated current are identical. Therefore, the short-circuit current is
the largest current which may be drawn from the solar cell.
The short-circuit current depends on a number of factors
which are described below:

the area of the solar cell. To remove the dependence of

the solar cell area, it is more common to list the short-
circuit current density (Jsc in mA/cm2) rather than the
short-circuit current;
the number of photons (i.e., the power of the incident
light source). Isc from a solar cell is directly dependant
on the light intensity as discussed in Effect of Light
Intensity;
the spectrum of the incident light. For most solar cell
measurement, the spectrum is standardised to the
AM1.5 spectrum;
the optical properties (absorption and reflection) of
the solar cell (discussed in Optical Losses); and
the collection probability of the solar cell, which IV curve of a solar cell showing the short-circuit current.
depends chiefly on the surface passivation and the
minority carrier lifetime in the base.

When comparing solar cells of the same material type, the most critical material parameter is the diffusion length and surface
passivation. In a cell with perfectly passivated surface and uniform generation, the equation for the short-circuit current can be
approximated as:

where G is the generation rate, and Ln and Lp are the electron and hole diffusion lengths respectively. Although this equation makes
several assumptions which are not true for the conditions encountered in most solar cells, the above equation nevertheless indicates
that the short-circuit current depends strongly on the generation rate and the diffusion length.

Silicon solar cells under an AM1.5 spectrum have a maximum possible current of 46 mA/cm2. Laboratory devices have measured
short-circuit currents of over 42 mA/cm2, and commercial solar cell have short-circuit currents between about 28 mA/cm2 and 35
mA/cm2.

In an ideal device every photon above the bandgap gives one charge carrier in the external circuit so the highest current is
for the lowest bandgap.

Illuminated Current and Short Circuit Current (IL or Isc ?)

IL is the light generated current inside the solar cell and is the correct term to use in the solar cell equation. At short circuit conditions
the externally measured current is Isc. Since Isc is usually equal to IL, the two are used interchangeably and for simplicity and the solar
cell equation is written with Isc in place of IL. In the case of very high series resistance (> 10 Ωcm2) Isc is less than IL and writing the
solar cell equation with Isc is incorrect.

Another assumption is that the illumination current IL is solely dependent on the incoming light and is independent of voltage across
the cell. However, IL varies with voltage in the case of drift-field solar cells and where carrier lifetime is a function of injection level
such as defected multicrystalline materials.

Open-Circuit Voltage
Add new comment
3 comment(s)
Español

The open-circuit voltage, VOC, is the maximum voltage available from a solar cell, and this occurs at zero current. The open-circuit
voltage corresponds to the amount of forward bias on the solar cell due to the bias of the solar cell junction with the light-generated
current. The open-circuit voltage is shown on the IV curve below.

IV curve of a solar cell showing the open-circuit voltage.

An equation for Voc is found by setting the net current equal to zero in the solar cell equation to give:

A casual inspection of the above equation might indicate that VOC goes up linearly with temperature. However, this is not the case as
I0 increases rapidly with temperature primarily due to changes in the intrinsic carrier concentration ni. The effect of temperature is
complicated and varies with cell technology. See the page “Effect of Temperature” for more details

VOC decreases with temperature. If temperature changes, I0 also changes.

Open Circuit Voltage Calculator 1

Dark Saturation Current, I0 = A Light

Generated Current, IL = A Ideality Factor,
n= Temperature, T = K
Open Circuit Voltage, VOC = V

The above equation shows that Voc depends on the saturation current of the solar cell and the light-generated current. While Isc
typically has a small variation, the key effect is the saturation current, since this may vary by orders of magnitude. The saturation
current, I0 depends on recombination in the solar cell. Open-circuit voltage is then a measure of the amount of recombination in the
device. Silicon solar cells on high quality single crystalline material have open-circuit voltages of up to 764 mV under one sun and
AM1.5 conditions1, while commercial devices on multicrystalline silicon typically have open-circuit voltages around 600 mV.

The VOC can also be determined from the carrier concentration 2:

where kT/q is the thermal voltage, NA is the doping concentration, Δn is the excess carrier concentration and ni is the intrinsic carrier
concentration. The determination of VOC from the carrier concentration is also termed Implied VOC.

Open Circuit Voltage Calculator 2

Doping Concentration, NA = cm-3 Excess Carrier

Concentration, Δn = cm-3 Temperature, T =
K Intrinsic Carrier Concentration, ni =
cm-3
Open Circuit Voltage, VOC = V

Voc as a Function of Bandgap, EG

Where the short-circuit current (ISC) decreases with increasing bandgap, the open-circuit voltage increases as the band gap increases.
In an ideal device the VOC is limited by radiative recombination and the analysis uses the principle of detailed balance to determine the
minimum possible value for J0.

The minimum value of the diode saturation current is given by 3:

where q is the electronic charge, σ is the Stefan–Boltzmann constant, k is Boltzmann constant, T is the temperature and

Evaluating the integral in the above equation is quite complex. The graph below uses the method outlined in 4
Diode saturation current as a function of band gap. The values are determined from detailed balance and place a limit on the open
circuit voltage of a solar cell.

The J0 calculated above can be directly plugged into the standard solar cell equation given at the top of the page to determine the VOC
so long as the voltage is less than the band gap, as is the case under one sun illumination.

VOC as function of bandgap for a cell with AM 0 and AM 1.5. The VOC increases with bandgap as the recombination current falls.
There is drop off in VOC at very high band gaps due to the very low ISC.

1. A. Augusto, Herasimenka, S. Y., King, R. R., Bowden, S. G., and Honsberg, C., “Analysis of the recombination mechanisms
of a silicon solar cell with low bandgap-voltage offset”, Journal of Applied Physics, vol. 121, no. 20, p. 205704, 2017.
2. R. A. Sinton and Cuevas, A., “Contactless determination of current–voltage characteristics and minority-carrier lifetimes in
semiconductors from quasi-steady-state photoconductance data”, Applied Physics Letters, vol. 69, pp. 2510-2512, 1996.
3. P. Baruch, De Vos, A., Landsberg, P. T., and Parrott, J. E., “On some thermodynamic aspects of photovoltaic solar energy
conversion”, Solar Energy Materials and Solar Cells, vol. 36, pp. 201-222, 1995.
4. M. Y. Levy and Honsberg, C. B., “Rapid and precise calculations of energy and particle flux for detailed-balance photovoltaic
applications”, Solid-State Electronics, vol. 50, pp. 1400-1405, 2006.
Fill Factor
Add new comment
Español

The short-circuit current and the open-circuit voltage are the maximum current and voltage respectively from a solar cell. However, at
both of these operating points, the power from the solar cell is zero. The "fill factor", more commonly known by its abbreviation "FF",
is a parameter which, in conjunction with Voc and Isc, determines the maximum power from a solar cell. The FF is defined as the ratio
of the maximum power from the solar cell to the product of Voc and Isc so that:

$$FF= \frac{P_{MP}}{V_{OC}\times I_{SC}}$$

Fill Factor
Edit
Delete
Clone

Read more about Fill Factor

Graphically, the FF is a measure of the "squareness" of the solar cell and is also the area of the largest rectangle which will fit in the IV
curve. The FF is illustrated below.

Graph of cell output current (red line) and power (blue line) as a function of voltage. Also shown are the cell short-circuit current (Isc)
and open-circuit voltage (VOC) points, as well as the maximum power point (Vmp, Imp). Click on the graph to see how the curve
changes for a cell with low FF.

As FF is a measure of the "squareness" of the IV curve, a solar cell with a higher voltage has a larger possible FF since the "rounded"
portion of the IV curve takes up less area. The maximum theoretical FF from a solar cell can be determined by differentiating the
power from a solar cell with respect to voltage and finding where this is equal to zero. Hence:

giving:

$$V_{MP} = V_{OC} - \frac{nkT}{q}ln(\frac{q V_{MP}}{nkT}+1)$$

The equation above requires Lambert functions to solve (see below) but a simpler approach is to use iteration to calculate VMP. The
equation above only relates Voc to VMP and extra equations are needed to find IMP and FF. A more commonly used expression for the
FF can be determined empirically as:1

Fill Factor - Empirical

 Edit
Delete
Clone

Read more about Fill Factor - Empirical

where voc is defined as a "normalized Voc":

Voc normalized
Edit
Delete
Clone

Read more about Voc normalized

Fill Factor Calculator 1

Input Parameters

Open-circuit voltage,Voc (volts) Ideality Factor,n (units) Temperature, T (K)

Results

Normalized VOC, voc (units):X

Fill Factor, FF:X

The above equations show that a higher voltage will have a higher possible FF. However, large variations in open-circuit voltage within
a given material system are relatively uncommon. For example, at one sun, the difference between the maximum open-circuit voltage
measured for a silicon laboratory device and a typical commercial solar cell is about 120 mV, giving maximum FF's respectively of
0.85 and 0.83. However, the variation in maximum FF can be significant for solar cells made from different materials. For example, a
GaAs solar cell may have a FF approaching 0.89.

The above equation also demonstrates the importance of the ideality factor, also known as the "n-factor" of a solar cell. The ideality
factor is a measure of the junction quality and the type of recombination in a solar cell. For the simple recombination mechanisms
discussed in Types of Recombination, the n-factor has a value of 1. However, some recombination mechanisms, particularly if they are
large, may introduce recombination mechanisms of 2. A high n-value not only degrades the FF, but since it will also usually signal
high recombination, it gives low open-circuit voltages.

A key limitation in the equations described above is that they represent a maximum possible FF, although in practice the FF will be
lower due to the presence of parasitic resistive losses, which are discussed in Effects of Parasitic Resistances. Therefore, the FF is most
commonly determined from measurement of the IV curve and is defined as the maximum power divided by the product of Isc*Voc,
i.e.:

Fill Factor
Edit
Delete
Clone

Read more about Fill Factor

 Fill Factor Calculator 2

Input Parameters

Open-circuit voltage, Voc (volts): Short-circuit current, Isc (amps): Voltage at max
power, Vmp (volts): Current at max power, Imp (amps):

Results

Resulting fill factor, FF: X

Exact determination of VMP

The equation for a solar cell is:

$$I = I_L-I_0\left[\exp\left(\frac{V}{nV_t}\right)-1\right]$$

\(Power = V \times I\) and in addition the -1 term has no effect at VMP

$$P = V I_L- V I_0\exp\left(\frac{V}{nV_t}\right)$$

VMP is when the derivative of the power with respect to V is zero:

$$0 = I_L- I_0\exp\left(\frac{V_{MP}}{nV_t}\right)\left(1+\frac{V_{MP}}{nV_t}\right)$$

V> > Vt and rearranging gives:

$$\frac{I_L}{I_0} =\exp\left(\frac{V_{MP}}{nV_t}\right)\left(\frac{V_{MP}}{nV_t}\right)$$

The Lambert W function provides the solution to a class of exponential functions.

$$Y = Xe^x \Leftrightarrow X = W (Y)$$

so we get:

$$\frac{V_{MP}}{nV_t} =W\left(\frac{I_L}{I_0}\right)$$

$$V_{MP} = nV_t W\left(\frac{I_L}{I_0}\right)$$

using the expression for VOC we can also write:

$$V_{MP} = nV_t W\left(\exp\left(\frac{V_{OC}}{nV_t}\right)\right)$$

Since the expression inside the W() is always real and positive we only ever need the principle branch of the Lambert W function, W0.
The Lambert W function is a transcendental function much like logarithm function. While it is not available on most calculators, it is
available on advanced mathematical packages such as Matlab or Python. With further analysis the Lambert W function can also be
used for other solar cell terms and in the presence of parasitic resistances 2.

1. M. A. Green, “Solar cell fill factors: General graph and empirical expressions”, Solid-State Electronics, vol. 24, pp. 788 - 789,
1981.
2. A. Jain, “Exact analytical solutions of the parameters of real solar cells using Lambert W-function”, Solar Energy Materials
and Solar Cells, vol. 81, no. 2, pp. 269 - 277, 2004.

Solar Cell Efficiency

Add new comment
1 comment(s)
Español

The efficiency is the most commonly used parameter to compare the performance of one
solar cell to another. Efficiency is defined as the ratio of energy output from the solar
cell to input energy from the sun. In addition to reflecting the performance of the solar
cell itself, the efficiency depends on the spectrum and intensity of the incident sunlight
and the temperature of the solar cell. Therefore, conditions under which efficiency is
measured must be carefully controlled in order to compare the performance of one
device to another. Terrestrial solar cells are measured under AM1.5 conditions and at a
temperature of 25°C. Solar cells intended for space use are measured under AM0
conditions. Recent top efficiency solar cell results are given in the page Solar Cell
Efficiency Results.

The efficiency of a solar cell is determined as the fraction of incident power which is
converted to electricity and is defined as:

Where:
Voc is the open-circuit voltage;
Isc is the short-circuit current;
FF is the fill factor and
η is the efficiency.

The input power for efficiency calculations is 1 kW/m2 or 100 mW/cm2. Thus the input
power for a 100 × 100 mm2 cell is 10 W and for a 156 × 156 mm2 cell is 24.3 W Testing silicon solar cells.

Efficiency Calculator

Input Parameters

Open Circuit Voltage, VOC (V) Short Circuit Current, ISC (A) Fill Factor, FF
Input Power, Pin (W)

Results

Maximum Power, Pmax = W Efficiency, η = %

Detailed Balance
Add new comment
Español

Detailed balance provides a technique to calculate the maximum efficiency of photovoltaic devices. Originally the method was
proposed by Shockley and Queisser in 1961 1. An extended version was published in 1984 by Tiedje et al. 2.

Detailed balance in its simplest and most common implementation makes several fundamental assumptions:

1. The mobility is infinite, allowing collection of carriers no matter where they are generated.
2. Complete absorption of all photons above the band gap.

The calculations for detailed balance calculations involve calculating the particle flux for different configurations of the Plank’s
equation. The general form of the equation is:
The general approach is to calculate the absorption flux and the flux emitted from the solar cell. The difference between these two
(multiplied by q) is the current from the solar cell.

Absorption flux

The absorption consists of two parts; one from the sun and the other from the other regions of the sky. Under maximum concentration,
the optics makes it such that the entire surrounding of the solar cell is illuminated by radiation of the same temperature of the sun.
Under conditions other than maximum concentration, one portion of the sky (shown in yellow below) is illuminated from the sun, and
the remainder is illuminated from a radiation source with the same temperature as the Earth. The maximum concentration is calculated
based on the size of the sun’s disc in the sky and is given by 46,300.

Schematic representation of approximations used in detailed balance. On the left image light is collected from the entire hemisphere
which corresponds to maximum concentration. For lower or no concentration the Sun can be seen as an arc giving lower particle flux.

The absorption from the sun is given by:

And the absorption from the black body radiation of the earth is:

The total absorption of the solar cell is:

Emission flux

The emission from the solar cell depends on the quasi-Fermi level separation (μ) of the solar cell. Under short circuit conditions, m is
zero. The emission is calculated by:

Calculation of efficiency for a fixed Eg and black body

The power from the solar cell depends on the band gap and on the quasi-Fermi level separation. For a given band gap, the quasi-Fermi
level separation must be varied to find the maximum power point, i.e., where
is at a maximum. This is done by varying m from 0 to close to the open circuit condition (where φ1 = φ2), and find where the power is
at a maximum.

The efficiency is then defined as:

Efficiency as a function of band gap

To find the efficiency as a function of band gap, the above procedure is repeated for each band gap. There is an range of bandgaps for
the optimum cell efficiency as shown in the graph below.

Limiting solar cell efficiency as a function of the material bandgap for one-sun illumination. The calculations assume that the only
recombination is radiative. In actual devices the efficiencies are lower due to other recombination mechanisms and losses in parasitic
resistances.

AM1.5 Spectrum

To find the efficiency under an AM1.5 spectra (or other measured spectra), φ1 is replaced by the summation of the photon flux for
energies above the band gap, and the power from the sun is replaced by the summation of the power in the measured spectra.

1. W. Shockley and Queisser, H. J., “Detailed Balance Limit of Efficiency of p-n Junction Solar Cells”, Journal of Applied
Physics, vol. 32, pp. 510-519, 1961.
2. T. Tiedje, Yablonovich, E., Cody, G. D., and Brooks, B. G., “Limiting Efficiency of Silicon Solar Cells”, IEEE
TRANSACTIONS ON ELECTRON DEVICES, vol. ED-31, 1984.

Resistive Effects
Add new comment
Español

Tandem Cells
 Add new comment
Español

One method to increase the efficiency of a solar cell is to split the spectrum and use a solar cell that is optimised to each section of the
spectrum.

Series connected tandem solar cell. Adding more devices allows for each device to be optimized to a narrower spectrum giving a
higher overall efficiency.

Tandem solar cells can either be individual cells or connected in series. Series connected cells are simpler to fabricate but the current is
the same though each cell so this contrains the band gaps that can be used. The most common arrangement for tandem cells is to grow
them monolithically so that all the cells are grown as layers on the on substrate and tunnel junctions connect the individual cells.

The maximum efficiency for a two junction tandem under the AM1.5G spectrum and without concentration is 47 %. At the peak
efficiency the top cell has a bandgap of 1.63 eV and the bottom cell has a bandgap of 0.96 eV.

As the number of bandgaps increases the efficiency of the stack also potentially increases. In reality, the semiconductor materials do
not exist to allow for arbitrary materials with a specific bandgap and of high quality.
Efficiency of a an ideal stack of solar cells as a function of the number of bandgaps 1. The spectrum used here is the direct spectrum
(AM 1.5D) where the diffuse radiation is excluded.

1. S. P. Bremner, Levy, M. Y., and Honsberg, C. B., “Analysis of tandem solar cell efficiencies under {AM1.5G} spectrum using
a rapid flux calculation method”, Progress in Photovoltaics: Research and Applications, vol. 16, pp. 225–233, 2008.

Characteristic Resistance
Add new comment
Español

The characteristic resistance of a solar cell is the output resistance of the solar cell at its maximum power point. If the resistance of the
load is equal to the characteristic resistance of the solar cell, then the maximum power is transferred to the load and the solar cell
operates at its maximum power point. It is a useful parameter in solar cell analysis, particularly when examining the impact of parasitic
loss mechanisms. The characteristic resistance is shown in the figure below.

The characteristic resistance of a solar cell is the inverse of the slope of the line, shown in the figure above, which, after Green 1, can
be given as VMP divided by IMP for most cells it can be approximated by VOC divided by ISC:

Characteristic Resistance
Edit
Delete
Clone
 Read more about Characteristic Resistance

Commercial silicon solar cells are typically very high current and low voltage. A 156 mm (6 inch) square solar cell has a current of
almost 9 amps and a maximum power point voltage of 0.6 volts giving a characteristic resistance of around 0.067 Ω. As a
consequence, connections to a single cell require resistances in the milliohm range.

1. M. A. Green, “Accuracy of Analytical Expressions for Solar Cell Fill Factors”, Solar Cells, vol. 7, pp. 337-340, 1982.

Effect of parasitic Resistances

Add new comment
Español

Resistive effects in solar cells reduce the efficiency of the solar cell by dissipating power in the resistances. The most common parasitic
resistances are series resistance and shunt resistance. The inclusion of the series and shunt resistance on the solar cell model is shown
in the figure below.

Parasitic series and shunt resistances in a solar cell circuit.

In most cases and for typical values of shunt and series resistance, the key impact of parasitic resistance is to reduce the fill factor. Both
the magnitude and impact of series and shunt resistance depend on the geometry of the solar cell, at the operating point of the solar
cell. Since the value of resistance will depend on the area of the solar cell, when comparing the series resistance of solar cells which
may have different areas, a common unit for resistance is in Ωcm2. This area-normalized resistance results from replacing current with
current density in Ohm's law as shown below:

The effects of series and shunt resistances are described in the following pages.

Series Resistance
Add new comment
1 comment(s)
Español

Series resistance in a solar cell has three causes: firstly, the movement of current through the emitter and base of the solar cell;
secondly, the contact resistance between the metal contact and the silicon; and finally the resistance of the top and rear metal contacts.
The main impact of series resistance is to reduce the fill factor, although excessively high values may also reduce the short-circuit
current.
Schematic of a solar cell with series resistance.

where: I is the cell output current, IL is the light generated current, V is the voltage across the cell terminals, T is the temperature, q and
k are constants, n is the ideality factor, and RS is the cell series resistance. The formula is an example of an implicit function due to the
appearance of the current, I, on both sides of the equation and requires numerical methods to solve.

The effect of the series resistance on the IV curve is shown below. To generate the plot the voltage across the diode is varied thereby
avoiding the need to solve an implicit equation.

Cell series resistance, RS = 1 Ω cm2

0 20

The effect of series resistance on fill factor. The area of the solar cell is 1 cm2 so that the units of resistance can be either ohm or ohm
cm2. The short circuit current (ISC) is unaffected b the series resistance until it is very large.

Series resistance does not affect the solar cell at open-circuit voltage since the overall current flow through the solar cell, and therefore
through the series resistance is zero. However, near the open-circuit voltage, the IV curve is strongly affected by the series resistance.
A straight-forward method of estimating the series resistance from a solar cell is to find the slope of the IV curve at the open-circuit
voltage point.

An equation for the FF as a function of series resistance can be determined by noting that for moderate values of series resistance, the
maximum power may be approximated as the power in the absence of series resistance minus the power lost in the series resistance.
The equation for the maximum power from a solar cell then becomes:
defining a normalized series resistance as;

gives the following equation which approximates the effect of series resistance on the output power of a solar cell;

Assuming that the open-circuit voltage and short-circuit current are not affected by the series resistance allows the impact of series
resistance on FF to be determined;

In the above equation the fill factor which is not affected by series resistance is denoted by FF0 and FF' is called FFS. The equation
then becomes;

An empirical equation, which is slightly more accurate for the relationship between FF0 and FFS is;

which is valid for rs < 0.4 and voc > 10.

The following calculator determines the effect of Rs on the solar cell fill factor. Typical values for area-normalized series resistance are
between 0.5 Ωcm2 for laboratory type solar cells and up to 1.3 0.5 Ωcm2 for commercial solar cells. The current levels in the solar cell
have a major impact on the losses due to series resistance and in the following calculator, examine the impact raising the current has on
the FF.

Series Resistance Calculator

Input Parameters

Open circuit Voltage, VOC (volts) Short circuit current, ISC (amps) Ideality factor,
n (units) Temperature, T (K) Ideal FF with no R, FF0
Rs (ohms)

Results

Cell characteristic resistance, RCH 18.5714 (ohms)

Normalized Voc, voc 25.1208 (units)
Normalized series resistance, rs 0.0538 (units)
Approximate fill factor, FF, with Rs FFapprox 0.7922
More accurate FF valid for rs < 0.4 and voc > 10 FFs 0.7882
Shunt Resistance
Add new comment
1 comment(s)
Español

Significant power losses caused by the presence of a shunt resistance, RSH, are typically due to manufacturing defects, rather than poor
solar cell design. Low shunt resistance causes power losses in solar cells by providing an alternate current path for the light-generated
current. Such a diversion reduces the amount of current flowing through the solar cell junction and reduces the voltage from the solar
cell. The effect of a shunt resistance is particularly severe at low light levels, since there will be less light-generated current. The loss
of this current to the shunt therefore has a larger impact. In addition, at lower voltages where the effective resistance of the solar cell is
high, the impact of a resistance in parallel is large.

Circuit diagram of a solar cell including the shunt resistance.

The equation for a solar cell in presence of a shunt resistance is:

where: I is the cell output current, IL is the light generated current, V is the voltage across the cell terminals, T is the temperature, q and
k are constants, n is the ideality factor, and RSH is the cell shunt resistance.

The effect of a low shunt resistance is shown in the animation below.

Shunt resistance is: 1.00e+6 Ω cm2

The effect of shunt resistance on fill factor in a solar cell. The area of the solar cell is 1 cm2, the cell series resistance is zero,
temperature is 300 K, and I0 is 1 x 10-12 A/cm2.

An estimate for the value of the shunt resistance of a solar cell can be determined from the slope of the IV curve near the short-circuit
current point.

The impact of the shunt resistance on the fill factor can be calculated in a manner similar to that used to find the impact of series
resistance on fill factor. The maximum power may be approximated as the power in the absence of shunt resistance, minus the power
lost in the shunt resistance. The equation for the maximum power from a solar cell then becomes;

Defining a normalized shunt resistance as;

Assuming that the open-circuit voltage and short-circuit current are not affected by the shunt resistance allows the impact of shunt
resistance on FF to be determined as;

In the above equation FF, the fill factor which is not affected by shunt resistance is denoted by FF0 and FF' is called FFSH. The
equation then becomes;

An empirical equation, which is slightly more accurate for the relationship between FF0 and FFSH is;

which is valid for rsh > 0.4.

The following calculator determines the effect of Rsh on the solar cell fill factor. Typical values for area-normalized shunt resistance
are in the MΩcm2 range for laboratory type solar cells, and 1000 Ωcm2 for commercial solar cells.
 Shunt Resistance Calculator

VOC = V ISC =
An= T=
K FF0 =
RSH =
Ohms

RCH = Ohms voc = rsh =

Ohms Approximate FF valid for rsh > 1.0
Rsh
FFapprox = More accurate estimation of FF
valid for rsh > 0.4
FFsh =

Impact of Both Series and Shunt Resistance

Add new comment

In the presence of both series and shunt resistances, the IV curve of the solar cell is given by;

and the circuit diagram of the solar cell is given as;

Parasitic series and shunt resistances in a solar cell circuit.

To combine the effect of both series and shunt resistances, the expression for FFsh, derived above, can be used, with FF0 replaced by
FFs 1. The overall equation then becomes;

where FFs is given by;

and by combining the above equations, the net equation for FF becomes;
 Series Resistance, Rseries = 1 Ω cm2

0.1 3

Shunt Resistance, Rshunt = 3 volts

10 1e5

The following calculator determines the effect of Rs and Rsh on the cell fill factor using the same variable definitions as above.
 Effect of Both Resistances Calculator

Input Parameters

VOC = V
ISC = A
n=
T= K
FF0 =

RS = Ohms
RSH = Ohms

Results

RCH = Ohms
voc =
rs = Ohms
rsh = Ohms
Approximate fill factor taking into account Rs and Rsh
FFapprox =
A more accurate estimation of FF
valid for rs < 0.4 and voc > 10
FFs =
Estimation of FF from Rshuntvalid for rsh > 0.4
FFsh =
More accurate estimation of FF taking into account Rs & Rsh
FF =

1. M. A. Green, Solar Cells - Operating Principles, Technology and System Application. Kensington, Australia: University of
NSW, 1992.

Other Effects
Add new comment
Español

Effect of Temperature
Add new comment
3 comment(s)
Español

Like all other semiconductor devices, solar cells are sensitive to temperature. Increases in temperature reduce the band gap of a
semiconductor, thereby effecting most of the semiconductor material parameters. The decrease in the band gap of a semiconductor
with increasing temperature can be viewed as increasing the energy of the electrons in the material. Lower energy is therefore needed
to break the bond. In the bond model of a semiconductor band gap, reduction in the bond energy also reduces the band gap. Therefore
increasing the temperature reduces the band gap.

In a solar cell, the parameter most affected by an increase in temperature is the open-circuit voltage. The impact of increasing
temperature is shown in the figure below.
The effect of temperature on the IV characteristics of a solar cell.

The open-circuit voltage decreases with temperature because of the temperature dependence of I0. The equation for I0 from one side of
a p-n junction is given by;

where:
q is the electronic charge given in the constants page;
A is the area;
D is the diffusivity of the minority carrier given for silicon as a function of doping in the Silicon Material Parameters page;
L is the minority carrier diffusion length;
ND is the doping; and
ni is the intrinsic carrier concentration given for silicon in the Silicon Material Parameters page.

In the above equation, many of the parameters have some temperature dependence, but the most significant effect is due to the intrinsic
carrier concentration, ni. The intrinsic carrier concentration depends on the band gap energy (with lower band gaps giving a higher
intrinsic carrier concentration), and on the energy which the carriers have (with higher temperatures giving higher intrinsic carrier
concentrations). The equation for the intrinsic carrier concentration is;

where:
T is the temperature;
h and k are constants given in the constants page;
me and mh are the effective masses of electrons and holes respectively;
EGO is the band gap linearly extrapolated to absolute zero; and
B is a constant which is essentially independent of temperature.

Substituting these equations back into the expression for I0, and assuming that the temperature dependencies of the other parameters
can be neglected, gives;

where B' is a temperature independent constant. A constant, γ, is used instead of the number 3 to incorporate the possible temperature
dependencies of the other material parameters. For silicon solar cells near room temperature, I0 approximately doubles for every 10 °C
increase in temperature.

The impact of I0 on the open-circuit voltage can be calculated by substituting the equation for I0 into the equation for Voc as shown
below;
where EG0 = qVG0. Assuming that dVoc/dT does not depend on dIsc/dT, dVoc/dT can be found as;

The above equation shows that the temperature sensitivity of a solar cell depends on the open circuit voltage of the solar cell, with
higher voltage solar cells being less affected by temperature. For silicon, EG0 is 1.2, and using γ as 3 gives a reduction in the open-
circuit voltage of about 2.2 mV/°C;

The short-circuit current, Isc, increases slightly with temperature, since the band gap energy, EG, decreases and more photons have
enough energy to create e-h pairs. However, this is a small effect and the temperature dependence of the short-circuit current from a
silicon solar cell is;

or 0.06% per °C for silicon.

The temperature dependency FF for silicon is approximated by the following equation;

The effect of temperature on the maximum power output, Pm, is;

or 0.4% to 0.5% per °C for silicon.

300 K or 25 °C ?

Most semiconductor modeling is done at 300 K since it is close to room temperature and a convenient number. However, solar cells are
typically measured almost 2 degrees lower at 25 °C (298.15 K). In most cases the difference is insignificant (only 4 mV of Voc) and
both are referred to as room temperature. Occasionally, the modeled results need to be adjusted to correlate with the measured results.

Temperature - Intrinsic Carrier Concentration Calculator

Temperature, T (K)
Intrinsic Carrier Concentration, ni: 8.59e9 cm-3
At 300 K, ni = 1.01 x 1010 cm-3 and kT/q = 25.852 mV
At 25 °C (298.15 K), ni = 8.6 x 109 cm-3 and kT/q = 25.693 mV

Measuring solar cell efficiency in Antarctica. Solar cells love cold sunny environments. (Photo Antony Schinckel)

Effect of Light Intensity

Add new comment
3 comment(s)

Changing the light intensity incident on a solar cell changes all solar cell parameters, including the short-circuit current, the open-
circuit voltage, the FF, the efficiency and the impact of series and shunt resistances. The light intensity on a solar cell is called the
number of suns, where 1 sun corresponds to standard illumination at AM1.5, or 1 kW/m2. For example a system with 10 kW/m2
incident on the solar cell would be operating at 10 suns, or at 10X. A PV module designed to operate under 1 sun conditions is called a
"flat plate" module while those using concentrated sunlight are called "concentrators".

Ideal Cell: Voc = 0.623 Isc = 35 mA/cm2 FF = 0.83

Real Cell: Voc = 0.612 Isc = 34.6 mA/cm2 FF = 0.67

 Shunt Resistance, Rshunt = 1000.00 ohms

10 1e6

Cell series resistance, Rs = 1 cm/2

0.1 18

Concentration of light on solar cell, conc = 1 suns

0 2

The effect of concentration on the IV characteristics of a solar cell. The series resistance has a greater effect on performance at high
intensity and the shunt resistance has a greater effect on cell performance at low light intensity.

Concentrators

A concentrator is a solar cell designed to operate under illumination greater than 1 sun. The incident sunlight is focused or guided by
optical elements such that a high intensity light beam shines on a small solar cell. Concentrators have several potential advantages,
including a higher efficiency potential than a one-sun solar cell and the possibility of lower cost. The short-circuit current from a solar
cell depends linearly on light intensity, such that a device operating under 10 suns would have 10 times the short-circuit current as the
same device under one sun operation. However, this effect does not provide an efficiency increase, since the incident power also
increases linearly with concentration. Instead, the efficiency benefits arise from the logarithmic dependence of the open-circuit voltage
on short circuit. Therefore, under concentration, Voc increases logarithmically with light intensity, as shown in the equation below;

where X is the concentration of sunlight.

From the equation above, a doubling of the light intensity (X=2) causes a 18 mV rise in VOC .

The cost of a concentrating PV system may be lower than a corresponding flat-plate PV system since only a small area of solar cells is
needed.

The efficiency benefits of concentration may be reduced by increased losses in series resistance as the short-circuit current increases
and also by the increased temperature operation of the solar cell. As losses due to short-circuit current depend on the square of the
current, power loss due to series resistance increases as the square of the concentration.

Low Light Intensity

Solar cells experience daily variations in light intensity, with the incident power from the sun varying between 0 and 1 kW/m2. At low
light levels, the effect of the shunt resistance becomes increasingly important. As the light intensity decreases, the bias point and
current through the solar cell also decreases, and the equivalent resistance of the solar cell may begin to approach the shunt resistance.
When these two resistances are similar, the fraction of the total current flowing through the shunt resistance increases, thereby
increasing the fractional power loss due to shunt resistance. Consequently, under cloudy conditions, a solar cell with a high shunt
resistance retains a greater fraction of its original power than a solar cell with a low shunt resistance1.

1. G. Bunea, Wilson, K., Meydbray, Y., Campbell, M., and Ceuster, D. D., “Low Light Performance of Mono-Crystalline Silicon
Solar Cells”, in 4th World Conference on Photovoltaic Energy Conference, Waikoloa, HI, 2006, pp. 1312–1314.

Ideality Factor
Add new comment
1 comment(s)
Español

The ideality factor of a diode is a measure of how closely the diode follows the ideal diode equation. The derivation of the simple
diode equation uses certain assumption about the cell. In practice, there are second order effects so that the diode does not follow the
simple diode equation and the ideality factor provides a way of describing them.
Recombination mechanisms

The ideal diode equation assumes that all the recombination occurs via band to band or recombination via traps in the bulk areas from
the device (i.e. not in the junction). Using that assumption the derivation produces the ideal diode equation below and the ideality
factor, n, is equal to one.

However recombination does occur in other ways and in other areas of the device. These recombinations produce ideality factors that
deviate from the ideal. Deriving the ideal diode equation by considering the number of carriers the need to come together during the
process produces the results in the table below.

Ideality
Recombination Type Description
factor
SRH, band to band (low level
1 Recombination limited by minority carrier.
injection)
SRH, band to band (high level
2 Recombination limited by both carrier types.
injection)

Two majority and one minority carriers required

Auger 2/3
for recombination.

Depletion region (junction) 2 two carriers limit recombination.

Chapter 4 Quiz
Questions 6
Attempts allowed Unlimited
Available Always
Pass rate 75 %
Backwards navigation Allowed
Start Quiz

Design of Silicon Cells

Add new comment
Español

Solar Cell Design Principles

Add new comment
Español

Solar cell design involves specifying the parameters of a solar cell structure in order to maximize efficiency, given a certain set of
constraints. These constraints will be defined by the working environment in which solar cells are produced. For example in a
commercial environment where the objective is to produce a competitively priced solar cell, the cost of fabricating a particular solar
cell structure must be taken into consideration. However, in a research environment where the objective is to produce a highly efficient
laboratory-type cell, maximizing efficiency rather than cost, is the main consideration.
Evolution of silicon solar cell efficiency.

The theoretical efficiency for photovoltaic conversion is in excess of 86.8% 1. However, the 86.8% figure uses detailed balance
calculations and does not describe device implementation. For silicon solar cells, a more realistic efficiency under one sun operation is
about 29% 2. The maximum efficiency measured for a silicon solar cell is currently 24.7% under AM1.5G. The difference between the
high theoretical efficiencies and the efficiencies measured from terrestrial solar cells is due mainly to two factors. The first is that the
theoretical maximum efficiency predictions assume that energy from each photon is optimally used, that there are no unabsorbed
photons and that each photon is absorbed in a material which has a band gap equal to the photon energy. This is achieved in theory by
modeling an infinite stack of solar cells of different band gap materials, each absorbing only the photons which correspond exactly to
its band gap.

The second factor is that the high theoretical efficiency predictions assume a high concentration ratio. Assuming that temperature and
resistive effects do not dominate in a concentrator solar cell, increasing the light intensity proportionally increases the short-circuit
current. Since the open-circuit voltage (Voc) also depends on the short-circuit current, Voc increases logarithmically with light level.
Furthermore, since the maximum fill factor (FF) increases with Voc, the maximum possible FF also increases with concentration. The
extra Voc and FF increases with concentration which allows concentrators to achieve higher efficiencies.

In designing such single junction solar cells, the principles for maximizing cell efficiency are:

increasing the amount of light collected by the cell that is turned into carriers;
increasing the collection of light-generated carriers by the p-n junction;
minimising the forward bias dark current;
extracting the current from the cell without resistive losses.

1. C. B. Honsberg, Corkish, R., and Bremner, S. P., “A New Generalized Detailed Balance Formulation to Calculate Solar Cell
Efficiency Limits”, 17th European Photovoltaic Solar Energy Conference. pp. 22-26, 2001.
2. R. M. Swanson, “Approaching the 29% limit efficiency of silicon solar cells”, Thirty-First IEEE Photovoltaic Specialists
Conference. 01/2005, Lake buena Vista, FL, USA, pp. 889-94, 2005.

Optical Losses
Add new comment
1 comment(s)
Español

Optical losses chiefly effect the power from a solar cell by lowering the short-circuit current. Optical losses consist of light which
could have generated an electron-hole pair, but does not, because the light is reflected from the front surface, or because it is not
absorbed in the solar cell. For the most common semiconductor solar cells, the entire visible spectrum (350 - 780 nm) has enough
energy to create electron-hole pairs and therefore all visible light would ideally be absorbed.
Sources of optical loss in a solar cell.

There are a number of ways to reduce the optical losses:

Top contact coverage of the cell surface can be minimised (although this may result in increased series resistance). This is
discussed in more detail in Series Resistance;
Anti-reflection coatings can be used on the top surface of the cell.
Reflection can be reduced by surface texturing.
The solar cell can be made thicker to increase absorption (although light that is absorbed more than a diffusion length from the
junction has a low collection probability and will not contribute to the short circuit current).
The optical path length in the solar cell may be increased by a combination of surface texturing and light trapping.

The reflection of a silicon surface is over 30% due to its high refractive index. The reflectivity, R, between two materials of different
refractive indices is determined by:

Reflectivity
Edit
Delete
Clone

Read more about Reflectivity

where n0 is the refractive index of the surroundings and nSi is the complex refractive index of silicon. For an unencapsulated cell n0 =
1. For an encapsulated cell n0 = 1.5. The refractive index of silicon changes with wavelength and is given in the chapter on material
properties.

Anti-Reflection Coatings
Add new comment
5 comment(s)
Español

Bare silicon has a high surface reflection of over 30%. The reflection is reduced by texturing and and by applying anti-reflection
coatings (ARC) to the surface1. Anti-reflection coatings on solar cells are similar to those used on other optical equipment such as
camera lenses. They consist of a thin layer of dielectric material, with a specially chosen thickness so that interference effects in the
coating cause the wave reflected from the anti-reflection coating top surface to be out of phase with the wave reflected from the
semiconductor surfaces. These out-of-phase reflected waves destructively interfere with one another, resulting in zero net reflected
energy. In addition to anti-reflection coatings, interference effects are also commonly encountered when a thin layer of oil on water
produces rainbow-like bands of color.
Use of a quarter wavelength anti-reflection coating to counter surface reflection.

The thickness of the anti-reflection coating is chosen so that the wavelength in the dielectric material is one quarter the wavelength of
the incoming wave. For a quarter wavelength anti-reflection coating of a transparent material with a refractive index n1 and light
incident on the coating with a free-space wavelength λ0, the thickness d1 which causes minimum reflection is calculated by:

ARC Thickness
Edit
Delete
Clone

Read more about ARC Thickness

ARC Thickness
Edit
Delete
Clone

ARC Thickness Calculator

Wavelength, λ (µm) = Refractive index of anti-reflection coating, n1:

Optimal ARC thickness, d1: 0.75 μm

Reflection is further minimized if the refractive index of the anti-reflection coating is the geometric mean of that of the materials on
either side; that is, glass or air and the semiconductor. This is expressed by:

ARC Refractive Index

Edit
Delete
Clone
 Read more about ARC Refractive Index

ARC Refractive Index

Edit
Delete
Clone

ARC Refractive Index Calculator

Refractive index of surrounding material, n0: Refractive index of semiconductor n2

Optimal refractive index of ARC, n1: 2.2913

For the reflectance at normal incidence we define a series of parameters: r1, r2, and θ. The surrounding region has a refractive index of
n0, the ARC has a refractive index of n1 and a thickness of t1, and the silicon has a refractive index of n2.

ARC Reflectance
Edit
Delete
Clone

Read more about ARC Reflectance

For a single layer ARC on a substrate the reflectivity is:

ARC Substrate Reflectivity

Edit
Delete
Clone

Read more about ARC Substrate Reflectivity

 Refractive index, n = 2 units

1 3

Cell thickness, t = 1.00e+2 µm. In silicon JSC =0.0366 A/cm²

10 200

The graph shows the effect of a single layer anti-reflection coating on silicon. Use the sliders to adjust the refractive index and
thickness of the layer. For simplicity this simulation assumes a constant refractive index for silicon at 3.5. In reality the refractive index
of silicon and the coating is a function of wavelength.

While the reflection for a given thickness, index of refraction, and wavelength can be reduced to zero using the equations above, the
index of refraction is dependent on wavelength and so zero reflection occurs only at a single wavelength. For photovoltaic applications,
the refractive index, and thickness are chosen in order to minimize reflection for a wavelength of 0.6 µm. This wavelength is chosen
since it is close to the peak power of the solar spectrum.

Comparison of surface reflection from a silicon solar cell, with and without a typical anti-reflection coating.
 1. G. Bauer, “Absolutwerte der optischen Absorptionskonstanten von Alkalihalogenidkristallen im Gebiet ihrer ultravioletten
Eigenfrequenzen”, Annalen der Physik, vol. 411, no. 4, pp. 434 - 464, 1934.

Anti-Reflection Coating Color

Add new comment
Español

The most common method to precisely measure the thickness of anti-reflection coatings is using ellipsometry which looks at the way
polarized light is reflected.

A simple method to measure anti-reflection coating is by simply looking at the film colour.1

Four multicrystalline wafers covered with films of silicon nitride. The difference in color is solely due to the thickness of the film. The
green wafers are very thick films and so don't appear in the color chart of the next figure.

The color of the film is affected by the thickness as well as the refractive index so the film color is merely a rough guide to thickness.
There are multiple tables for determining the films thickness for films of silicon dioxide or silicon nitride such as those shown below. It
is common to fabricate film standards so that variations in process conditions can be quickly detected by comparing samples to the
standard.

Color of silicon nitride films with a refractive index of ~2.05 as a function of film thickness under fluorescent lighting for normal
incident light. The chart is merely a guide as the color of films on films in commercial production will be affected by texturing and
changes in the refractive index. (Plot generated using applet at BYU)

Color Chart for Films of Silicon Dioxide (SiO2) under fluorescent lighting2

Film
Colour and Comments
Thickness (µm)
0.05 Tan
0.07 Brown

0.10 Dark violet to red violet

0.12 Royal Blue
0.15 Light blue to metallic blue

0.17 Metallic to very light yellow green

0.20 Light gold to yellow; slightly metallic
0.22 Gold with slight yellow orange

0.25 Orange to melon

0.27 Red violet
0.30 Blue to violet blue

0.31 Blue

0.32 Blue to blue green

0.34 Light green

0.35 Green to yellow green

0.36 Yellow green

0.37 Green yellow

0.39 Yellow
0.41 Light orange

0.42 Carnation pink

0.44 Violet red

0.46 Red violet

0.47 Violet
0.48 Blue violet

0.49 Blue

0.50 Blue green

0.52 Green (broad)

0.54 Yellow green

0.56 Green yellow

"Yellowish"
0.57
(actually at times appears to be light creamy gray or metallic)
0.58 Light orange or yellow to pink borderline

0.60 Carnation pink

0.63 Violet red

0.68 "Bluish" (actually appears more like a mixture between violet red and blue green and looks grayish)

0.72 Blue green to green (quite broad)

0.77 "Yellowish"
 0.80 Orange (rather broad for orange)
0.82 Salmon

0.85 Dull, light red violet

0.86 Violet
0.87 Blue violet

0.89 Blue
0.92 Blue green
0.95 Dull yellow green

0.97 Yellow to "yellowish"

0.99 Orange

1.00 Carnation pink

1.02 Violet red

1.05 Red violet

1.06 Violet

1.07 Blue violet

1.10 Green

1.11 Yellow green

1.12 Green
1.18 Violet

1.19 Red Violet

1.21 Violet red

1.24 Carnation pink to salmon

1.25 Orange
1.28 "Yellowish"

1.32 Sky blue to green blue

1.40 Orange

1.45 Violet

1.46 Blue violet

1.50 Blue

1.54 Dull yellow green

1. J. Henrie, Kellis, S., Schultz, S., and Hawkins, A., “Electronic color charts for dielectric films on silicon”, Optics Express, vol.
12, pp. 1464–1469, 2004.
2. W. A. Pliskin and Conrad, E. E., “Nondestructive determination of thickness and refractive index of transparent films”, IBM
Journal of Research Devices, vol. 8, pp. 43–51, 1964.

Double Layer Anti Reflection Coatings

Add new comment
 1 comment(s)
Español

A further reduction in reflectivity is achieved through a double layer anti-reflection coating (DLARC). Popular DLARC coatings are
zinc sulfide (ZnS) with magnesium flouride (MgF) or layers of silicon nitride with varying refractive index. However, this is usually
too expensive for most commercial solar cells.

Double layer anti-reflection film on silicon wafer. The layers are usually deposited on a textured substrate to decrease the reflectivity
further.

The equations for multiple anti-reflection coatings are more complicated than that for a single layer 1. First we define a series of
parameters: r1, r2, r3, θ1 and θ2. As in the diagram above, the surrounding region has a refractive index of n0, the next layer has a
refractive index of n1 and a thickness of t1, the layer immediately above the silicon has a refractive index of n2 and a thickness of t2
and the silicon has a refractive index of n3.

The reflectivity is then calculated from the above parameters using the following formula:

The graph below simulates a double layer antireflection coating. By adjusting the refractive index and thickness of the two layers it is
possible to produce two minima and a overall reflectance of less than 3%.
 Refractive index of layer 1, n1= 1.3 units

1 3

Thickness of layer 1, t1= 100 nm

10 300

Refractive index of layer 2, n2= 2.2 unit

1 3

Thickness of layer 2, t2= 200 nm

10 300

Interactive graph showing the effect of thickness and refractive index of a double layer anti-reflection coating (DLARC). The substrate
has a constant refractive index of 3.5 and approximates silicon. Under air, the reflection is minimized by setting the top layer to 105 nm
with a refractive index of 1.4 and setting the bottom layer to 55 nm and a reflective index of 2.4.

Further reductions in reflectivity can be achieved with stacks of coatings or by grading the index of the layers. In the optical industry
stacks of over 10 layers are used for very low reflection losses in photographic equipment. However, the performance benefits of
multiple layers are marginal. In most cases the real challenge is to incorporate the anti-reflection coatings in with the surface
passivation layers.

OPAL22 at pvlighthouse.com.au provides a simulator for multilayer stacks. It uses concepts similar to what is presented on these pages
but has much more complete mathematical models including the effect of refractive index as a function of wavelength.

1. E. Y. Wang, Yu, F. T. S., Sims, V. L., Brandhorst, E. W., and Broder, J. D., “Optimum Design of Anti-reflection coating for
silicon solar cells”, 10th IEEE Photovoltaic Specialists Conference. pp. 168-171, 1973.
2. K. R. McIntosh and Baker-Finch, S. C., “OPAL 2: Rapid optical simulation of silicon solar cells”, in 2012 IEEE 38th
Photovoltaic Specialists Conference (PVSC)2012 38th IEEE Photovoltaic Specialists Conference, Austin, TX, USA, 2012.

Surface Texturing
Add new comment
Español

Surface texturing, either in combination with an anti-reflection coating or by itself, can also be used to minimize reflection. Any
"roughening" of the surface reduces reflection by increasing the chances of reflected light bouncing back onto the surface, rather than
out to the surrounding air.1

Surface texturing can be accomplished in a number of ways. A single crystalline substrate can betextured by etching along the faces of
the crystal planes. The crystalline structure of silicon results in a surface made up of pyramids if the surface is appropriately aligned
with respect to the internal atoms. One such pyramid is illustrated in the drawing below. An electron microscope photograph of a
textured silicon surface is shown in the photograph below. This type of texturing is called "random pyramid" texture2, and is
commonly used in industry for single crystalline wafers.

A square based pyramid which forms the surface of an appropriately textured crystalline silicon solar cell.

Scanning electron microscope photograph of a textured silicon surface. Image Courtesy of The School of Photovoltaic & Renewable
Energy Engineering, University of New South Wales.

Another type of surface texturing used is known as "inverted pyramid" texturing3,4. Using this texturing scheme, the pyramids are
etched down into the silicon surface rather than etched pointing upwards from the surface. A photograph of such a textured surface is
shown below.
Scanning electron microscope photograph of a textured silicon surface. Image Courtesy of The School of Photovoltaic & Renewable
Energy Engineering, University of New South Wales.

For multicrystalline wafers, only a small fraction of the surface will have the required orientation of <100> and consequently these
techniques are less effective on multicrystalline wafers. However, multicrystalline wafers can be textured using a photolithographic
technique5 as well as mechanically sculpting the front surface using dicing saws6 or lasers7 to cut the surface into an appropriate
shape. A micrograph of a photolithographic texturing scheme is shown below.

Scanning electron microscope photograph of a textured multicrystalline silicon surface. Image Courtesy of The School of Photovoltaic
& Renewable Energy Engineering, University of New South Wales.

The modeling of textured substrates is covered by simulation programs at pvlighthouse.com.au and associated references8

1. B. Dale and Rudenberg, H. G., “High efficiency silicon solar cells”, in Proceedings of the 14th Annual Power Sources
Conference, 1960, p. 22.
2. W. L. Bailey, Coleman, M. G., Harris, C. B., and Lesk, I. A., “United States Patent: 4137123 - Texture etching of silicon:
method”. 1979.
3. P. Campbell and Green, M. A., “Light trapping properties of pyramidally textured surfaces”, Journal of Applied Physics, vol.
62, no. 1, p. 243, 1987.
4. P. Campbell and Green, M. A., “High performance light trapping textures for monocrystalline silicon solar cells”, Solar
Energy Materials and Solar Cells, vol. 65, no. 1-4, pp. 369 - 375, 2001.
5. J. Zhao, A., W., Dai, X., Green, M. A., and Wenham, S. R., “Improvements in Silicon Solar Cell Performance”, 22nd IEEE PV
Specialists Conference. pp. 399-402, 1991.
6. S. R. Wenham and Green, M. A., “Buried contact solar cell”. 1988.
 7. J. C. Zolper, Narayanan, S., Wenham, S. R., and Green, M. A., “16.7% efficient, laser textured, buried contact polycrystalline
silicon solar cell”, Applied Physics Letters, vol. 55, p. 2363, 1989.
8. S. C. Baker-Finch, McIntosh, K. R., and Terry, M. L., “Isotextured Silicon Solar Cell Analysis and Modeling 1: Optics”, IEEE
Journal of Photovoltaics, vol. 2, no. 4, pp. 457 - 464, 2012.

Material Thickness
Add new comment
Español

While the reduction of reflection is an essential part of achieving a high efficiency solar cell, it is also essential to absorb all the light in
the silicon solar cell. The amount of light absorbed depends on the optical path length and the absorption coefficient. The animation
below shows the dependence of photon absorption on device thickness for a silicon solar cell.

The device simulated is a cell with no front surface reflection losses so that all incident light enters the cell. The electronic properties
are assumed to be perfect (infinite diffusion length) so that all light generated carriers are collected. Finally, the light only passes once
through the cell. In reality, thin cells are usually designed with a reflector on the rear so that light makes multiple passes across the cell
and the absorption is increased. In the case of ideal lambertian light trapping the path length is effectively increased by 4n2 1. For
silicon with a refractive index of 3.5, light trapping increases the path length by a factor of ~50. See light trapping page.

thickness =

-1 4

For silicon material in excess of 10 mm thick, essentially all the light with energy above the band gap is absorbed. The 100% of the
total current refers to the fact that at 10 mm, all the light which can be absorbed in silicon, is absorbed. In material of 10 µm thick, only
30% of the total available current is absorbed. The photons which are lost are the orange and red photons.

1. T. Tiedje, Yablonovich, E., Cody, G. D., and Brooks, B. G., “Limiting Efficiency of Silicon Solar Cells”, IEEE
TRANSACTIONS ON ELECTRON DEVICES, vol. ED-31, 1984.

Light Trapping
Add new comment
1 comment(s)
Español

The optimum device thickness is not controlled solely by the need to absorb all the light. For example, if the light is not absorbed
within a diffusion length of the junction, then the light-generated carriers are lost to recombination. In addition, as discussed in the
Voltage Losses Due to Recombination, a thinner solar cell which retains the absorption of the thicker device may have a higher
voltage. Consequently, an optimum solar cell structure will typically have "light trapping" in which the optical path length is several
times the actual device thickness, where the optical path length of a device refers to the distance that an unabsorbed photon may travel
within the device before it escapes out of the device. This is usually defined in terms of device thickness. For example, a solar cell with
no light trapping features may have an optical path length of one device thickness, while a solar cell with good light trapping may have
an optical path length of 50, indicating that light bounces back and forth within the cell many times.

Light trapping is usually achieved by changing the angle at which light travels in the solar cell by having it be incident on an angled
surface. A textured surface will not only reduce reflection as previously described but will also couple light obliquely into the silicon,
thus giving a longer optical path length than the physical device thickness. The angle at which light is refracted into the semiconductor
material is, according to Snell's Law, as follows:

Snell's Law
Edit
Delete
Clone

Read more about Snell's Law

where θ1 and θ2 are the angles for the light incident on the interface relative to the normal plane of the interface within the mediums
with refractive indices n1 and n2 respectively. θ1 and θ2 are shown in the animation below.

Refraction of a ray of light at a dielectric boundary. You can adjust the angle of incidence and see how this affects the angle of the ray
transmitted to the second medium by clicking in the right side of the graph and dragging the mouse to change the angle. When n2 has a
higher refractive index than n1 the refracted ray is closer to normal than the incident ray.

By rearranging Snell's law above, the angle at which light enters the solar cell (the angle of refracted light) can be calculated:

Snell's Law (rearranged)

 Edit
Delete
Clone

Read more about Snell's Law (rearranged)

In a textured single crystalline solar cell, the presence of crystallographic planes make the angle θ1 equal to 36° as shown below.

Reflection and transmission of light for a textured silicon solar cell.

The amount of light reflected at an interface is calculated from the fresnel reflection formula. For light polarised in the parrallel to the
surface the amount of reflected light is:

Fresnel Reflection (parrallel)

Edit
Delete
Clone

Read more about Fresnel Reflection (parrallel)

For light polarised perpendicular to the surface the amount reflected is:

Fresnel Reflection (perpendicular)

Edit
Delete
Clone

Read more about Fresnel Reflection (perpendicular)

For unpolarised light the reflected amount is the average of the two:

Fresnel Reflection (plane and perpendicular)

Edit
Delete
Clone
 Read more about Fresnel Reflection (plane and perpendicular)

Light Trapping Calculator

Refractive index of incident medium, n1 =

Refractive index of transmitted medium, n2 =
Incident angle, θ1 = degrees
Refracted angle, θ2 = degrees Proportion
of light reflected, R = Proportion of light
transmitted, T =

If light passes from a high refractive index medium to a low refractive index medium, there is the possibility of total internal reflection
(TIR). The angle at which this occurs is the critical angle and is found by setting θ2 in Snell's law to 0.

Snell's Law (Critical Angle)

Edit
Delete
Clone

Read more about Snell's Law (Critical Angle)

Total Internal Reflection Calculator

Refractive index of incident medium, n1 =

Refractive index of secondary medium, n2 =
Critical angle for total internal reflection to occur,
θ1 = degrees

Using total internal reflection, light can be trapped inside the cell and make multiple passes through the cell, thus allowing even a thin
solar cell to maintain a high optical path length.

Lambertian Rear Reflectors

 Add new comment
Español

A Lambertian back reflector is a special type of rear reflector which randomizes the direction of the reflected light. High reflection off
the rear cell surface reduces absorption in the rear cell contacts or transmission from the rear, allowing the light to bounce back into the
cell for possible absorption. Randomising the direction of light allows much of the reflected light to be totally internally reflected.
Light reaching the top surface at an angle greater than the critical angle for total internal reflection is reflected again towards the back
surface. Light absorption can be dramatically increased in this way, since the pathlength of the incident light can be enhanced by a
factor up to 4n2 where n is the index of refraction for the semiconductor 1. This allows an optical path length of approximately 50
times the physical devices thickness and thus is an effective light trapping scheme. A Lambertian rear surface is illustrated in the figure
below.

Light trapping using a randomised reflector on the rear of the cell. Light less than the critical angle escapes the cell but light greater
than the critical angle is totally internally reflected inside the cell. In actual devices, the front surface is also textured using schemes
such as the random pyramids mentioned earlier.

The graph below shows the short-circuit current of a silicon solar cell with and without light trapping. In the case without light
trapping, the light makes only one pass through the solar cell with some of the light (particularly in the infrared) passing straight
through the cell without being absorbed. In the case with ideal light trapping the path length is increased by 4n2 (4 × 3.52 ≈ 50) and the
short-circuit current is much higher. In practice the level of light trapping is somewhere between these two limits.

Light trapping increases the short-circuit current (JSC) of the solar cell - particularly for thin devices.

1. E. Yablonovich and Cody, G. D., “Intensity Enhancement in Textured Optical Sheets for Solar Cells”, IEEE Transactions on
Electron Devices, vol. ED-29, pp. 300-305, 1982.
Recombination Losses
Add new comment
1 comment(s)
Español

Recombination losses effect both the current collection (and therefore the short-circuit current) as well as the forward bias injection
current (and therefore the open-circuit voltage). Recombination is frequently classified according to the region of the cell in which it
occurs. Typically, recombination at the surface (surface recombination) or in the bulk of the solar cell (bulk recombination) are the
main areas of recombination. The depletion region is another area in which recombination can occur (depletion region recombination).
Bulk and surface recombination are shown in the animation below.

Recombination and collection in a solar cell. The animation proceeds automatically. There is no need to click to continue
or restart.

Current Losses Due to Recombination

Add new comment
2 comment(s)
Español

In order for the p-n junction to be able to collect all of the light-generated carriers, both surface and bulk recombination must be
minimised. In silicon solar cells, the two conditions commonly required for such current collection are:

1. the carrier must be generated within a diffusion length of the junction, so that it will be able to diffuse to the junction before
recombining; and
2. in the case of a localized high recombination site (such as at an unpassivated surface or at a grain boundary in multicrystalline
devices), the carrier must be generated closer to the junction than to the recombination site. For less severe localised
recombination sites, (such as a passivated surface), carriers can be generated closer to the recombination site while still being
able to diffuse to the junction and be collected without recombining.

The presence of localized recombination sites at both the front and the rear surfaces of a silicon solar cell means that photons of
different energy will have different collection probabilities. Since blue light has a high absorption coefficient and is absorbed very
close to the front surface, it is not likely to generate minority carriers that can be collected by the junction if the front surface is a site of
high recombination. Similarly, a high rear surface recombination will primarily affect carriers generated by infrared light, which can
generate carriers deep in the device. The quantum efficiency of a solar cell quantifies the effect of recombination on the light
generation current. The quantum efficiency of a silicon solar cell is shown below.
Typical quantum efficiency in an ideal and actual solar cell, illustrating the impact of optical and recombination losses.

Voltage Losses due to Recombination

Add new comment
Español

The open-circuit voltage is the voltage at which the forward bias diffusion current is exactly equal to the short circuit current. The
forward bias diffusion current is dependent on the amount recombination in a p-n junction and increasing the recombination increases
the forward bias current. Consequently, high recombination increases the forward bias diffusion current, which in turn reduces the
open-circuit voltage. The material parameter which gives the recombination in forward bias is the diode saturation current. The
recombination is controlled by the number of minority carriers at the junction edge, how fast they move away from the junction and
how quickly they recombine. Consequently, the dark forward bias current, an hence the open-circuit voltage is affected by the
following parameters:

1. the number of minority carriers at the junction edge. The number of minority carriers injected from the other side is simply the
number of minority carriers in equilibrium multiplied by an exponential factor which depends on the voltage and the
temperature. Therefore, minimising the equilibrium minority carrier concentration reduces recombination. Minimizing the
equilibrium carrier concentration is achieved by increasing the doping;
2. the diffusion length in the material. A low diffusion length means that minority carriers disappear from the junction edge quickly
due to recombination, thus allowing more carriers to cross and increasing the forward bias current. Consequently, to minimise
recombination and achieve a high voltage, a high diffusion length is required. The diffusion length depends on the types of
material, the processing history of the wafer and the doping in the wafer. High doping reduces the diffusion length, thus
introducing a trade-off between maintaining a high diffusion length (which affects both the current and voltage) and achieving a
high voltage;
3. the presence of localised recombination sources within a diffusion length of the junction. A high recombination source close to
the junction (usually a surface or a grain boundary) will allow carriers to move to this recombination source very quickly and
recombine, thus dramatically increasing the recombination current. The impact of surface recombination is reduced by
passivating the surfaces.

The net effect of previous trade-offs is shown in the graphs below.

Effect of doping (ND) on diffusion length and open-circuit voltage assuming well passivated surfaces.

Surface Recombination
Add new comment
Español

Surface recombination can have a major impact both on the short-circuit current and on the open-circuit voltage. High recombination
rates at the top surface have a particularly detrimental impact on the short-circuit current since top surface also corresponds to the
highest generation region of carriers in the solar cell. Lowering the high top surface recombination is typically accomplished by
reducing the number of dangling silicon bonds at the top surface by using "passivating" layer on the top surface. The majority of the
electronics industry relies on the use of a thermally grown silicon dioxide layer to passivate the surface due to the low defect states at
the interface1. For commercial solar cells, dielectric layers such as silicon nitride are commonly used.

Techniques for reducing the impact of surface recombination.

Since the passivating layer for silicon solar cells is usually an insulator, any region which has an ohmic metal contact cannot be
passivated using silicon dioxide. Instead, under the top contacts the effect of the surface recombination can be minimised by increasing
the doping. While typically such a high doping severely degrades the diffusion length, the contact regions do not participate in carrier
generation and hence the impact on carrier collection is unimportant. In addition, in cases where a high recombination surface is close
to the junction, the lowest recombination option is to increase the doping as high as possible.

Back Surface Field

A similar effect is employed at the rear surface to minimise the impact of rear surface recombination velocity on voltage and current if
the rear surface is closer than a diffusion length to the junction. A "back surface field" (BSF) consists of a higher doped region at the
rear surface of the solar cell. The interface between the high and low doped region behaves like a p-n junction and an electric field
forms at the interface which introduces a barrier to minority carrier flow to the rear surface. The minority carrier concentration is thus
maintained at higher levels in the bulk of the device and the BSF has a net effect of passivating the rear surface.2

1. W. D. Eades and Swanson, R. M., “Calculation of surface generation and recombination velocities at the Si-SiO2 interface”,
Journal of Applied Physics, vol. 58, p. 4267, 1985.
2. J. G. Fossum, “Physical operation of back-surface-field silicon solar cells”, IEEE Transactions on Electron Devices, vol. 24,
pp. 322 - 325, 1977.

Series Resistance
Add new comment
Español

In addition to maximizing absorption and minimizing recombination, the final condition necessary to design a high efficiency solar cell
is to minimize parasitic resistive losses. Both shunt and series resistance losses decrease the fill factor and efficiency of a solar cell. A
detrimentally low shunt resistance is a processing defect rather than a design parameter. However, the series resistance, controlled by
the top contact design and emitter resistance, needs to be carefully designed for each type and size of solar cell structure in order to
optimize solar cell efficiency.

The series resistance of a solar cell consists of several components as shown in the diagram below. Of these components, the emitter
and top grid (consisting of the finger and busbar resistance) dominate the overall series resistance and are therefore most heavily
optimized in solar cell design.
Resistive components and electron flow in a solar cell.

The metallic top contacts are necessary to collect the current generated by a solar cell. "Busbars" are connected directly to the external
leads, while "fingers" are finer areas of metalization which collect current for delivery to the busbars. The key design trade-off in top
contact design is the balance between the increased resistive losses associated with a widely spaced grid and the increased reflection
caused by a high fraction of metal coverage of the top surface.

Top contact design in a solar cell. The busbars connect the fingers together and pass the generated current to the external electrical
contacts.

Base Resistance
Add new comment
1 comment(s)
Español

Generated current typically flows perpendicular to the cell surface from the bulk of the cell and then laterally through the top doped
layer until it is collected at a top surface contact.

The resistance and current of the base is assumed to be constant. The resistance to the current of the bulk component of the cell, or the
"bulk resistance", Rb, is defined as:

taking into account the thickness of the material. Where:

l = length of conducting (resistive) path

ρb= "bulk resistivity" (inverse of conductivity) of the bulk cell material (0.5 - 5.0 Ω cm for a typical silicon solar cell)
A = cell area, and
w = width of bulk region of cell.
Sheet Resistivity
Add new comment
Español

For the emitter layer, the resistivity as well as the thickness of the layer will often be unknown, making the resistance of the top layer
difficult to calculate from the resistivity and thickness. However, a value known as the "sheet resistivity", which depends on both the
resistivity and the thickness, can be readily measured for the top surface n-type layer. For a uniformly doped layer, the sheet resistivity
is defined as:

where
ρ is the resistivity of the layer; and
t is the thickness of the layer.

The sheet resistivity is normally expressed as ohms/square or Ω/□.

The resistance of a square conductive sheet is the same no matter what size it is so long as it remains a square.

For non-uniformly doped n-type layers, ie., if ρ is non-uniform:

The sheet resistivity of an emitter layer is typically measured with a four-point-probe.

Emitter Resistance
Add new comment
Español

Based on the sheet resistivity, the power loss due to the emitter resistance can be calculated as a function of finger spacing in the top
contact. However, the distance that current flows in the emitter is not constant. Current can be collected from the base close to the
finger and therefore has only a short distance to flow to the finger or, alternatively, if the current enters the emitter between the fingers,
then the length of the resistive path seen by such a carrier is half the grid spacing.

Idealised current flow from point of generation to external contact in a solar cell. The emitter is typically much thinner than shown in
the diagram.

The incremental power loss in the section dy is given by:

The differential resistance is given by:

where

ρ is the sheet resistivity in Ω/sqr

b is the distance along the finger; and
y the distance between two grid fingers as shown below.

Dimensions needed for calculating power loss due to the lateral resistance of the top layer.

The current also depends on y and I(y) is the lateral current flow, which is zero at the midpoint between grating lines and increases
linearly to its maximum at the grating line, under uniform illumination. The equation for the current is:

where
J is the current density;
b is the distance along the finger; and
y the distance between two grid fingers as shown above.

The total power loss is therefore:

where
S is the spacing between grid lines.

Emitter Resistance Loss Calculator

Input Parameters

Current Density, J (A/cm2) Finger Length, b (cm) Emitter Sheet Resistivity, ρ

(Ω/square ) Finger Spacing, S (cm)

Results

Power Loss, Ploss mW

At the maximum power point, the generated power is:

The fractional power loss is given by:

Hence, the minimum spacing for the top contact grid can be calculated. For example, for a typical silicon solar cell where ρ= 40 Ω/sq,
Jmp = 30 mA/cm2, Vmp = 450 mV, to have a power loss in the emitter of less than 4% the finger spacing should be less than 4 mm.

Fractional Emitter Power Loss Calculator

Current density at max power, Jmp = A/cm2

Voltage at max power, Vmp = V Emitter Sheet
Resistivity, ρ = Ω/square Finger Spacing, S =
cm
Fractional power loss P% = %

Contact resistance
Add new comment
Español

Contact resistance losses occur at the interface between the silicon solar cell and the metal contact. To keep top contact losses low, the
top N+ layer must be as heavily doped as possible. However, a high doping level creates other problems. If a high level of phosphorus
is diffused into silicon, the excess phosphorus lies at the surface of the cell, creating a "dead layer", where light generated carriers have
little chance of being collected. Many commercial cells have a poor "blue" response due to this "dead layer". Therefore, the region
under the contacts should be heavily doped, while the doping of the emitter is controlled by the trade-offs between achieving a low
saturation current in the emitter and maintaining a high emitter diffusion length.

Points of contact resistance losses at interface between grid lines and semiconductor.

In commercial screen printed solar cells the contact resistance varies across the wafer. The physics of silver paste firing are quite
complicated so small differences in surface topology and local heating cause large variations in the quality of the silver-silicon bond.

The contact resistance of a solar cell can be measured in a variety of ways. One popular method is to bias the cell at the maximum
power point an then measure voltage drops along the cell 1. The method can be automated to produce a map of the device showing
regions of very large contact resistance.
Corescan of a large area screen printed solar cell. Due to unoptimised processing conditions parts of the cell at the center have a larger
contact resistance.

1. A. S. H. van der Heide and et al, “Mapping of contact resistance and locating shunts on solar cells using Resistance Analysis
by Mapping of Potential (RAMP) techniques”, 16th European Photovoltaic Solar Energy Conference. Glasgow (United
Kingdom), p. 1438, 2000.

Finger Resistance
Add new comment
Español

To provide higher conductivity the top of a cell has a series of regularly spaced finger. While tapered fingers theoretically provide
lower losses technology limitations mean that fingers are usually uniform in width. The resistive loss in a finger is calculated as below.
At the end of this page there is a graph showing how the parameters affect power loss in the cell.

Calculation of Power Loss in the Fingers

Calculation of the power loss in a single finger. The width is assumed constant and it is assumed that the current is uniformly generated
and that it flows perpendicularly into the finger, i.e., no current flow directly into the busbar.

Consider an element dx at a distance x from the end of the finger.

The current through the element dx is: ,

where Jmp is the current at maximum power point and Sf is the finger spacing.

The resistance of the element dx is: ,

where wf is the finger width, df is the finger depth (or height) and ρf is the effective resistivity of the metal.

The power loss in the element dx is:

Integrating x from 0 to L gives the power loss in the finger:

Optimization of Finger Spacing

Add new comment
Español

Combining the equations for resistive losses allows use to deteremine the total power loss in the top contact grid. In the following
graphical simulation adjust the various parameters and determine the total power loss. For a typical cell type, say a screen printed cell,
the metal resistivity will be fixed and the finger width is controlled by the screen size. Typical values for the specific resistivity of
silver are 3 × 10-8 Ω m 1. For non-rectangular fingers the width is set to the actual width and an equivalent height is used to get the
correct cross sectional area.

The simulation does not include the contact loss between the metal and the semiconductor or the resistive losses in the bus bars.

Finger Length, L = 2 cm
 1 10

Emitter Sheet Resistivity, ρ□ = 100 Ω/□

20 200

Metal Resistivity, ρmetal = 3×10-8 Ω m

1 20

Finger Width, wf = 14 µm

1 200

Finger Depth, df = 14 µm

1 100

The interactive graph determines the total power and the optimum finger spacing.

1. A. Mette and et al, “Series resistance characterization of industrial silicon solar cells with screen-printed contacts using
hotmelt paste”, Progress in Photovoltaics: Research and Applications, vol. 15, pp. 493-505, 2007.

Metal Grid Pattern

Add new comment
1 comment(s)
Español

The design of the top contact involves not only the minimization of the finger and busbar resistance, but the overall reduction of losses
associated with the top contact. These include resistive losses in the emitter, resistive losses in the metal top contact and shading losses.
The critical features of the top contact design which determine how the magnitude of these losses are the finger and busbar spacing, the
metal height-to-width aspect ratio, the minimum metal line width and the resistivity of the metal. These are shown in the figure below.

Key features of a top surface contacting scheme.

Impact of Finger Spacing on Emitter Resistance

An important factor in top contact design is that of resistive losses in the emitter. As shown in the Emitter Resistance page, the power
loss from the emitter depends on the cube of the line spacing, and therefore a short distance between the fingers is desirable for a low
emitter resistance.

Grid Resistance

The grid resistance is determined by the resistivity of the metal used to make the metal contact, the pattern of the metalization and on
the aspect ratio of the metalization scheme. A low resistivity and a high metal height-to-width aspect ratio are desirable in solar cells,
but in practice are limited by the fabrication technology used to make the solar cell.
Shading Losses

Shading losses are caused by the presence of metal on the top surface of the solar cell which prevents light from entering the solar cell.
The shading losses are determined by the transparency of the top surface, which, for a planar top surface, is defined as the fraction of
the top surface covered by metal. The transparency is determined by the width of the metal lines on the surface and on the spacing of
the metal lines. An important practical limitation is the minimum linewidth associated with a particular metalization technology. For
identical transparencies, a narrow line-width technology can have closer finger spacing, thus reducing the emitter resistance losses.

Design Rules

While a multitude of top contacting schemes exist, for practical reasons most top surface metalization patterns are relatively simple and
highly symmetrical. A symmetrical contacting scheme can be broken down into unit cells and several broad design rules can be
determined. It can be shown 1 that:

the optimum width of the busbar, WB, occurs when the resistive loss in the busbar equals its shadowing loss;
a tapered busbar has lower losses than a busbar of constant width; and
the smaller the unit cell, the smaller finger width, WF , and the smaller the finger spacings, S, the lower the losses.

Schematic of a top contact design showing busbars and fingers 1.

A program for designing and developing the front surface grid pattern is available at the PV Lighthouse Metal Grid Calculator

1. a. b. H. B. Serreze, “Optimizing Solar Cell Performance by Simultaneous Consideration of Grid Pattern Design and
Interconnect Configurations”, 13th IEEE Photovoltaic Specialists Conference. Washington, D.C., USA, pp. 1-8, 1978.

Silicon Solar Cell Parameters

Add new comment
Español

For silicon solar cells, the basic design constraints on surface reflection, carrier collection, recombination and parasitic resistances
result in an optimum device of about 25% theoretical efficiency. A schematic of such an optimum device is shown below.
Basic schematic of a silicon solar cell. The top layer is referred to as the emitter and the bulk material is referred to as the base.

Basic Cell Design Compromises

Substrate Material
(usually silicon)

Bulk crystalline silicon dominates the current photovoltaic market, in part due to the prominence of silicon in the integrated circuit
market. As is also the case for transistors, silicon does not have optimum material parameters. In particular, silicon's band gap is
slightly too low for an optimum solar cell and since silicon is an indirect material, it has a low absorption co-efficient. While the low
absorption co-efficient can be overcome by light trapping, silicon is also difficult to grow into thin sheets. However, silicon's
abundance, and its domination of the semiconductor manufacturing industry has made it difficult for other materials to compete.

Cell Thickness
(100-500 µm)

An optimum silicon solar cell with light trapping and very good surface passivation is about 100 µm thick. However, thickness
between 200 and 500µm are typically used, partly for practical issues such as making and handling thin wafers, and partly for surface
passivation reasons.

Doping of Base
(1 Ω·cm)

A higher base doping leads to a higher Voc and lower resistance, but higher levels of doping result in damage to the crystal.

Reflection Control
(front surface typically textured)

The front surface is textured to increase the amount of light coupled into the cell.

Emitter Dopant
(n-type)

N-type silicon has a higher surface quality than p-type silicon so it is placed at the front of the cell where most of the light is absorbed.
Thus the top of the cell is the negative terminal and the rear of the cell is the positive terminal.

Emitter Thickness
(<1 μm)

A large fraction of light is absorbed close to the front surface. By making the front layer very thin, a large fraction of the carriers
generated by the incoming light are created within a diffusion length of the p-n junction.
Doping Level of Emitter
(100 Ω/☐)

The front junction is doped to a level sufficient to conduct away the generated electricity without resistive loses. However, excessive
levels of doping reduces the material's quality to the extent that carriers recombine before reaching the junction.

Grid Pattern.
(fingers 20 to 200 µm width, placed 1 - 5 mm apart)

The resistivity of silicon is too high to conduct away all the current generated, so a lower resistivity metal grid is placed on the surface
to conduct away the current. The metal grid shades the cell from the incoming light so there is a compromise between light collection
and resistance of the metal grid.

Rear Contact.

The rear contact is much less important than the front contact since it is much further away from the junction and does not need to be
transparent. The design of the rear contact is becoming increasingly important as overall efficiency increases and the cells become
thinner.

Efficiency and Solar Cell Cost

Add new comment
Español

Under laboratory conditions and with current state-of-the-art technology, it is possible to produce single crystal silicon solar cells close
to 25% efficient. However, commercially mass produced cells are typically only 13-14% efficient. The overriding reason for this
difference in efficiency is that the research techniques used in the laboratory are not suitable for commercial production within the
photovoltaic industry and therefore lower cost techniques, which result in lower efficiency, are used.

Solar cell research continues to improve the efficiency of solar cells, with targets aimed towards the currently accepted limit of
25-30%. Efficiency results for commercially produced solar cells lag some years behind efficiency results for laboratory produced
cells. However, it is estimated module efficiencies over 20% could be produced commercially. For a given module cost, more efficient
modules are also more cost effective because fewer additional costs (such as lade area, wiring costs, etc.) are needed to generate the
same amount of power. Ultimately, this means that the more efficient a PV module is, and thus the more cost effective it becomes, the
more attractive photovoltaics will be as an electricity source for both domestic and industrial use. The impact of module efficiency on
electricity cost is shown in the figure below.
With higher efficiency modules, the cost per unit area can be much higher for a given cost target of electricity in kWh. To achieve the
proposed target with 10% efficient modules requires that the modules be less than $10/m2. With modules of 20% efficiency, it is still
possible to meet the proposed target with modules that are $75/m21.

1. Citekey DOE1987 not found

Chapter 5 Quiz
Questions 8
Attempts allowed Unlimited
Available Always
Pass rate 75 %
Backwards navigation Allowed
Start Quiz

Manufacturing Si Cells
Add new comment
Español

First photovoltaic Devices

Add new comment
Español

(Taken from Green1 which makes use of Benjamin2, Shive3 and Wolf4)
Edmond Becquerel appears to have been the first to demonstrate the photovoltaic effect5 6. Working in his father's laboratory as a
nineteen year old, he generated electricity by illuminating an electrode with different types of light, including sunlight (see the figure
below). Best results were obtained with blue or ultraviolet light and when electrodes were coated with light sensitive material such as
AgCl or AgBr. Although he usually used platinum electrodes, he also observed some response with silver electrodes. He subsequently
found a use for the photovoltaic effect by developing an "actinograph" which was used to record the temperature of heated bodies by
measuring the emitted light intensity.

Diagram of apparatus described by Becquerel (1839)

The next significant photovoltaic development arose from the interest in the photoconductive effect in selenium. While investigating
this effect, Adams and Day (1877)7 noted an anomaly they thought could be explained by the generation of internal voltages. They
investigated this anomaly more carefully using samples as shown below. Heated platinum contacts were pushed into opposite ends of
small cylinders of vitreous selenium. The objective of one experiment conducted by Adams and Day upon such specimens was to see
'whether it would be possible to start a current in the selenium merely by the action of light'.

Sample geometry used by Adams and Day (1876) for the investigation of the photoelectric effects in selenium.

The result was positive! This was the first demonstration of the photovoltaic effect in an all solid-state system. Adams and Day
attributed the photogenerated currents to light induced crystallization of the outer layers of the selenium bar. Several decades were to
pass before the development of physics allowed more insight into this process.

The next significant step forward came seven years later with the work of Fritts (1883) 8. By compressing molten selenium between
plates made from two different metals, Fritts was able to prepare thin Se films which adhered to one of the two plates, but not to the
other. By pressing a gold leaf to the exposed selenium surface, he thereby prepared the first "thin-film" photovoltaic devices. These
first thin-film devices were as large as 30 cm2 in area.
Thin-film selenium demonstrated by Fritts in 1883.

He was also the first to recognize the enormous potential of photovoltaic devices. He saw that the devices could be fabricated at very
low cost and noted that 'the current, if not wanted immediately, can either be "stored" where produced, in storage batteries … or
transmitted … to a distance, and there used, or stored'.

It was, however, to be nearly fifty years before there was another significant burst of activity in this area.

While studying photoconductive effects in cuprous oxide layers grown on copper, the rectifying action of the copper-cuprous oxide
junction was discovered. This led to the development of large-area rectifiers, followed soon after by large-area photocells. Grondahl 9
describes the development of both copper-cuprous oxide rectifiers and photovoltaic cells.

The figure below shows the very simple structure used by the earlier cells based on the copper-cuprous oxide junction. A coil of Pb
wire is used to give a grid contact to the illuminated surface of the cell. This approach was subsequently refined by sputtering the metal
on the outer surface and removing a part of it so as 'to form a grid of any desired fineness'. These developments seem to have
stimulated a great deal of activity in this area. Grondahl 9 documents 38 publications dealing with copper-cuprous oxide photovoltaic
cells over the period 1930-32. 10

Early Grondahl-Geiger copper-cuprous oxide photovoltaic cell (circa 1927).

This activity also seems to have reawakened interest in selenium as a photovoltaic material. In particular, Bergmann 11 reported
improved selenium devices in 1931. These proved superior to the copper-based devices and became the commercially dominant
product. In 1939, a thallous-sulphide cell of similar performance was also reported by Nix 10. The structure of this device and of the
most efficient selenium and copper-cuprous oxide devices took the form shown in the figure below.

Structure of the most efficient photovoltaic devices developed during the 1930's.

1. M. A. Green, “Photovoltaics: Coming of Age”, 21st IEEE Photovoltaic Specialists Conference. Orlando, USA, pp. 1-8, 1990.
2. P. Benjamin, “Voltaic Cell, Chapter XIV”, New York: Wiley, 1983.
3. J. N. Shive, “Semiconductor Devices, Chapter 8”, New Jersey: Van Nostrand, 1959.
 4. M. Wolf, “Historical Development of Solar Cells”. IEEE Press, 1976.
5. A. E. Becquerel, “Recherches sur les effets de la radiation chimique de la lumiere solaire au moyen des courants electriques”,
Comptes Rendus de L´Academie des Sciences, vol. 9, pp. 145-149, 1839.
6. A. E. Becquerel, “Memoire sur les effects d´electriques produits sous l´influence des rayons solaires”, Annalen der Physick
und Chemie, vol. 54, pp. 35-42, 1841.
7. W. G. Adams and Day, R. E., “The Action of Light on Selenium”, Proceedings of the Royal Society, London, vol. A25, p. 113,
1877.
8. C. E. Fritts, “On a New Form of Selenium Photocell”, American J. of Science, vol. 26, p. 465, 1883.
9. a. b. L. O. Grondahl, “The Copper-Cuprous-Oxide Rectifier and Photoelectric Cell”, Review of Modern Physics, vol. 5, p. 141,
1933.
10. a. b. F. C. Nix and Treptwo, A. W., “A Thallous Sulphide Photo EMF Cell”, Journal Opt. Society of America, vol. 29, p. 457,
1939.
11. L. Bergmann, “Uber eine neue Selen- Sperrschicht Photozelle”, Physikalische Zeitschrift, vol. 32, p. 286, 1931.

Early Silicon Cells

Add new comment
Español

(Coninuation of the previous page. Taken from 1)

At about the same time that selenium cells were under development in the 1930's, silicon was receiving increasing attention for its use
in point-contact rectifiers. The rectifying properties of sharpened metal point contacts to various crystals had been known since at least
1874 2. In the early days of radio, such crystal rectifiers were the most commonly used detectors in radio receivers, but with the
development of thermionic tubes, crystal rectifiers were replaced in all but ultra-high frequency applications. Tungsten points proved
to be the most suitable material for making contact with silicon surfaces. Research into the purity of silicon also led to further
understanding of its properties.

In studies of recrystallized melts of pure silicon prepared for this purpose, Ohl 3 discovered the presence of a well-defined barrier in
ingots grown from commercially available, high-purity silicon.

(a) Cast ingot showing natural junction formed by impurity segregation during melting; (b) photovoltaic device cut perpendicular to
junction; (c) device cut parallel to junction; (d) top surface of device cut parallel to junction.

In 1941, before even this limited understanding of dopants, silicon photovoltaic devices based on these natural junctions were
described 3. In the figure above, (a) shows the naturally grown junction in a cast ingot of silicon. Ingots were produced from acid-
leached metallurgical-grade silicon which were cooled from the top after melting. Cells were prepared by cutting from the ingot as
indicated in (a). Alternatively, devices could be cut parallel to the junction as shown in (c).
This natural barrier was first found in rods cut from ingots for resistivity measurements. The rods showed good photovoltaic response,
were found to have a high thermoelectric coefficient, and had good rectifying properties. One end of the rod developed a negative
potential when illuminated or heated and had to be biased negatively in order to show low resistance to current flow across the barrier
or across a point contact to this material. Material with these properties became known as negative- or n-type silicon, and material of
the opposite type was named positive-, or p-type. Subsequently the role of donor and acceptor impurities in producing these properties
was shown.

Although these devices performed similarly to the thin-film devices then available, the preparation method did not obviously lend itself
to convenient manufacture. However, it was clear that if a suitable method could be found to activate large areas of silicon surface
uniformly, competitive cells could be made. In 1952, improved silicon solar cells were reported by Kingsbury and Ohl 4 using purer
silicon to prevent growth junction formation and the ion bombardment of the surface to form the rectifying junction.

Meanwhile, development of crystal growth techniques and techniques for forming junctions by diffusion resulted in the announcement
of the first modern silicon cell in 1954 by Chapin, Fuller and Pearson 5. These cells had the dual rear contact structure of the figure
below and had an efficiency of 6%, about 15 times that of earlier devices, opening the first real prospects for power generation using
photovoltaics. Considerable interest was aroused 6. However, given the immaturity of the silicon preparation industry, it soon became
apparent that the initial enthusiasm was premature. Nevertheless, the cells proved suitable for space use and this formed their major
application until the early 1970's.

Early silicon solar cell structure.

1. M. A. Green, “Photovoltaics: Coming of Age”, 21st IEEE Photovoltaic Specialists Conference. Orlando, USA, pp. 1-8, 1990.
2. F. Braun, “On Conductance in Metal Sulphides”, Ann. d. Physik, vol. 153, p. 556, 1874.
3. a. b. R. S. Ohl, “Light-Sensitive Electric Device”, U.S. Patent, vol. 2, p. 402, 602, 1941.
4. E. F. Kingsbury and Ohl, R. S., “Photoelectric Properties of Tonically Bombarded Silicon”, Bell Systems Technical Journal,
vol. 31, pp. 802-815, 1952.
5. D. M. Chapin, Fuller, C. S., and Pearson, G. L., “A New Silicon P-N Junction Photocell for Converting Solar Radiation into
Electrical Power”, Journal of Applied Physics, vol. 25, pp. 676-677, 1954.
6. “Bell Laboratories Record”. 1955.

Refining Silicon
Add new comment
2 comment(s)
Español

Silicon dioxide (SiO2) is the most abundant mineral in the earth's crust. The manufacture of the hyperpure silicon for photovoltaics
occurs in two stages. The oxygen is removed to produce metallurgical grade silicon. It is further refined to produce semiconductor
grade silicon. An intermediate grade with impurity levels between metallurgical silicon and semiconductor grade silicon is often
termed solar grade silicon.

Metallurgical Grade Silicon

Silica is the dioxide form of silicon (SiO2) and occurs naturally in the form of quartz. While beach sand is also largely quartz, the most
common raw material for electronic grade is high purity quartz rock. Ideally the silica has low concentrations of iron, aluminum and
other metals. The silica is reduced (oxygen removed) through a reaction with carbon in the form of coal, charcoal and heating to
1500-2000 °C in an electrode arc furnace.

SiO2 + C → Si + CO2

The resulting silicon is metallurgical grade silicon (MG-Si). It is 98% pure and is used extensively in the metallurgical industry.

An even greater production of silicon is in the form of ferrosilicon that is manufactured using a similar process to that described above
but is done in the presence of iron. Ferrosilicon is used extensively in metals manufacturing. In 2013 the total production of silicon was
7.6 million tonnes and 80 % of that was in the form of ferrosilicon.1
Of the 1.8 million tonnes of metallurgical silicon produced in 2010, 12 % was for the production of silicon solar cells.2

Metallurgical (MG) silicon is produced at the rate of millions of tons/year at a low economic cost of few $/kg and an energy cost of
14–16 kWh/kg. As such, it is 98–99% pure, with a major contamination of carbon, alkali-earth and transition metals, and hundreds of
ppmw of B and P.3

The transition metals in the silicon result in deep levels in the bandgap and the high recombination activity make metallurgical grade
silicon unsuitable for use in electronics. In addition, boron and phosphorous dopant impurities are much too high in concentration
(>50–100 ppmw) to allow suitable compensation procedures 4

The impurity levels in metallurgical grade silicon5 vary widely due to process variation and the source raw materials of silica and
carbon

Electronic Grade Silicon

A small amount of the metallurgical grade silicon is further refined for the semiconductor industry. Powdered MG-Si is reacted with
anhydrous HCl at 300 °C in a fluidized bed reactor to form SiHCl3

Si + 3HCl → SiHCl3 + H2
During this reaction impurities such as Fe, Al, and B react to form their halides (e.g. FeCl3, AlCl3, and
BCl3). The SiHCl3 has a low boiling point of 31.8 °C and distillation is used to purify the SiHCl3 from the
impurity halides. The resulting SiHCl3 now has electrically active impurities(such as Al, P, B, Fe, Cu or Au) of less than 1 ppba.

Finally, the pure SiHCl3 is reacted with hydrogen at 1100°C for ~200 – 300 hours to produce a very pure form of silicon.

SiHCl3 + H2 →Si + 3 HCl

The reaction takes place inside large vacuum chambers and the silicon is deposited onto thin polysilicon rods (small grain size silicon)
to produce high-purity polysilicon rods of diameter 150-200mm. The process was first developed by Siemens in the 60's and is often
referred to as the Siemens process.

The resulting rods of semiconductor grade silicon are broken up to form the feedstock for the crystallisation process. The production of
semiconductor grade silicon requires a lot of energy. Solar cells can tolerate higher levels of impurity than integrated circuit fabrication
and there are proposals for alternative processes to create a "solar-grade" silicon.

1. L. A. Corathers, “Minerals Yearbook, Vol. I, Metals & Minerals:”. U.S. Government Printing Office, p. 144, 2013.
2. “Silicon and Ferrosilicon: Global Industry Markets and Outlook, 13th edition”, Roskill, London, 2011.
3. S. Pizzini, “Towards solar grade silicon: Challenges and benefits for low cost photovoltaics”, Solar Energy Materials and
Solar Cells, vol. 94, no. 9, pp. 1528 - 1533, 2010.
4. S. Pizzini and Calligarich, C., “On the Effect of Impurities on the Photovoltaic Behavior of Solar-Grade Silicon”, Journal of
The Electrochemical Society, vol. 131, no. 9, p. 2128, 1984.
5. C. Smith and Barron, A., “Synthesis and Purification of Bulk Semiconductors”, 2012. [Online]. Available: http://cnx.org
/content/m23936/1.7/.

Types of Silicon
Add new comment
Español

Silicon or other semiconductor materials used for solar cells can be single crystalline, multicrystalline, polycrystalline or amorphous.
The key difference between these materials is the degree to which the semiconductor has a regular, perfectly ordered crystal structure,
and therefore semiconductor material may be classified according to the size of the crystals making up the material.

Terms for Crystalline Silicon Solar Cells 1.

Terminology for various types of crystalline silicon (c-Si).

Descriptor Symbol Grain Size Common Growth Techniques

Single crystal sc-Si >10cm Czochralski (CZ) float zone (FZ)

Multicrystalline mc-Si 1mm-10cm Cast, sheet, ribbon

Polycrystalline pc-Si 1µm-1mm Chemical-vapour deposition

Microcrystalline µc-Si <1µm Plasma deposition

1. P. A. Basore, “Defining terms for crystalline silicon solar cells”, Progress in Photovoltaics: Research and Applications, vol. 2,
pp. 177-179, 1994.

Single Crystalline Silicon

Add new comment
Español

The majority of silicon solar cells are fabricated from silicon wafers, which may be either single-crystalline or multi-crystalline.
Single-crystalline wafers typically have better material parameters but are also more expensive. Crystalline silicon has an ordered
crystal structure, with each atom ideally lying in a pre-determined position. Crystalline silicon exhibits predictable and uniform
behaviour but because of the careful and slow manufacturing processes required, it is also the most expensive type of silicon.
The regular arrangement of silicon atoms in single-crystalline silicon produces a well-defined band structure. Each silicon atom has
four electrons in the outer shell. Pairs of electrons from neighbouring atoms are shared so each atom shares four bonds with the
neighbouring atoms.

Single crystalline silicon is usually grown as a large cylindrical ingot producing circular or semi-square solar cells. The semi-square
cell started out circular but has had the edges cut off so that a number of cells can be more efficiently packed into a rectangular module.

Orientation and Doping

In single crystalline silicon material the crystal orientation is defined by Miller indices. A particular crystal plane is noted using
parenthesis such as (100). Silicon has a cubic symmetrical cubic structure and so (100), (010) etc are equivalent planes and collectively
referred to using braces {100}. Similarly, the crystal directions are defined using square brackets, e.g. [100] and referred collectively
using triangular brackets, <100>.

In solar cells the preferred orientation is <100> as this can be easily textured to produce pyramids that reduce the surface reflectivity.
However, some crystal growth processes such as dendritic web <111> produce material with other orientations.

To denote the crystal directions, single crystal wafers often have flats to denote the orientation of the wafer and the doping. The most
common standard is the SEMI standard:

1. If the minor flat is 180° from the major flat the wafer is n-type <100>
2. If the minor flat is 90° to the left or right the wafer is p-type <100>.
3. If the minor flat is 45° up on the left or right the wafer is n-type <111>
4. If there are no minor flats the wafer is p-type <111>

Czochralski Silicon
 Add new comment
Español

Single crystalline substrates are typically differentiated by the process by which they are made. Czochralski (Cz)1 wafers are the most
commonly used type of silicon wafer, and are used by both the solar and integrated circuit industry. The process of making a large
single crystalline silicon ingot by the Czochralski process is shown below. The use of quartz crucibles in the manufacture of Cz
substrates causes the incorporation of ppm (1018 cm-3) oxygen into the silicon ingot. The oxygen itself is relatively benign but creates
complexes with boron doping that degrades the carrier lifetime.2,3 N-type ingots fabricated with phosphorous dopants have similar
oxygen concentrations but do not show the degradation effect nor do wafers with lower resistivity or gallium dopants. 4
Top of Czochralski ingot. The bottom cylindrical section has been cut off to make wafers. Such "tops and tails" left over from growing
the semiconductor industry are a large source of silicon supply for the photovoltaics industry.

1. J. Czochralski, “Ein neues Verfahren zur Messung der Kristallisationsgeschwindigheit der Metalle”, Zeitschrift für
physikalische Chemie, vol. 92, pp. 219–221, 1918.
2. S. W. Glunz, Rein, S., Warta, W., Knobloch, J., and Wettling, W., “Degradation of carrier lifetime in Cz silicon solar cells”,
Solar Energy Materials and Solar Cells, vol. 65, pp. 219 - 229, 2001.
3. K. Bothe, Sinton, R., and Schmidt, J., “Fundamental boron-oxygen-related carrier lifetime limit in mono- and multicrystalline
silicon”, Progress in Photovoltaics: Research and Applications, vol. 13, pp. 287 - 296, 2005.
4. J. Zhao, Wang, A., and Green, M. A., “24.5% efficiency PERT silicon solar cells on SEH MCZ substrates and cell
performance on other SEH CZ and FZ substrates”, Solar Energy Materials and Solar Cells, vol. 66, pp. 27 - 36, 2001.

Float Zone Silicon

Add new comment
Español

Even though the CZ process is commonly used for commercial substrates, it has several disadvantages for high efficiency laboratory or
niche market solar cells. CZ wafers contain a large amount of oxygen in the silicon wafer. Oxygen impurities reduce the minority
carrier lifetime in the solar cell, thus reducing the voltage, current and efficiency. In addition, the oxygen and complexes of the oxygen
with other elements may become active at higher temperatures, making the wafers sensitive to high temperature processing. To
overcome these problems, Float Zone (FZ) wafers may be used1. In this process, a molten region is slowly passed along a rod or bar of
silicon. Impurities in the molten region tend to stay in the molten region rather than be incorporated into the solidified region, thus
allowing a very pure single crystal region to be left after the molten region has passed. Due to the difficulty in growing large diameter
ingots and the often higher cost, FZ wafers are typically only used for laboratory cells and are less common in commercial
production.2

Schematic of Float Zone wafer growth.

1. W. G. Pfann, “Zone-refining”, Trans. AIME, vol. 194, 1952.

2. M. A. Green, “The path to 25% silicon solar cell efficiency: History of silicon cell evolution”, Progress in Photovoltaics:
Research and Applications, vol. 17, pp. 183-189, 2009.

Multi Crystalline Silicon

 Add new comment
1 comment(s)
Español

Techniques for the production of multicrystalline silicon are more simple, and therefore cheaper, than those required for single crystal
material. However, the material quality of multicrystalline material is lower than that of single crystalline material due to the presence
of grain boundaries. Grain boundaries introduce high localized regions of recombination due to the introduction of extra defect energy
levels into the band gap, thus reducing the overall minority carrier lifetime from the material. In addition, grain boundaries reduce solar
cell performance by blocking carrier flows and providing shunting paths for current flow across the p-n junction.

Growth of a multicrystalline slab of silicon. The animation plays automatically. It is captured from an animation format
that is not compatible with modern browsers.

It used to be thought that large grain crystals were the most suitable for multicrystalline silicon solar cells since larger crystals meant
fewer grain boundaries. However, in recent years it was found that smaller grains gave lower stress at the ground boundaries so they
were less electrically active (lower recombination). Presently, most multicystalline silicon for solar cells is grown using a process
where the growth is seeded to produce smaller grains and referred to as "high performance multi"1
Slab of multicrystalline silicon after growth. The slab is further cut up into bricks and then the bricks are sliced into wafers.

To avoid significant recombination losses at grain boundaries, grain sizes on the order of at least a few millimeters are required 2. This
also allows single grains to extend from front to back of the cell, providing less resistance to carrier flow and generally decreasing the
length of grain boundaries per unit of cell. Such multicrystalline material is widely used for commercial solar cell production.

At the boundary between two crystal grains, the bonds are strained, degrading the electronic properties.
A 10 x 10 cm2 multicrystalline wafer. The wafer has been textured so that grains of different orientation show up as light and dark.

1. Y. M. Yang, Yu, A., Hsu, B., Hsu, W. C., Yang, A., and Lan, C. W., “Development of high-performance multicrystalline
silicon for photovoltaic industry”, Progress in Photovoltaics: Research and Applications, vol. 23, no. 3, pp. 340 - 351, 2015.
2. H. C. Card and Yang, E. S., “Electronic processes at grain boundaries in polycrystalline semiconductors under optical
illumination”, IEEE Transactions on Electron Devices, vol. ED-24, pp. 397-402, 1977.

Wafer Slicing
Add new comment
Español

Once an ingot has been grown it is then sliced up into wafers. In the case of the multicrystalline silicon, large slabs are grown which
are then sliced up into smaller ingot blocks.
Large multicrystalline silicon block being sliced up into smaller bricks. The smaller bricks are then cut up into wafers with a wire saw.

Brick of multicrystalline silicon cut from slab and before being cut up into wafers.
Silicon brick being sliced up into wafers. Moving a mouse over the picture shows the end of the process.

Other Wafering Techniques

Add new comment
1 comment(s)
Español

The process of growing large silicon ingots and then slicing them up into wafers continues to dominate the photovoltaic market.
However there are many other processes that try to grow wafers from the outset and so avoiding the cutting process.

Edge Defined Film Fed Growth (EFG)

The edge defined film fed growth technique uses a die to define thickness of a sheet of silicon as shown in the figure below. Careful
adjustment of the temperature profile of the graphite die causes the sheets of silicon to crystallize with large grains.

In EFG the growth is defined by the gap between two graphite dies.

The simplest implementation of the technique produces a large cylinder of material. Traditionally these cylinders were large polygons
but recent experiments have included the use of large cylinders. The large cylinders produce curved material that is then flattened out
during the cell stringing and encapsulation process.

Large crystal cylinders grown using edge defined film fed growth (EFG) process

Solid State Diffusion

Add new comment
Español

Solid state diffusion is a straight forward process and the typical method for introducing dopant atoms into semiconductors. In silicon
solar cell processing starting substrates are typically uniformly doped with boron giving a p-type base. The n-type emitter layer is
formed through phosphorous doping (see Doping).

Calculation of Diffusion Profiles (Ghandi1)

In its simplest form the diffusion process follows Fick's law:

where j is the flux density (atoms cm-2), D is the diffusion coefficient (cm2 s-1), N is the concentration volume (atoms cm-3 ) and x is
the distance (cm).

The profiles can then be calculated for specific cases. Typical cases are an unlimited source such as heating a wafer in the presence of a
phosphorus saturated carrier gas and then turning off the source and driving in the phosphorous atoms on the surface.

Diffusion from an Unlimited Source

Diffusions from an unlimited source commonly produce a shallow junction with a very high surface concentration of phosphorus
atoms. The diffusion is described by the complementary error function.
 Solid state diffusion. Heating the wafer at a high temperature in an atmosphere containing dopant atoms causes some of
the atoms to be incorporated into the top surface of the wafer.

where N0 is the impurity concentration at the surface (atoms cm-3 ), D is the diffusivity (cm2 s-1 ), x is the depth (cm)and t is the time
(sec). A simple one-step diffusion is useful where there is no surface passivation of the device.

Diffusion from a limited source

Diffusions often consist of a two step process: a short pre-deposition as outlined above, followed by a longer drive in at a higher
temperature to provide a deep lightly doped emitter. A simplified analysis of the drive-in assumes that it is at a higher temperature and
that the dopant atoms incorporated in the pre-deposition simply redistribute. The final profile is a Gaussian and is described by:

Diffusion from limited source

Edit
Delete
Clone

Read more about Diffusion from limited source

Second order effects cause deviations from the simple models 2 and computer simulations are employed.
 Deposition Temperature: 900 °C

700 1000

Drivein Temperature: 1000 °C

700 1200

Pre-Deposition time : 10 min

0 3

Drive in time : 50 min

0 50

Doping profiles resulting from a phosphorous pre deposition step plus a high temperature drive in. The calculations assume that the
drive in temperature is greater than the pre deposition temperature. The calculations are approximate and do not include second-order
effects such as the "kink".3

1. S. K. Ghandhi, {VLSI} Fabrication Principles: Silicon and Gallium Arsenide, 2nd Edition, 2nd ed. Wiley-Interscience, 1994.
2. C. Hu and White, R. M., Solar Cells: From Basic to Advanced Systems. New York: McGraw-Hill, 1983.
3. J. C. C. Tsai, “Shallow phosphorus diffusion profiles in silicon”, Proceedings of the IEEE, vol. 57, no. 9, pp. 1499 - 1506,
1969.

Screen Printed Solar Cells

Add new comment
Español

Screen-printed solar cells were first developed in the 1970's. As such, they are the best established, most mature solar cell fabrication
technology, and screen-printed solar cells currently dominate the market for terrestrial photovoltaic modules. The key advantage of
screen-printing is the relative simplicity of the process.

There are a variety of processes for manufacturing screen-printed solar cells. The production technique given in the animation below is
one of the simplest techniques and has since been improved upon by many manufacturers and research laboratories.
Animation showing a basic technique for fabricating screen printed solar cells.

There are many variations to the scheme shown above which give higher efficiencies, lower costs or both. Some techniques have
already been introduced into commercial production while others are making progress from the labs to the production lines.

1. Phosphorous Diffusion
Screen-printed solar cells typically use a simple homogeneous diffusion to form the emitter where the doping is the same
beneath the metal contacts and between the fingers. To maintain low contact resistance, a high surface concentration of
phosphorous is required below the screen-printed contact. However, the high surface concentration of phosphorous produces a
"dead layer" that reduces the cell blue response. Newer cell designs can contact shallower emitters, thus improving the cell blue
response. Selective emitters with higher doping below the metal contacts have also been proposed 1, 2 - but none have yet been
introduced into commercial production.
2. Surface Texturing to Reduce Reflection
Wafers cut from a single crystal of silicon (monocrystalline material) are easily textured to reduce reflection by etching pyramids
on the wafer surface with a chemical solution. While such etching is ideal for monocrystalline CZ wafers, it relies on the correct
crystal orientation, and so is only marginally effective on the randomly orientated grains of multicrystalline material. Various
schemes have been proposed to texture multicrystalline materials by using one of the following processes:

1. mechanical texturing of the wafer surface with cutting tools or lasers 3, 4, 5;

2. isotropic chemical etching based on defects rather than crystal orientation 6;
3. isotropic chemical etching in combination with a photolithographic mask 7, 8;
4. plasma etching 9.
3. Antireflection Coatings and Fire Through Contacts
Antireflection coatings are particularly beneficial for multicrystalline material that cannot be easily textured. Two common
antireflection coatings are titanium dioxide (TiO2) and silicon nitride (SiNx). The coatings are applied through simple techniques
like spraying or chemical vapour deposition. In addition to the optical benefits, dielectric coatings can also improve the electrical
properties of the cell by surface passivation. By screen-printing over the antireflection coating with a paste containing cutting
agents, the metal contacts can fire though the antireflection coating and bond to the underlying silicon. This process is very
simple and has the added advantage of contacting shallower emitters 10.
4. Edge Isolation
There are various techniques for edge isolation such as plasma etching, laser cutting, or masking the border to prevent a
diffusion from occurring around the edge in the first place.
5. Rear Contact
A full aluminium layer printed on the rear on the cell, with subsequent alloying through firing, produces a back surface field
(BSF) and improves the cell bulk through gettering. However, the aluminium is expensive and a second print of Al/Ag is
required for solderable contact. In most production, the rear contact is simply made using an Al/Ag grid printed in a single step.
6. Substrate
Screen-printing has been used on a variety of substrates. The simplicity of the sequence makes screen-printing ideal for poorer
quality substrates such as multicrystalline material as well as CZ. The general trend is to move to larger size substrates - up to 15
x 15 cm2 for multicrystalline materials and wafers as thin as 200 µm.
Close up of a screen used for printing the front contact of a solar cell. During printing, metal paste is forced through the wire mesh in
unmasked areas. The size of the wire mesh determines the minimum width of the fingers. Finger widths are typically 100 to 200 µm.

Close up of a finished screen-printed solar cell. The fingers have a spacing of approximately 3 mm. An extra metal contact strip is
soldered to the busbar during encapsulation to lower the cell series resistance.
Front view of a completed screen-printed solar cell. As the cell is manufactured from a multicrystalline substrate, the different grain
orientations can be clearly seen. The square shape of a multicrystalline substrate simplifies the packing of cells into a module.

Rear view of a finished screen-printed solar cell. The cell can either have a grid from a single print of Al/Ag paste with no BSF, or a
coverage of aluminium that gives a BSF but requires a second print for solderable contacts. Click on the image to switch between the
two views.

1. J. Horzel, Szlufcik, J., Nijs, J., and Mertens, R., “A simple processing sequence for selective emitters”, Twenty Sixth IEEE
Photovoltaic Specialists Conference. New York, NY, USA, pp. 139-142, 1997.
2. D. S. Ruby, Yang, P., Zaidi, S., Brueck, S., Roy, M., and Narayanan, S., “Improved Performance of Self-Aligned, Selective-
Emitter Silicon Solar Cells”, 2nd World Conference and Exhibition on Photovoltaic Solar Energy Conversion. Vienna, Austria,
1998.
3. S. Narayanan et al., “18% efficient polycrystalline silicon solar cells”, Twenty First IEEE Photovoltaic Specialists Conference,
vol. 1. pp. 678-680, 1990.
4. G. Willeke, Nussbaumer, H., Bender, H., and Bucher, E., “A simple and effective light trapping technique for polycrystalline
silicon solar cells”, Solar Energy Materials and Solar Cells, vol. 26, pp. 345 - 356, 1992.
5. R. Hezel, “Recent progress in MIS solar cells”, Progress in Photovoltaics: Research and Applications, vol. 5, pp. 109-120,
1997.
6. R. Einhaus, Vazsonyi, E., Szlufcik, J., Nijs, J., and Mertens, R., “Isotropic texturing of multicrystalline silicon wafers with
acidic texturing solutions”, Twenty Sixth IEEE Photovoltaic Specialists Conference. New York, NY, USA, pp. 167-170, 1451,
1997.
7. M. J. Stocks, Carr, A. J., and Blakers, A. W., “Texturing of polycrystalline silicon”, Solar Energy Materials and Solar Cells,
 vol. 40, pp. 33 - 42, 1996.
8. J. Zhao, Wang, A., and Green, M. A., “19.8% Efficient Multicrystalline Silicon Solar Cells with Honeycomb Textured Front
Surface”, 2nd World Conference and Exhibition on Photovoltaic Solar Energy Conversion. Vienna, Austria, 1998.
9. K. Fukui, Inomata, Y., and Shirasawa, K., “Surface texturing using reactive ion etching for multicrystalline silicon solar cells”,
Twenty Sixth IEEE Photovoltaic Specialists Conference. New York, NY, USA, pp. 1451, 47-50, 1997.
10. J. Szlufcik, Sivoththaman, S., Nlis, J. F., Mertens, R. P., and Van-Overstraeten, R., “Low-cost industrial technologies of
crystalline silicon solar cells”, Proceedings-of-the-IEEE, vol. 85. pp. 711-730, 1997.

Buried Contact Solar Cells

Add new comment
Español

The buried contact solar cell is a high efficiency commercial solar cell technology based on a plated metal contact inside a laser-
formed groove. The buried contact technology overcomes many of the disadvantages associated with screen-printed contacts and this
allows buried contact solar cell to have performance up to 25% better than commercial screen-printed solar cells. A schematic of a
buried contact solar cell is shown in the figure below.

Cross-section of Laser Grooved, Buried Contact Solar Cell.

A key high efficiency feature of the buried contact solar cell is that the metal is buried in a laser-formed groove inside the silicon solar
cell. This allows for a large metal height-to-width aspect ratio. A large metal contact aspect ratio in turn allows a large volume of metal
to be used in the contact finger, without having a wide strip of metal on the top surface. Therefore, a high metal aspect ratio allows a
large number of closely spaced metal fingers, while still retaining a high transparency. For example, on a large area device, a screen
printed solar cell may have shading losses as high as 10 to 15%, while in a buried contact structure, the shading losses will only be 2 to
3%. These lower shading losses allow low reflection and therefore higher short-circuit currents.
Cross section of a partially plated laser groove.

In addition to good reflection properties, the buried contact technology also allows low parasitic resistance losses due to its high metal
aspect ratio, its fine finger spacing and its plated metal for the contacts. As shown in the Emitter Resistance page, the emitter resistance
is reduced in a buried contact solar cell since a narrower finger spacing dramatically reduces the emitter resistance losses. The metal
grid resistance is also low since the finger resistance is reduced by the large volume of metal in the grooves and by the use of copper,
which has a lower resistivity than the metal paste used in screen printing. As well, the contact resistance of a buried contact solar cell is
lower than that in screen printed solar cells due to the formation of a nickel silicide at the semiconductor-metal interface and the large
metal-silicon contact area. Overall, these reduced resistive losses allow large area solar cells with high FFs.

When compared to a screen-printed cell, the metalization scheme of a buried contact solar cell also improves the cell's emitter. To
minimise resistive losses, the emitter region of a screen-printed solar cell is very heavily doped and results in a "dead" layer at the
surface of the solar cell. Since emitter losses are low in a buried contact structure, the emitter doping can be optimized for high open-
circuit voltages and short-circuit currents. Furthermore, a buried contact structure includes a self-aligned, selective emitter, which
thereby reduces the contact recombination and also contributes to high open-circuit voltages.

The efficiency advantages of buried contact technology provide significant cost and performance benefits. In terms of $/W, the cost of
a buried contact solar cell is the same as a screen-printed solar cell 1. However, due to the inclusion of certain area-related costs as well
as fixed costs in a PV system, a higher efficiency solar cell technology results in lower cost electricity. An additional advantage of
buried contact technology is that it can be used for concentrator systems 2.

The production sequence for laser grooved buried contact solar cells is shown in the animation below.
Animation showing the manufacture of a buried contact solar cell. The cell thickness is greatly enlarged for clarity

1. D. Jordan and Nagle, J. P., “Buried contact concentrator solar cells”, Progress in Photovoltaics: Research and Applications,
vol. 2, pp. 171-176, 1994.
2. J. H. Wohlgemuth and Narayanan, S., “Buried contact concentrator solar cells”, Twenty Second IEEE Photovoltaic Specialists
Conference, vol. 1. pp. 273-277, 1991.

High Efficiency Solar Cells

Add new comment
Español

High efficiency cells cost considerably more to produce than standard silicon cells and are typically used in solar cars or space
applications.

Honda dream, the winning car in the 1996 World Solar Challenge. The custom made cells for the car are greater than 20% efficient.
(Photograph PVSRC)

Some of the techniques and design features used in the laboratory fabrication of silicon solar cells, to produce the highest possible
efficiencies include:

1. lightly phosphorus diffused emitters, to minimise recombination losses and avoid the existence of a "dead layer" at the cell
surface;
2. closely spaced metal lines, to minimise emitter lateral resistive power losses;
3. very fine metal lines, typically less than 20 µm wide, to minimise shading losses;
4. polished or lapped surfaces to allow top metal grid patterning via photolithography;
5. small area devices and good metal conductivities, to minimise resistive losses in the metal grid;
6. low metal contact areas and heavy doping at the surface of the silicon beneath the metal contact to minimise recombination;
7. use of elaborate metallization schemes, such as titanium/palladium/silver, that give very low contact resistances;
8. good rear surface passivation, to reduce recombination;
9. use of anti-reflection coatings, which can reduce surface reflection from 30% to well below 10%.

A number of cell designs exist which incorporate advanced laboratory features. Two approaches that have been used by niche markets
such as solar cars are the PERL cells produced at University of New South Wales, and the rear-contact cells developed at Stanford
University and SunPower.

PERL Solar Cells

The passivated emitter with rear locally diffused (PERL) cell uses micro-electronic techniques to produce cells with efficiencies
approaching 25% under the standard AM1.5 spectrum. The passivated emitter refers to the high quality oxide at the front surface that
significantly lowers the number of carriers recombining at the surface. The rear is locally diffused only at the metal contacts to
minimise recombination at the rear while maintaining good electrical contact.

Schematic of high efficiency laboratory cell.

Electron microscope image of the top surface of a PERL Cell showing a broken electroplated finger. The total finger width is less than
20 µm and the contact width is 3 µ>m.

While expensive to fabricate, it is possible to produce cells with very high efficiencies. A cell for a solar car had the following
characteristics:

Area: 22 cm2
Efficiency: 23.5%
Voc: 703 mV
Isc: 914 mA
Jsc: 41.3 mA
Vmp: 600 mV
FF: 0.81
Imp: 868 mA
IV curve for a solar car cell.

Rear Contact Solar Cells

Add new comment
Español

Rear contact solar cells achieve potentially higher efficiency by moving all or part of the front contact grids to the rear of the device.
The higher efficiency potentially results from the reduced shading on the front of the cell and is especially useful in high current cells
such as concentrators or large areas. There are several configurations.

Interdigitated back contact solar cells (IBC)

Rear contact solar cells eliminate shading losses altogether by putting both contacts on the rear of the cell. By using a thin solar cell
made from high quality material, electron-hole pairs generated by light that is absorbed at the front surface can still be collected at the
rear of the cell 1. Such cells are especially useful in concentrator applications where the effect of cell series resistance is greater.

An additional benefit is that cells with both contacts on the rear are easier to interconnect and can be placed closer together in the
module since there is no need for a space between the cells.

Back Contact Solar Cell as used in commercial production.

1. P. J. Verlinden, Swanson, R. M., and Crane, R. A., “7000 High Efficiency Cells for a Dream”, Progress in Photovoltaics:
Research and Applications, vol. 2, pp. 143 - 152, 1994.

Source Material
Add new comment

In the past it was common for the source material for solar cell production to be off cuts from the integrated circuit industry. The
source material has to be checked for impurities and to make sure that it has the right dopant type and resistivity. Almost all cells are
presently made on p-type substrates. Notable exceptions are the rear contact cells made by Sunpower and amorphous/crystalline silicon
heterojunction cells manufactured by Sanyo. The solar cell industry has grown very rapidly in the past few years and now uses more
silicon than the entire integrated circuit industry. The rapid expansion has pushed up the price of raw crystalline silicon feed stock from
under $20/kg to over $200/kg.

Silicon scrap comes from a wide variety of sources. One source is off cuts and scrap material from the semiconductor industry.

Wafer scraps from the production line are recycled for growing new ingots

Growing Ingots
Add new comment

The ingot growth for Multicrystalline silicon is quite simple, melt the silicon in a large crucible and let it cool slowly to form a large
crystal. The specifics of furnace design allows the ingot to cool slowly so that very large grains (> 1 cm) are formed.
Tub used for growing silicon. The dimensions are about 50 cm x 50 cm x 25 cm deep. The tub has to withstand the melting point of
silicon at 1415 °C. For comparison iron melts at 1538 °C.

Crystal growing furnaces. The system is loaded and unloaded using the hoists at the bottom.
Loading the growth tub into the furnace.
Finished silicon ingot.

Sawing the Ingot into Bricks

Add new comment

The large silicon ingot is sawn into more manageable bricks.

The individual bricks are now ready to saw up into wafers

Wafer Slicing
Add new comment

Slicing up the bricks into wafers is a delicate operation. Each wafer is up to 15 x 15 cm2 and under a third of a mm (300 µm) thick.
Modern solar cell factories use wire saws rather than the internal diameter blade saws previously used for the semiconductor industry.
In fact, the semiconductor industry is now moving to the wire saw due to their superior technology.
A large industrial wire saw. The wire sawing is done on the right hand side but the bricks are not loaded yet. The left hand side holds
the wire and the control circuitry.

Close up of the wires. The wires are covered with slurry and are wound around the drums. When running the drums spin at high speed
and the silicon bricks are pushed down from the top. Move the mouse over the picture for a closer view. The feed wire is just visible
coming in the top left of the spool.
Close up of the wire saw with the bricks loaded and ready to go.

Texturing
Add new comment

Cutting silicon into wafers leaves the surface covered with cutting slurry and the surface is damaged due to the action of the saw.
Wafers are cleaned in a hot solution of sodium hydroxide that removes the surface contamination and the first 10 µm of damaged
silicon. The wafers are then textured in a more dilute solution of sodium hydroxide with isopropanol as a wetting agent. For
multicrystalline wafers, acidic texturing is often used as it gives a more uniform etch rate across gain boundaries.

0:00 / 0:23

After the wafers are cut into slices they are then put into cassettes for saw damage removal and texturing
.

The cassettes are further loaded into holders for moving through the production line. During the texturing process hydrogen is released
and bubbles adhere to the wafer. The mesh place on top of the cassettes stops the wafers from floating out of the cassettes.

0:00 / 0:08

Wafers during the texturing process. Normally there is a protective blind in place to prevent sodium hydroxide splashes.
 0:00 / 0:25

The protective blind rolls back and the wafers are removed from the texturing bath. Wafers are put through several rinse cycles and an
acid rinse to neutralise the sodium hydroxide.

0:00 / 0:26

Finally the wafers are loaded into a centrifuge for a final rinse and spin dry.

Emitter Diffusion
Add new comment

The emitter diffusion process is performed in a variety of ways. In this case a phosphorous containing coating is applied to the surface.
The wafers are then put in a belt furnace to diffuse a small amount of phosphorous into the silicon surface.
 0:00 / 1:13

Loading the phosphorous coating. The trend in production lines is to have as much automation as possible and to move the wafers
through the lines on a belt. The loading and unloading of cassettes is a major source of yield loss.

0:00 / 0:13

Wafers moving from the diffusion coating furnace to the high temperature diffusion furnace. The diffusion coating is clear so the
wafers do not look markedly different from the previous video
 0:00 / 0:10

The wafers travel through the diffusion furnace for roughly an hour.

0:00 / 0:28

Finally the wafers are loaded into a centrifuge for a final rinse and spin dry.
 0:00 / 0:25

After the diffusion process the wafers are loaded back into cassettes for an acid etch to remove the diffusion glass.

Edge Isolation
Add new comment

The edge isolation process removes the phosphorous diffusion around the edge of the cell so that the front emitter is electrically
isolated from the cell rear. A common way to achieve this is to stack the wafers on top of each other then plasma etch using CF4 and
O2

0:00 / 0:48

Loading the plasma etching system and then etching the wafers

Anti Reflection Coatings

Add new comment

An antireflection of silicon nitride is typically deposited using chemical vapour deposition process (CVD). Precursor gases of silane
(SiH4) and ammonia (NH3) are fed into a chamber and break down due to temperature (LPCVD) or due to a plasma enhancement
(PECVD). Other systems use microwaves to cause the silane/ammonia reaction to take place. The complete reaction is:

3SiH4 + 4NH3 -> Si3N4 + 12H2

but the usual reaction to produce a non-stoichiometric film with the incorporation of large amounts of hydrogen (SixNy:H).

Older cell designs use titanium dioxide (TiO2), which provides a good antireflection coating and is simpler to apply but does not
provide surface or bulk passivation.

Wafers being deposited with silicon nitride antireflection coating giving a blue color.

0:00 / 2:24

A full view of the anti-reflection coating machine from loading to unloading

Screen Print Front

 Add new comment

Silver paste is forced through a patterned screen. Those areas with gaps in the pattern leaves a metal pattern on the surface.

0:00 / 0:27

Screen printer in operation. The cells move along a conveyor belt. Here they enter on the right and exit on the left.

0:00 / 0:37

Close up of the screen printing operation. The cells are undernearth the printer so are not visible from the top.
 0:00 / 0:32

After printing the paste is still wet and is easily smudged. Here the wafers are loaded into a drier to evaporate off the organic binders in
the paste. Driers operate at a low temperature of around 200 °C.

0:00 / 0:33

After the wafers are dried they are loaded back onto the conveyor belt to continue their movement in through the production line.
 0:00 / 0:16

Zooming in on a cell after the front screen print is finished. At this stage, the silver still exists as a powder resting on the cell. A later
firing process at high temperature bonds the silver to the silicon.

Screen Print Rear Aluminium

Add new comment

The rear is printed in two parts. A thick layer of aluminium paste covers most of the cell and provides a back surface field (BSF).
Silver strips are also printed on the cell for later soldering to the interconnect metal tabs.

0:00 / 0:07

After the front print is complete the cells are flipped over to print on the rear.
 0:00 / 0:37

Printing the aluminium paste on the rear. The cell is just visible through the screen as it moves through the line. The aluminium
printing is an expensive step since it is very thick and covers almost the whole cell.

A cell immediately after printing. Small gaps are left in the aluminium to print silver strips later in the process. Note how wet the paste
is.
 0:00 / 0:25

Drying the aluminium paste. Careful handling is essential since aluminium is a p-type dopant and any paste smudges that get on the
front of the cell will shunt the cell completely.

Screen Print Rear Silver

Add new comment

0:00 / 0:14

Transferring wafers to the final screen printer. Such automated transfer machinery greatly improves yield - a factor almost as important
as efficiency in a commercial process.
 0:00 / 0:25

Screen printing the silver on the rear

0:00 / 0:19

Unloading the final dry process.

Firing
Add new comment
 0:00 / 0:33

Firing the metal loading and unloading

0:00 / 0:23
 0:00 / 0:16

Overall view of the SP line

Testing
Add new comment

0:00 / 0:17

after the cells are tested they are sorted into bins

Module
Add new comment

Individual solar cells are protected from the weather by encapsulating into a module. Each cell is only around 0.5 volts, to obtain
sufficient voltage the cell are connected together in series using flat wires called tabs.
After the cell is finished it strips are added

Modules and Arrays

Add new comment

7. Modules and Arrays

Introduction
Add new comment

A PV module consists of individual solar cells electrically connected together to increase their power output. They are packaged so that
they are protected from the environment and so that the user is protected from electrical shock. However, several aspects of PV module
design which may reduce either the power output of the module or its lifetime need to be identified. The following chapter will
examine how solar cells are encapsulated into PV modules and examines some of the issues which arise as a result of interconnection
and encapsulation. The most important effects in PV modules or arrays are:

1. losses due to the interconnection of mismatched solar cells;

2. the temperature of the module; and
3. failure modes of PV modules.

Module Structure
Add new comment
3 comment(s)

A PV module consists of a number of interconnected solar cells encapsulated into a single, long-lasting, stable unit. The key purpose of
encapsulating a set of electrically connected solar cells is to protect them and their interconnecting wires from the typically harsh
environment in which they are used. For example, solar cells, since they are relatively thin, are prone to mechanical damage unless
protected. In addition, the metal grid on the top surface of the solar cell and the wires interconnecting the individual solar cells may be
corroded by water or water vapor. The two key functions of encapsulation are to prevent mechanical damage to the solar cells and to
prevent water or water vapor from corroding the electrical contacts.
Many different types of PV modules exist and the module structure is often
different for different types of solar cells or for different applications. For
example, amorphous silicon solar cells are often encapsulated into a flexible
array, while bulk silicon solar cells for remote power applications are usually
rigid with glass front surfaces.

The most common modules have either 60 cells or 72 cells with three bypass
diodes. 60 cell modules were originally designed for ease of handling in
residential applications and heavier 72 cell modules for large utility
installations where cranes and hydraulic lift are available. However, it is quite
possible to use 72 cell modules in residential installations so long as the rest of
the system is designed to handle the large size.

Module lifetimes and warranties on bulk silicon PV modules are over 20

years, indicating the robustness of an encapsulated PV module. A typical
warranty will guarantee that the module produces 90% of its rated output for
the first 10 years and 80% of its rated output up to 25 years. A third party
reinsurance company ensures these warranties are valid in the event the
manufacturer goes bankrupt.

Module Materials
Add new comment
2 comment(s)

Most PV bulk silicon PV modules consist of a transparent top surface, an

encapsulant, a rear layer and a frame around the outer edge. In most modules, A PV array for a remote power system usually
the top surface is glass, the encapsulant is EVA (ethyl vinyl acetate) and the consists of multiple PV modules.
rear layer is Tedlar, as shown below.

A typical bulk silicon PV module used in outdoor remote power applications.

Typical bulk silicon module materials.

Front Surface Materials

The front surface of a PV module must have a high transmission in the wavelengths which can be used by the solar cells in the PV
module. For silicon solar cells, the top surface must have high transmission of light in the wavelength range of 350 nm to 1200 nm. In
addition, the reflection from the front surface should be low. While theoretically this reflection could be reduced by applying an anti-
reflection coating to the top surface, in practice these coatings are not robust enough to withstand the conditions in which most PV
systems are used. An alternative technique to reduce reflection is to "roughen" or texture the surface. However, in this case the dust and
dirt is more likely to attach itself to the top surface, and less likely to be dislodged by wind or rain. These modules are not therefore
"self-cleaning", and the advantages of reduced reflection are quickly outweighed by losses incurred due to increased top surface
soiling.

In addition to its reflection and transmission properties, the top surface material should be impervious to water, should have good
impact resistance, should be stable under prolonged UV exposure and should have a low thermal resistivity. Water or water vapor
ingress into a PV module will corrode the metal contacts and interconnects, and consequently will dramatically reduce the lifetime of
the PV module. In most modules the front surface is used to provide the mechanical strength and rigidity, therefore either the top
surface or the rear surface must be mechanically rigid in order to support the solar cells and the wiring.

There are several choices for a top surface material including acrylic, polymers and glass. Tempered, low iron-content glass is most
commonly used as it is low cost, strong, stable, highly transparent, impervious to water and gases and has good self-cleaning
properties.

Encapsulant

An encapsulant is used to provide adhesion between the solar cells, the top surface and the rear surface of the PV module. The
encapsulant should be stable at elevated temperatures and high UV exposure. It should also be optically transparent and should have a
low thermal resistance. EVA (ethyl vinyl acetate) is the most commonly used encapsulant material. EVA comes in thin sheets which
are inserted between the solar cells and the top surface and the rear surface. This sandwich is then heated to 150 °C to polymerize the
EVA and bond the module together.

Rear Surface

The key characteristics of the rear surface of the PV module are that it must have low thermal resistance and that it must prevent the
ingress of water or water vapour. In most modules, a thin polymer sheet, typically Tedlar, is used as the rear surface. Some PV
modules, known as bifacial modules are designed to accept light from either the front or the rear of the solar cell. In bifacial modules
both the front and the rear must be optically transparent.

Frame

A final structural component of the module is the edging or framing of the module. A conventional PV module frame is typically made
of aluminium. The frame structure should be free of projections which could result in the lodgement of water, dust or other matter.
Several types of silicon PV modules.

Packing Density
Add new comment

The packing density of solar cells in a PV module refers to the area of the module that is covered with solar cells compared to that
which is blank. The packing density affects the output power of the module as well as its operating temperature. The packing density
depends on the shape of the solar cells used. For example, single crystalline solar cells are round or semi-square, while multicrystalline
silicon wafers are usually square. Therefore, if single-crystalline solar cells are not cut squarely, the packing density of a single
crystalline module will be lower than that of a multicrystalline module. The relative packing density possible with round verses square
cells is illustrated below.

The packing density of round and square cells. The round ingots of Cz material give a low packing density so the edges are cut off to
produce semi-square cells and higher packing density. Multicrystalline material is cast in square blocks, giving a high packing density.

Sparsely packed cells in a module with a white rear surface can also provide marginal increases in output via the "zero depth
concentrator" effect 1, illustrated below. Some of the light striking regions of the module between cells and cell contacts is scattered
and channelled to active regions of the module.
The "zero-depth concentration effect" in modules with sparsely packed cells and a white rear surface.

1. SERI, Photovoltaics for Residential Applications. Golden, Colorado: Solar Energy Research Institute, 1984.

Module Circuit Design

Add new comment
2 comment(s)

A bulk silicon PV module consists of multiple individual solar cells connected, nearly always in series, to increase the power and
voltage above that from a single solar cell. The voltage of a PV module is usually chosen to be compatible with a 12V battery. An
individual silicon solar cell has a voltage at the maximum power point around 0.5V under 25 °C and AM1.5 illumination. Taking into
account an expected reduction in PV module voltage due to temperature and the fact that a battery may require voltages of 15V or
more to charge, most modules contain 36 solar cells in series. This gives an open-circuit voltage of about 21V under standard test
conditions, and an operating voltage at maximum power and operating temperature of about 17 or 18V. The remaining excess voltage
is included to account for voltage drops caused by other elements of the PV system, including operation away from maximum power
point and reductions in light intensity.

In a typical module, 36 cells are connected in series to produce a voltage sufficient to charge a 12V battery.

The voltage from the PV module is determined by the number of solar cells and the current from the module depends primarily on the
size of the solar cells. At AM1.5 and under optimum tilt conditions, the current density from a commercial solar cell is approximately
between 30 mA/cm2 to 36 mA/cm2. Single crystal solar cells are often 15.6 × 15.6 cm2, giving a total current of almost 9 – 10A from a
module.

The table below shows the output of typical modules at STC. IMP and ISC do not change that much but VMP and VOC scale with the
number of cells in the module.

Cells PMAX VMPP IMPP VOC ISC Efficiency

72 340 Wp 37.9 V 8.97 A 47.3 V 9.35 A 17.5%

60 280 Wp 31.4 V 8.91 A 39.3 V 9.38 A 17.1%

36 170 Wp 19.2 V 8.85 A 23.4 V 9.35 A 17%

If all the solar cells in a module have identical electrical characteristics, and they all experience the same insolation and temperature,
then all the cells will be operating at exactly the same current and voltage. In this case, the IV curve of the PV module has the same
shape as that of the individual cells, except that the voltage and current are increased. The equation for the circuit becomes:

where:
N is the number of cells in series;
M is the number of cells in parallel;
IT is the total current from the circuit;
 Modules for residential or large fields usually contain either 60 or 72 cells. There are other sizes such as 96 cell modules
but they are much less common.

V
T is the total voltage from the circuit;
I0 is the saturation current from a single solar cell;
IL is the short-circuit current from a single solar cell;
n is the ideality factor of a single solar cell;
and q, k, and T are constants as given in the constants page.

The overall IV curve of a set of identical connected solar cells is shown below. The total current is simply the current of an individual
cell multiplied by the number of cells in parallel. Such that: ISC total = ISC × M. The total voltage is the voltage of an individual cell
multiplied but the number of cells in series. Such that:

$$I_{SC}(total) = I_{SC}(cell)\times M$$

$$I_{MP}(total) = I_{MP}(cell)\times M$$

$$V_{OC}(total) = V_{OC}(cell)\times N$$

$$V_{MP}(total) = V_{MP}(cell)\times N$$

If the cells are identical then the fill factor does not change when the cells are in parallel or series. However, there is usually mismatch
in the cells so the fill factor is lower when the cells are combined. The cell mismatch may come from manufacturing or from
differences in light on the cells where one cell has more light than another.

I-V curve for N cells in series x M cells in parallel.

 Cells in series: 2 units

1 30

Cells in parallel: 2 units

1 5

Mismatch Effects
Add new comment

Mismatch losses are caused by the interconnection of solar cells or modules which do not have identical properties or which
experience different conditions from one another. Mismatch losses are a serious problem in PV modules and arrays under some
conditions because the output of the entire PV module under worst case conditions is determined by the solar cell with the lowest
output. For example, when one solar cell is shaded while the remainder in the module are not, the power being generated by the "good"
solar cells can be dissipated by the lower performance cell rather than powering the load. This in turn can lead to highly localised
power dissipation and the resultant local heating may cause irreversible damage to the module.
Shading of one region of a module compared to another is a major cause of mismatch in PV modules.

Mismatch in PV modules occurs when the electrical parameters of one solar cell are significantly altered from those of the remaining
devices. The impact and power loss due to mismatch depend on:

the operating point of the PV module;

the circuit configuration; and
the parameter (or parameters) which are different from the remainder of the solar cells.

Differences in any part of the IV curve between one solar cell and another may lead to mismatch losses at some operating point. A
non-ideal IV curve and the operating regime of the solar cell is shown below. Although mismatch may occur in any of the cell
parameters shown below, large mismatches are most commonly caused by differences in either the short-circuit current or open-circuit
voltage. The impact of the mismatch depends on both the circuit configuration and on the type of mismatch, and is demonstrated in
more detail in the following pages.

The comparison of an ideal and a non-ideal solar cell. For mismatch, the greatest difference is when the cell is driven into reverse
voltage bias.

Mismatch for Cells Connected in series

 Add new comment
5 comment(s)

As most PV modules are series-connected, series mismatches are the most common type of mismatch encountered. Of the two simplest
types of mismatch considered (mismatch in short-circuit current or in open-circuit voltage), a mismatch in the short-circuit current is
more common, as it can easily be caused by shading part of the module. This type of mismatch is also the most severe.

For two cells connected in series, the current through the two cells is the same. The total voltage produced is the sum of the individual
cell voltages. Since the current must be the same, a mismatch in current means that the total current from the configuration is equal to
the lowest current.

Open Circuit Voltage Mismatch for Cells Connected in Series

A mismatch in the open-circuit voltage of series-connected cells is a relatively benign form of mismatch. As shown in the animation
below, at short-circuit current, the overall current from the PV module is unaffected. At the maximum power point, the overall power is
reduced because the poor cell is generating less power. As the two cells are connected in series, the current through the two solar cells
is the same, and the overall voltage is found by adding the two voltages at a particular current.

In the animation, cell 2 has a lower output voltage than cell 1.

Short-Circuit Current Mismatch for Cells Connected in Series

A mismatch in the short-circuit current of series connected solar cells can, depending on the operating point of the module and the
degree of mismatch, have a drastic impact on the PV module. As shown in the animation below, at open-circuit voltage, the impact of a
reduced short-circuit current is relatively minor. There is a minor change in the open-circuit voltage due to the logarithmic dependence
of open-circuit voltage on short-circuit current. However, as the current through the two cells must be the same, the overall current
from the combination cannot exceed that of the poor cell. Therefore, the current from the combination cannot exceed the short-circuit
current of the poor cell. At low voltages where this condition is likely to occur, the extra current-generating capability of the good cells
is not dissipated in each individual cell (as would normally occur at short circuit), but instead is dissipated in the poor cell.

Overall, in a series connected configuration with current mismatch, severe power reductions are experienced if the poor cell produces
less current than the maximum power current of the good cells and also if the combination is operated at short circuit or low voltages,
the high power dissipation in the poor cell can cause irreversible damage to the module. These effects are illustrated in the two
animations below.
Current mismatch for two cells in series can be quite serious and quite common. The Isc of the combination is limited to the Isc of the
lowest cell.

An easy method of calculating the combined short-circuit current of series connected mismatched cells. The current at the point of
intersection represents the short-circuit current of the series combination (ie. V1+V2=0).

Shading
Add new comment
3 comment(s)

shading is a problem in PV modules since shading just one cell in the module can reduce the power output to zero

Shading of a Single Cell

The output of a cell declines when shaded by a tree branch, building or module dust. The output declines proportionally to the amount
of shading. For completely opaque objects such as a leaf, the decline in current output of the cell is proportional to the amount of the
cell that is obscured.
Move the grey rectangle over the solar cell. The current is reduced by the amount of shading.

Shading of a Module

The cells in a module are all connected in series. Shading a single cell causes the current in the string of cells to fall to the level of the
shaded cell.

Hot Spot Heating

Add new comment
2 comment(s)

Hot-spot heating occurs when there is one low current solar cell in a string of at least several high short-circuit current solar cells, as
shown in the figure below.

One shaded cell in a string reduces the current through the good cells, causing the good cells to produce higher voltages that can often
reverse bias the bad cell.

If the operating current of the overall series string approaches the short-circuit current of the "bad" cell, the overall current becomes
limited by the bad cell. The extra current produced by the good cells then forward biases the good solar cells. If the series string is
short circuited, then the forward bias across all of these cells reverse biases the shaded cell. Hot-spot heating occurs when a large
number of series connected cells cause a large reverse bias across the shaded cell, leading to large dissipation of power in the poor cell.
Essentially the entire generating capacity of all the good cells is dissipated in the poor cell. The enormous power dissipation occurring
in a small area results in local overheating, or "hot-spots", which in turn leads to destructive effects, such as cell or glass cracking,
melting of solder or degradation of the solar cell.
Heat dissipated in a shaded cell caused the module to crack.

Bypass Diodes
Add new comment
1 comment(s)

The destructive effects of hot-spot heating may be circumvented through the use of a bypass diode. A bypass diode is connected in
parallel, but with opposite polarity, to a solar cell as shown below. Under normal operation, each solar cell will be forward biased and
therefore the bypass diode will be reverse biased and will effectively be an open circuit. However, if a solar cell is reverse biased due
to a mismatch in short-circuit current between several series connected cells, then the bypass diode conducts, thereby allowing the
current from the good solar cells to flow in the external circuit rather than forward biasing each good cell. The maximum reverse bias
across the poor cell is reduced to about a single diode drop, thus limiting the current and preventing hot-spot heating. The operation of
a bypass diode and the effect on an IV curve are shown in the animation below.

Current flow for two cells in series and the effect of a bypass diode. The animation progresses automatically from one condition to
another.

The effect of a bypass diode on an IV curve can be determined by first finding the IV curve of a single solar cell with a bypass diode
and then combining this curve with other solar cell IV curves. The bypass diode affects the solar cell only in reverse bias. If the reverse
bias is greater than the knee voltage of the solar cell, then the diode turns on and conducts current. The combined IV curve is shown in
the figure below.
IV curve of solar cell with bypass diode.
Preventing hot-spot heating with a bypass diode. For clarity, the example uses a total of 10 cells with 9 unshaded and 1 shaded. A
typical module contains 36 cells and the effects of current mismatch are even worse without the bypass diode, but are less important
with the bypass diode. The animation moves automatically. You do not need to click to continue.

In practice, however, one bypass diode per solar cell is generally too expensive and instead bypass diodes are usually placed across
groups of solar cells. The voltage across the shaded or low current solar cell is equal to the forward bias voltage of the other series cells
which share the same bypass diode plus the voltage of the bypass diode. This is shown in the figure below. The voltage across the
unshaded solar cells depends on the degree of shading on the low current cell. For example, if the cell is completely shaded, then the
unshaded solar cells will be forward biased by their short circuit current and the voltage will be about 0.6V. If the poor cell is only
partially shaded, the some of the current from the good cells can flow through the circuit, and the remainder is used to forward bias
each solar cell junction, causing a lower forward bias voltage across each cell. The maximum power dissipation in the shaded cell is
approximately equal to the generating capability of all cells in the group. The maximum group size per diode, without causing damage,
is about 15 cells/bypass diode, for silicon cells. For a normal 36 cell module, therefore, 2 bypass diodes are used to ensure the module
will not be vulnerable to "hot-spot" damage.

Bypass diodes across groups of solar cells. The voltage across the unshaded solar cells depends on the degree of shading of the poor
cell. In the figure above, 0.5V is arbitrarily shown.

Mismatch for Cells Connected in Parallel

Add new comment
1 comment(s)

In small modules, the cells are in placed in series so parallel mismatch is not an issue. Modules are paralleled in large arrays so the
mismatch usually applies at a module level rather than at a cell level. For cells or modules in parallel:

and
Cells connected in parallel. The voltage across the cell combination is always the same and the total current from the combination is
the sum of the currents in the individual cells.

Cell1 VOC, = 0.61 volts

0.6 0.7

Cell2 VOC, = 0.65 volts

0.6 0.7

Voltage mismatch for two cells in parallel. The individual cells are in red and blue. The green curve is the IV curve of the combination.
The VOC of the combination lies between the VOC’s for the individual cells.
An easy method of calculating the combined open circuit voltage (Voc) of mismatched cells in parallel. The curve for one of the cells
is reflected in the voltage axis so that the intersection point (where I1+I2=0) is the Voc of the parallel configuration.

Mismatch Effects in Arrays

Add new comment
1 comment(s)

In a larger PV array, individual PV modules are connected in both series and parallel. A series-connected set of solar cells or modules
is called a "string". The combination of series and parallel connections may lead to several problems in PV arrays. One potential
problem arises from an open-circuit in one of the series strings. The current from the parallel connected string (often called a "block")
will then have a lower current than the remaining blocks in the module. This is electrically identical to the case of one shaded solar cell
in series with several good cells, and the power from the entire block of solar cells is lost. The figure below shows this effect.

Potential mismatch effects in larger PV arrays. Although all modules may be identical and the array does not experience any shading,
mismatch and hot spot effects may still occur.

Parallel connections in combination with mismatch effects may also lead to problems if the by-pass diodes are not rated to handle the
current of the entire parallel connected array. For example, in parallel strings with series connected modules, the by-pass diodes of the
series connected modules become connected in parallel, as shown in the figure below. A mismatch in the series connected modules
will cause current to flow in a by-pass diode, thereby heating this diode. However, heating the by-pass diode reduces the effective
resistance. Most of the current will now flow through the slightly hotter set of by-pass diodes. These by-pass diodes then become even
hotter, further reducing their resistance and increasing the current flow. Eventually almost all the current may flow through one set of
by-pass diodes. If the diodes are not rated to handle the current from the parallel combination of modules, they will burn out and allows
damage to the PV modules to occur.

Bypass diodes in paralleled modules. There are typically two bypass diodes in each 36 cell module.

In addition to the use of by-pass diodes to prevent mismatch losses, an additional diode, called a blocking diode, may be used to
minimize mismatch losses. A blocking diode, shown in the figure below, is typically used to prevent the module from loading the
battery at night by preventing current flow from the battery through the PV array. With parallel connected modules, each string to be
connected in parallel should have its own blocking diode. This not only reduces the required current carrying capability of the blocking
diode, but also prevents current flowing from one parallel string into a lower-current string and therefore helps to minimize mismatch
losses arising in parallel connected arrays.

Impact of blocking diodes in parallel connected modules.

PV Module Temperature
Add new comment

An unwanted side-effect of the encapsulation of solar cells into a PV module is that the encapsulation alters the heat flow into and out
of the PV module, thereby increasing the operating temperature of the PV module. These increases in temperature have a major impact
on the PV module by reducing its voltage, thereby lowering the output power. In addition, increases in temperature are implicated in
several failure or degradation modes of PV modules, as elevated temperatures increase stresses associated with thermal expansion and
also increase degradation rates by a factor of about two for each 10°C increase in temperature.

Thermographic image of sixteen cell module with integral bypass diode cells under reverse bias conditions. Each color change
corresponds to a 4oC change in temperature.

The operating temperature of a module is determined by the equilibrium between the heat produced by the PV module, the heat lost to
the environment and the ambient operating temperature. The heat produced by the module depends on the operating point of the
module, the optical properties of the module and solar cells, and the packing density of the solar cells in the PV module. The heat lost
to the environment can proceed via one of three mechanisms; conduction, convection, and radiation. These loss mechanisms depend on
the thermal resistance of the module materials, the emissive properties of the PV module, and the ambient conditions (particularly wind
speed) in which the module is mounted. These factors are discussed in the following pages.

Heat Generation in PV Modules

Add new comment
1 comment(s)

A PV module exposed to sunlight generates heat as well as electricity. For a typical commercial PV module operating at its maximum
power point, only 10 to 15% of the incident sunlight is converted into electricity, with much of the remainder being converted into
heat. The factors which affect the heating of the module are:

1. the reflection from the top surface of the module;

2. the electrical operating point of the module;
3. absorption of sunlight by the PV module in regions which are not covered by solar cells;
4. absorption of low energy (infrared) light in the module or solar cells; and
5. the packing density of the solar cells.

Front Surface Reflection

Light reflected from the front surface of the module does not contribute to the electrical power generated. Such light is considered an
electrical loss mechanism which needs to be minimized. Neither does reflected light contribute to heating of the PV module. The
maximum temperature rise of the module is therefore calculated as the incident power multiplied by the reflection. For typical PV
modules with a glass top surface, the reflected light contains about 4% of the incident energy.

Operating Point and Efficiency of the Module

The operating point and efficiency of the solar cell determine the fraction of the light absorbed by the solar cell that is converted into
electricity. If the solar cell is operating at short-circuit current or at open-circuit voltage, then it is generating no electricity and hence
all the power absorbed by the solar cell is converted into heat.
Absorption of Light by the PV Module

The amount of light absorbed by the parts of the module other than the solar cells will also
contribute to the heating of the module. How much light is absorbed and how much is
reflected is determined by the color and material of the rear backing layer of the module.

Absorption of Infra-red Light

Light which has an energy below that of the band gap of the solar cells cannot contribute
to electrical power, but if it is absorbed by the solar cells or by the module, this light will
contribute to heating. The aluminium at the rear of the solar cell tends to absorb this
infrared light. In solar cells which do not have full aluminium coverage at the rear of the
solar cell, the infrared may pass through the solar cell and exit from the module.

Packing Factor of the Solar Cells

Solar cells are specifically designed to be efficient absorbers of solar radiation. The cells
will generate significant amounts of heat, usually higher than the module encapsulation
and rear backing layer. Therefore, a higher packing factor of solar cells increases the
generated heat per unit area.

Heat Loss in PV Modules

Add new comment
Sunlight incident on a solar panel
The operating temperature of a PV module is an equilibrium between the heat generated
generates heat as well as electricity.
by the PV module and the heat loss to the surrounding environment. There are three main
mechanisms of heat loss: conduction, convection and radiation.

The module temperature is determined by the equilibrium between heat generated in the PV module by the sun and the conduction,
convection and radiative heat loss from the module.

Heat Conduction

Conductive heat losses are due to thermal gradients between the PV module and other materials (including the surrounding air) with
which the PV module is in contact. The ability of the PV module to transfer heat to its surroundings is characterized by the thermal
resistance and configuration of the materials used to encapsulate the solar cells.

Conductive heat flow is analogous to conductive current flow in an electrical circuit. In conductive heat flow, the temperature
differential is the driving force behind the conductive flow of heat in a material with a given thermal resistance, while in an electric
circuit the voltage differential causes a current flow in a material with a particular electrical resistance. Therefore, the relationship
between temperature and heat (i.e., power) is given by an equation similar to that relating voltage and current across a resistor.
Assuming that a material is uniform and in a steady state, the equation between heat transfer and temperature is given by:

where:
Pheat is the heat (power) generated by the PV module discussed in Heat Generation in PV Modules;
Φ is the thermal resistance of the emitting surface in °C W-1; and
ΔT is the temperature difference between the two materials in °C.

The thermal resistance of the module depends on the thickness of the material and its thermal resistivity (or conductivity). Thermal
resistance is similar to electrical resistance and the equation for thermal resistance is:

where:
A is the area of the surface conducting heat;
l is the length of the material through which heat must travel; and
k is the thermal conductivity in units of W m-1 °C-1.

To find the thermal resistance of a more complicated structure, the individual thermal resistances may be added in series or in parallel.
For example, since both the front and the rear surface conduct heat from the module to the ambient, these two mechanisms operate in
parallel with one another and the thermal resistance of the front and rear accumulate in parallel. Alternatively, in a module, the thermal
resistance of the encapsulant and that of the front glass would add in series. A diagram of the thermal resistance of a simple PV module
neglecting the conductance of the frames and edging is shown above.

Convection

Convective heat transfer arises from the transport of heat away from a surface as the result of one material moving across the surface of
another. In PV modules, convective heat transfer is due to wind blowing across the surface of the module. The heat which is transferred
by this process is given by the equation:

where:
A is the area of contact between the two materials;
h is the convection heat transfer co-efficient in units of W m-2 °C-1; and
ΔT is the temperature difference between the two materials in °C.

Unlike the thermal resistance, h is complicated to calculate directly and is often an experimentally determined parameter for a
particular system and conditions.

Radiation

A final way in which the PV module may transfer heat to the surrounding environment is through radiation. As discussed in the
Blackbody Radiation page, any object will emit radiation based on its temperature. The power density emitted by a blackbody is given
by the equation:

where:
P is the power generated as heat by the PV module;
σ is the Stefan-Boltzmann constant as given in the Constants page; and
T is the temperature of the solar cell in K.

However, a PV module is not an ideal blackbody and to account for non-ideal blackbodies, the blackbody equation is modified by
including a parameter called the emissivity, ε, of the material or object. A blackbody, which is perfect emitter (and absorber) of energy
has an emissivity of 1. An emissivity of an object can often be gauged by its absorption properties, as the two will often be very
similar. For example metals, which tend to have reduced absorption, also have a lower emissivity, usually in the range of 0.03.
Including the emissivity in the equation for emitted power density from a surface gives:

where:
ε is the emissivity of the surface; and
the remainder of the parameters are as above.

The net heat or power lost from the module due to radiation is the difference between the heat emitted from the surroundings to the
module and the heat emitted from the PV module to the surroundings, or in mathematical format:

where:
Tsc is the temperature of the solar cell;
Tamb is the temperature of the ambient surrounding the solar cell; and
the remainder of the parameters are as above.

Nominal Operating Cell Temperature

Add new comment

A PV module will be typically rated at 25 °C under 1 kW/m2. However, when operating in the field, they typically operate at higher
temperatures and at somewhat lower insolation conditions. In order to determine the power output of the solar cell, it is important to
determine the expected operating temperature of the PV module. The Nominal Operating Cell Temperature (NOCT) is defined as the
temperature reached by open circuited cells in a module under the conditions as listed below:

1. Irradiance on cell surface = 800 W/m2

2. Air Temperature = 20°C
3. Wind Velocity = 1 m/s
4. Mounting = open back side.

The equations for solar radiation and temperature difference between the module and air show that both conduction and convective
losses are linear with incident solar insolation for a given wind speed, provided that the thermal resistance and heat transfer coefficient
do not vary strongly with temperature. The NOCT for best case, worst case and average PV modules are shown below. The best case
includes aluminium fins at the rear of the module for cooling which reduces the thermal resistance and increases the surface area for
convection.
Temperature increases, above ambient levels, with increasing solar irradiance for different module types 1.

The best module operated at a NOCT of 33°C, the worst at 58°C and the typical module at 48°C respectively. An approximate
expression for calculating the cell temperature is given by 2:

where:

S = insolation in mW/cm2. Module temperature will be lower than this when wind velocity is high, but higher under still conditions.

Operating Temperature Calculator

Ambient air temperature, Tair = °C

Cell temperature at NOCT = °C Insolation
level, S= mW/cm2
Cell temperature, Tcell = °C

Impact of Module Design on NOCT

Module design, including module materials and packing density, have a major impact on the NOCT. For example, a rear surface with a
lower packing density and reduced thermal resistance may make a temperature difference of 5°C or more.

Impact of Mounting Conditions

Both conductive and convective heat transfer are significantly affected by the mounting conditions of the PV module. A rear surface
which cannot exchange heat with the ambient (i.e., a covered rear surface such as that directly mounted on a roof with no air gap), will
effectively have an infinite rear thermal resistance. Similarly, convection in these conditions is limited to the convection from the front
of the module. Roof integrated mounting thus causes higher operating temperature, often increasing the temperature of the modules by
10°C.
 1. J. R. G. Ross and Smokler, M. I., “Flat-Plate Solar Array Project Final Report”, pp. 86-31, 1986.
2. R. G. Ross, “Flat-Plate Photovoltaic Array Design Optimization”, 14th IEEE Photovoltaic Specialists Conference. San Diego,
CA, pp. 1126-1132, 1980.

Thermal Expansion and Thermal Stresses

Add new comment

Thermal expansion is another important temperature effect which must be taken into account when modules are designed.

Use of stress relief loops to

accommodate expansion
between cells with increases in
temperature.

The spacing between cells tries

to increase an amount δ given
by:

where:
αG and αC are the expansion
coefficients of the glass and the
cell respectively;
D is the cell width; and
C is the cell centre to centre distance.

Typically, interconnections between cells are looped, as shown, to minimise cyclic stress. Double interconnects are used to protect
against the probability of fatigue failure caused by such stress.

In addition to interconnect stresses, all module interfaces are subject to temperature-related cyclic stress which may eventually lead to
delamination.

Electrical and Mechanical Insulation

Add new comment

The encapsulation system has to be able to withstand voltage differences at least as large as the system voltages. Metal frames must
also be earthed, as internal and terminal potentials can be well above the earth potential. Any leakage currents to earth must be low to
prevent interference with earth leakage safety devices.

Solar modules must have adequate strength and rigidity to allow normal handling before and during installation. If glass is used for the
top surface, it must be tempered, since the central areas of the module become hotter than areas near the frame. This places tension at
the edges, and can cause cracking. In an array, the modules must be able to accommodate some degree of twisting in the mounting
structure, as well as to withstand wind induced vibrations and the loads imposed by high winds, snow and ice.
Possible module twisting on a distorted mounting frame 1.

The standards set by the Australian Standards AS 4509-1999 include:

1. Static load - 3.9 kPa for 1 hour front then back (equivalent to 200 km/hr winds).
2. Dynamic load - 2.5 kPa front then back over 2,500 - 10,000 cycles (160 km/hr winds).
3. Hail impact damage - 2.5 cm diameter ice ball at terminal velocity 23.2 m/s (~ 80 km/hr).

1. Citekey JPL1981 not found

Degradation and Failure Modes

Add new comment
1 comment(s)

Because it has no moving parts (the major source of reliability issues in other types of electrical generating systems), a PV module's
operating life is largely determined by the stability and resistance to corrosion of the materials from which it is constructed.
Manufacturer's guarantees of up to 20 years indicate the quality of bulk silicon PV modules currently being produced. Nevertheless,
there are several failure modes and degradation mechanisms which may reduce the power output or cause the module to fail. Nearly all
of these mechanisms are related to water ingress or temperature stress.

Example of PV module degradation or failure. Shown here is the degradation of the antireflection coating of a solar cell caused by
water vapour ingress.

Reversible Reductions in Output Power

A PV module may be producing reduced output for reversible reasons. It may be subject to shading, for example, by a tree which has
grown in front of it. The front surface may be soiled (PV modules will generally experience up to 10% loss of output due to front
surface soiling). One module may have failed, or the interconnects between modules may have changed the operating point of the
array. However, these reductions in power are all reversible, provided that the original cause is rectified.
Soiling of the top surface of the array may cause either mismatch losses or a more uniform reduction in power output.

Degradation and Failure of PV Modules

Degradation mechanisms may involve either a gradual reduction in the output power of a PV module over time or an overall reduction
in power due to failure of an individual solar cell in the module.

Solar Cell Degradation

A gradual degradation in module performance can be caused by:

1. increases in RS due to decreased adherence of contacts or corrosion (usually caused by water vapor);
2. decreases in RSH due to metal migration through the p-n junction; or
3. antireflection coating deterioration.

Short-Circuited Cells

Short circuiting can occur at cell interconnections, as illustrated below. This is also a common failure mode for thin film cells since top
and rear contacts are much closer together and stand more chance of being shorted together by pin-holes or regions of corroded or
damaged cell material.

Cell failure through interconnect shorting.

Open-Circuited Cells

This is a common failure mode, although redundant contact points plus "interconnect-busbars" allow the cell to continue functioning.
Cell cracking can be caused by:

1. thermal stress;
2. hail; or
3. damage during processing and assembly, resulting in "latent cracks", which are not detectable on manufacturing inspection, but
appear sometime later.
Cracked cell indicating how "interconnect" busbars can help prevent open-circuit failure.

Interconnect Open-Circuits

Fatigue due to cyclic thermal stress and wind loading leads to interconnect open circuit failures.

Module Open-Circuits

Open circuit failures also occur in the module structure, typically in the bus wiring or junction box.

Module Short-Circuits

Although each module is tested before sale, module short circuits are often the result of manufacturing defects. They occur due to
insulation degradation with weathering, resulting in delamination, cracking or electrochemical corrosion.

Module Glass Breakage

Shattering of the top glass surface can occur due to vandalism, thermal stress, handling, wind or hail.

Module Delamination

A common failure mode in early generations of modules, module delamination is now less of a problem. It is usually caused by
reductions in bond strength, either environmentally induced by moisture or photothermal aging and stress which is induced by
differential thermal and humidity expansion.

Hot-Spot Failures

Mismatched, cracked or shaded cells can lead to hot-spot failures, as discussed previously in Hot Spot Heating.

By-Pass Diode Failure

By-pass diodes, used to overcome cell mismatching problems, can themselves fail, usually due to overheating, often due to undersizing
1. The problem is minimised if junction temperatures are kept below 128°C.

Encapsulant Failure

UV absorbers and other encapsulant stabilizers ensure a long life for module encapsulating materials. However, slow depletion, by
leaching and diffusion does occur and, once concentrations fall below a critical level, rapid degradation of the encapsulant materials
occurs. In particular, browning of the EVA layer, accompanied by a build-up of acetic acid, has caused gradual reductions in the output
of some arrays, especially those in concentrating systems 2

1. S. Durand, “Attaining Thirty-Year Photovoltaic System Lifetime”, Progress in Photovoltaics: Research and Applications,
1994.
2. H. J. Wenger, Schaefer, J., Rosenthal, A., Hammond, B., and Schlueter, L., “Decline of the Carrisa Plains PV Power Plant:
 The Impact of Concentrating Sunlight on Flat Plates”, 22nd IEEE Photovoltaic Specialists Conference. Las Vegas, USA, pp.
586-592, 1991.

Module Measurement
Safety First

Do not measure a module when it is installed in an array. The voltage is up to 1000 V and deadly.
Only measure one module at a time. Do not measure modules connected together.
Check that the module voltage is below 50 V by consulting the label.
Modules get hot in sunlight. Be careful about hot surfaces and/or wear gloves.

The accurate measurement of a module is quite challenging. Special testers give the standard test conditions: 1000 W/m2 of sunlight,
25 °C and spectrum of Airmass 1.5 (AM 1.5). The guides on the following pages are to give an indication that the module is
functional. Further tests would be needed to ensure that the module is performing to specification.

Interpretation of a module label

A module label shows the specifications of a module under standard test conditions. It is usually on the back of the
module. Standards vary slightly throughout the world but most labels will follow this format.

Modules from major manufacturers are measured according to standards. A typical module label is shown above. Measurements are at
STC unless otherwise stated.
Solar Module Type
Usually looking up the module type will give a more detailed data sheet on the module. In this case, we can tell its a 295 W
module and there are 60 cells in the module. As there are 60 cells it is more likely to be used in residential applications than a
large centralized power station.
Maximum Power (Pmax)
295 watts is the power of the module under STC at the maximum power point. In the field, the module will likely be at a higher
temperature so the actual power will be lower.
Power Tolerance
Under STC the module might have a power up to 3% higher. A higher power is not always a good thing as it could overload the
power electronics.
Maximum Power Voltage (VMP)
When operating at Pmax the voltage is 32.4 volts.
Maximum Power Current (ISC)
When operating at Pmax the current is 9.1 amps. Note that 32.4 V × 9.1 A = 295 W.
Open Circuit Voltage (VOC)
At STC the no-load voltage is 39.7 volts. This is the voltage that would be measured on a sunny day when the light intensity is
close to 1000 W/m2 and the temperature is 25 °C.
Maximum Power Current (ISC)
When the leads of the panel are shorted the current is 9.61 amps.
Nominal Operating Cell Temperature (NOCT)
In the field the module will operate at 45°C under the specific conditions of NOCT, which is different from STC. NOCT
has Irradiance 800 W/m2, air temperature or 20°C, wind velocity of 1 m/s and an open back.

The rest of the details cover how the module is installed and that it passes safety standards.

STC
STC – Standard test conditions. The label finally notes that the standard test conditions are 1000 W/m2 of sunlight, 25 °C and
spectrum of Airmass 1.5 (AM 1.5). All modern modules are tested using these conditions.

The label usually does not give the fill factor but we can calculate it. In this case it is (32.4 × 9.1)/(39.7 × 9.61) = 0.772. Most modules
will have a fill factor from 0.7 to 0.8.

Acknowlegement

The module measurement content was developed by the QESST Research Experience for Teachers (RET) program in Summer 2018.
Team members (alphabetical order):

Scott Currier (Fourth Grade Science Teacher, Highland Lakes School, Deer Valley Unified School District
Lauren D'Amico (Science Teacher, Barcelona Middle School, Alhambra Elementary School District)
Mark Calhoun (Physics Teacher, Camelback High School, Phoenix Union School District)
Elliot Hall (Science Teacher, Barcelona Middle School, Alhambra School District)
Alyssa Johnson (Akimel-A-al Middle School, Kyrene Elementary School District)
Milt Johnson (Physics and Engineering Teacher, Bioscience High School, and Maricopa Community College Instructor)
Leah Moran (Sonoran Trails Middle School, Cave Creek Unified School District)
Meredith Morrissey (Science Teacher, Tempe High School, Tempe Unified School District)
Myra Ramos (Math and Science Teacher, Alhambra Elementary School District)
Tamara Waller (Fourth Grade Teacher, Alhambra Elementary School District)
Allison Wolf (High School Science and Sustainability Teacher, Tempe High School, Tempe Unified School District)

Module Measurement without Load

Simple Module Measurement with a Multimeter
Read the safety instructions before proceeding.

Measuring the full power output of a solar module requires a load. However, as a first step, we can use a simple multimeter to measure
with no load to get the open current voltage, (VOC) and short circuit current (ISC). For large outdoor modules, any multimeter with a
current scale that goes to 10 A (amps) and 50 V (Volts) will work. See below for the requirements on smaller modules. A thermal gun
for measuring the temperature of the panel is useful on especially hot days. Choose a sunny day and point the module so it faces the
sun. Make sure none of the module is shaded. Even shading part of one corner of the module will cause a dramatic loss in output.

In a typical multimeter, the negative terminal is the black lead and labeled COM. Plug the red lead into the V terminal for measuring
voltage and
the 10 A for
large currents.
If you are not
familiar with
a

A handheld multimeter. Measurement of a solar panel uses the 10 A and the 200 V settings. (picture source)

multimeter, here is a website that describes what a multimeter is and here is a link about how to use a multimeter. There is also a video
on youtube.

A temperature gun is only really needed on very hot days.

The temperature should be measured on the back of the
module as the reflection of the sun from the module will
distort the reading. The material emissivity will also affect
the reading so that point the sensor at a metal surface will
give a different reading to back sheet even if they are at the
same temperature. In most cases, the gun can be pointed
anywhere on the back-sheet and give a reliable reading.
However, a piece of black tape on the module can also be
used to give a more reliable reading.

Measuring Voc
For VOC, set the multimeter to a DC voltage scale greater
than the expected module voltage. Use the 200 V scale
setting if unsure. Connect multimeter leads as shown below
to the solar panel leads (power to power and ground to
ground) and record the voltage. The VOC readings will
decrease as the module heats up and this variation will be
accounted for below. Record the temperature of the surface
of the panel and the time of day in which your
measurements were taken. Temperature measurement gun (picture source)

Measuring Isc
Disconnect the multimeter from the module before changing
the setting. For ISC, set the multimeter to a DC current scale
greater than the expected module current. Use the 10 A
scale setting if unsure. Change the power (red) lead to the
10 A socket to prevent blowing the fuse inside the
multimeter when taking your measurements. Connect the
multimeter leads to the solar panel leads and record the
voltage.

A video on how to measure current with a multimeter can

be found on youtube.

Discussion
Multiplying the VOC and ISC together gives a rough
estimate of the power. In this example case, the Voc equals
35.8 V, the current equals 10.07 and the product is 363.5 W.
However, we must also take into account the fill factor. The
maximum power (PMAX, or sometimes written MPP)
equation is:

$$P_{MAX} = V_{OC} \times I_{SC} \times FF$$

We cannot measure the fill factor without a load but it

is typically around 0.7, so our PMAX is in this case 254.5 W.
By determining the actual VOC, ISC, and estimated PMAX of
the panel and comparing it to the rating of the module, one
can quickly evaluate if a panel is performing near optimal.
In the image below, a picture of this panel’s rating is
included. Since the panel’s VOC and ISC are quite similar to The 200 V, direct voltage setting.
the results determined in these measurements, it is likely
that the panel
is performing
well.

Correct lead connections for voltage measurement. If the voltage reading is negative, try switching the leads

Limitations
Although this is a quick and simple way to learn some
information about the quality of a module, there are some
distinct limitations. While this method shows that the solar
module is functional and is putting out some amount of
power, it fails to characterize the module’s efficiency or
give any further specifications. This approach also fails to
generate an IV curve, which means that while PMAX can be
estimated, it cannot be determined exactly, or show how the
panel might be affected by shunt or series resistance. For a
better understanding of how the panel will operate with a
load and to get these missing data points, refer to the next
page. The measurement of VOC will be accurate to within
10% but ISC might be out by 50% or more.

Variations in VOC

The VOC is very dependent on the temperature and, to a

lesser extent, on the sun’s intensity. Each cell in the module
will fall by about 2.2 mV/°C. The standard for module
measurement is 25 °C so we need to correct by 2.2 mV per
cell for each degree above 25 °C. To correct for temperature
in a silicon module use the following formula:

$$V_{corrected} = V_{measured} + 0.0022 \times N \times

(T_{measured} - 25)$$

where Vmeasured is the voltage measured at VOC , Tmeasured

is the module temperature in Celsius and N is the number of
cells in the module.

Variation in ISC

ISC varies in proportion with the sun’s intensity, which A module with a multimeter attached.
varies with location and the time of day. Measuring the
sun’s intensity is challenging and is covered on subsequent
pages. A rough estimate is provided by measuring more than one module. If they all have about the same measured ISC then it is likely
they are all functioning correctly. Measurement on a cloudy day will give about 10-20% of the rated output. Measurements in the shade
or in a garage

Acknowledgment

The module measurement content was developed by the QESST Research Experience for Teachers (RET) program in Summer 2018.
Team members (alphabetical order):

Scott Currier (Fourth Grade Science Teacher, Highland Lakes School, Deer Valley Unified School District
Lauren D'Amico (Science Teacher, Barcelona Middle School, Alhambra Elementary School District)
Mark Calhoun (Physics Teacher, Camelback High School, Phoenix Union School District)
Elliot Hall (Science Teacher, Barcelona Middle School, Alhambra School District)
Alyssa Johnson (Akimel-A-al Middle School, Kyrene Elementary School District)
Milt Johnson (Physics and Engineering Teacher, Bioscience High School, and Maricopa Community College Instructor)
Leah Moran (Sonoran Trails Middle School, Cave Creek Unified School District)
Meredith Morrissey (Science Teacher, Tempe High School, Tempe Unified School District)
Myra Ramos (Math and Science Teacher, Alhambra Elementary School District)
Tamara Waller (Fourth Grade Teacher, Alhambra Elementary School District)
Allison Wolf (High School Science and Sustainability Teacher, Tempe High School, Tempe Unified School District)

Module Measurement with Load

This page is being worked on March 2019

This module measurement method uses the variable of resistance to determine the power IV curve. By changing the resistance of the
module load and measuring voltage and current, the
power IV curve can be generated for a specific
panel. This method will ultimately allow the user of
the module to compare and contrast the factory
curves provided for that module. A power IV curve
can provide the maximum power point (PMAX) of
the module, which can be contrasted with the
manufacturer’s specifications of the expected power
output. Because there is a power IV curve, this
method can also help identify shunt and series
resistance in a module. A discussion of the effects of
resistance on a solar module can be found here.

Measuring with a Load

Ideally, we want to operate the module at the
maximum power point. The module voltage is VMP
and the module current is IMP. We already know the
power output of the module PMAX but we also need
the load resistance, Rload, which is found from ohms
law:

$$R_{load} = \frac{V_{MP}}{I_{MP}} $$

In the case of the module below, where VMP equals

32.4 V and IMP equals 9.1 A, Rload should be 3.5
ohms.

The resistors used should be around the maximum

power voltage (Vmp) and the maximum power
current (Imp). If the Vmp and Imp are not known,
open circuit voltage (Voc) and short circuit current
(Isc) is a good approximation. Usually, the back of A module with a multimeter attached. The 10 A, direct current setting
the module will have a label listing VMP and the

Leads correctly connected for testing direct current. Be sure that the lead into the multimeter is connected into the 10A
socket.
IMP. The VOC and ISC can be found using a
multimeter.

Modules produce hundreds of watts or power so

high wattage power resistors are needed.
However, the measurements are only for short
periods so we don’t need power resistors of the
full power rating. Cooling the resistors with
water also helps. The ones used in the example
results were designed to withstand 100 Watts and
varied between 0.5 Ω, 1 Ω, and 4 Ω. When
connected in various series combinations, the
resistance ranged from 1 Ω to 19 Ω. About a fifth
to five times the calculated resistance at PMAX
calculated above.

Materials needed

Two Multimeters - One for voltage and one

for current
Water to cool the resistors
Two or more alligator clip cords
Protective gloves - modules and resistors
get hot
Resistors
Solar Module

The schematic for placing the meters is shown

below. The resistors heat up during the
measurement and their value will likely drift.
Having two meters, one for current and one for
voltage means that we do not need precision
resistors or have to the uncertainty of drift during
the measurement.

Measurement procedure Example of a panel's ratings provided by the manufacturer.

Place the module in a position where it will

have full sunlight and shading. The easiest is to lie the module flat near solar noon.
Set one meter to the voltage scale and one to the current scale.
Don’t add the resistors just yet. Follow the instructions on the previous page to measure VOC and ISC
Now add the resistors using the schematic above and record the voltage and current for each resistor.
Change the resistor values to cover the IV curve.
After measuring with all the resistors, remeasure VOC and ISC

After the VOC and ISC are measured, the resistors can be swapped in and out to gather results. Connect the black lead from the current
reading multimeter to one end of a resistor. Using an additional black alligator cord, connect one clip to the positive end of the panel.
There will be one free clip leftover. That free clip will be used to connect to the resistor thus completing the circuit.

Results and Discussion

Record the cell voltage and current for each resistor. The power is \( P = I \times V \). The resistance is also calculated from \( R = V/I
\) but it is not necessary.

Voltage (V) Current (A) Power (W) Resistance

 (Ω)
0.00 9.33 0.00 0.00
6.4 9.32 59.6 0.69
13.4 9.36 125.4 1.4
24.2 8.80 213.0 2.8
26.0 7.60 197.6 3.4
27.7 6.74 186.7 4.1
28.3 6.07 171.8 4.7
29.4 4.66 137.0 6.3
30.3 3.69 111.8 8.2
30.0 3.46 103.9 8.7
30.6 2.95 90.3 10.4
31.0 2.53 78.4 12.3
31.0 2.42 75.0 12.8
31.1 2.16 67.2 14.4
31.3 1.92 60.1 16.3
31.3 1.80 56.3 17.4
31.3 1.70 53.2 18.4
32.6 0.00 0.00 0.00
Most modules have a label on the back giving the performance under standard test conditions (STC).
Graph the current

A collection of power resistors for measuring a module IV curve. Placing the resistors in series and parallel gives a
multitude of values.

against the voltage and the power against the voltage

Recommended materials for a module measurement as listed above

Schematic of the connections to the module.

The red meter is reading voltage and the blue meter is reading current. Use alligator clips on the resistors to easily change
values.

connect resistors in series. Use an additional black alligator clip to connect the resistors for a value needed.
Handle the resistors with caution, they will be hot. Use protective gloves and spray the resistors with water to help with
cooling. The measurements take a while so shade is preferable for the people but absolutely no shade must fall on the
module.
Discussion

The module PMAX is the maximum in the power column, which is 213 W in the example above. The fill factor (FF) is calculated from
the equation:

$$FF = \frac{P_{MAX}}{V_{OC} \times I_{SC}} $$

For the measurement above, ISC is the first line of the table (where V = 0) and equals 9.33 A and VOC is the last line of the table
(where I = 0 ) and equals 32.6 V. The resulting FF is 0.70. Comparing the measured results to the specifications listed on the module
label we have:

Parameter Specification Measured

PMAX (W) 295 213
VOC (V) 39.7 32.6
ISC (A) 9.61 9.33
FF 0.77 0.7

The previous page discusses the variation in VOC and ISC as a result of temperature and light intensity respectively. The measured FF
was 10% lower than the specification, which is caused by the considerable scatter in the data. It is difficult to keep the module
temperature and light intensity constant during the measurement. For instance, a person moving around the module is enough to alter
the light on the module (even if they do not cast a shadow) by blocking some of the diffuse light.

Acknowledgment

The module measurement content was developed by the QESST Research Experience for Teachers (RET) program in Summer 2018.
Team members (alphabetical order):

Scott Currier (Fourth Grade Science Teacher, Highland Lakes School, Deer Valley Unified School District
Lauren D'Amico (Science Teacher, Barcelona Middle School, Alhambra Elementary School District)
Mark Calhoun (Physics Teacher, Camelback High School, Phoenix Union School District)
Elliot Hall (Science Teacher, Barcelona Middle School, Alhambra School District)
Alyssa Johnson (Akimel-A-al Middle School, Kyrene Elementary School District)
Milt Johnson (Physics and Engineering Teacher, Bioscience High School, and Maricopa Community College Instructor)
Leah Moran (Sonoran Trails Middle School, Cave Creek Unified School District)
Meredith Morrissey (Science Teacher, Tempe High School, Tempe Unified School District)
 Myra Ramos (Math and Science Teacher, Alhambra Elementary School District)
Tamara Waller (Fourth Grade Teacher, Alhambra Elementary School District)
Allison Wolf (High School Science and Sustainability Teacher, Tempe High School, Tempe Unified School District)

Characterisation
Add new comment

Characterisation module

Introduction
Add new comment

Characterisation Types

There are three basic types of characterisation in use:

1. Full Area Measurement

Here the entire solar cell or large fraction wafer is measured giving one data point per wafer. The measurements are typically fast
enough for inline characterisation on every wafer going through a production line and the data produced is suitable for statistical
process control (SPC). an example of full area measurement is an IED test at the end of solar cell processing, the measurement
of a parameter such as reflectivity on part of a wafer, or a lifetime test using a coil based sensor. The disadvantage of full area
measurements is that it does not reveal area related problems such as poor printing on part of a wafer.
2. Mapping
mapping techniques rely on point by point measurements to scan across a wafer surface. They may provide a great deal of
accuracy about each point on the wafer and reveal defects such as grain boundaries or other imperfections. The drawback of
mapping techniques is the time taken to cover an entire wafer. For example if a measurement takes 0.1 seconds and a resolution
of 100 µm is required for a 1500x1500 array of data points then the time to map an entire wafer will be 62 hours. The long
sample times usually makes wafer mapping unsuitable for in-line characterization. An example of wafer mapping is an LBIC
system where a laser is scanned across the surface of the wafer and the current is read out of each data point. due to the times
required for two-dimensional images mapping techniques are often simplified by using line scans.
3. Imaging
Imaging away for is somewhat similar to taking a photograph the measurement technique uses a sensor array to sample multiple
points simultaneously. the advantage of imaging techniques is that large arrays of data points can be taken in quite small short
periods of time. The disadvantage is the cost of the sensors. The role of imaging techniques is increasing due to the appearance
of low cost silicon CCD cameras similar to those used in digital cameras.

Electrical Characterisation

The first section discusses the electrical measurement of solar cells with the three basic techniques:

1. Illuminated IV, where the cell is illuminated at one sun and basic cell parameters measured
2. Dark IV, where the cell is in the dark and the cell IV characteristic traced
3. Jsc Voc where the Voc and Isc are recorded at different illumination levels

These three techniques have numerous names and technical variations.

Measurement of Solar Cell Efficiency

Add new comment

The most fundamental of solar cell characterization techniques is the measurement of cell efficiency. Standardized testing allows the
comparison of devices manufactured at different companies and laboratories with different technologies to be compared.
The standards for cell testing are:

1. Air mass 1.5 spectrum (AM1.5) for terrestrial cells and Air Mass 0 (AM0) for space cells.
2. Intensity of 100 mW/cm2 (1 kW/m2, also known as one-sun of illumination)
3. Cell temperature of 25 °C (not 300 K)
4. Four point probe to remove the effect of probe/cell contact resistance
Constructing a system that meets all the above criteria simultaneously is difficult and expensive. Most research laboratories have
simple custom-built testers that only roughly approximate the above conditions. Results that are tested "in-house" are typically quite
approximate. Periodically companies and research institutions will send devices with record efficiencies to certified testing laboratories
for confirmation. Such efficiencies are published as "confirmed efficiencies" along with the name of the test center and the testing date
(see appendix).

Basic structure of a simple IV tester. The current and voltage are measured separately to overcome contact resistance problems.

In the diagram of the IV tester the electronics are shown as all one unit, however, the voltage and current can be measured separately
from the power source. Adjusting the height of the lamps rather than the power to the lamps keeps the spectrum constant. However
small adjustments to the lamp power are possible without unduly affecting the light spectrum.

IV tester components and Sources of Error

The following pages discuss the individual components of a measurement system and the source of error each component may
introduce.

Illumination Sources
Add new comment
1 comment(s)

Measuring solar cells requires a stable light source that closely matches the conditions of sunlight. Not only the intensity but also the
spectrum must be matched to a standard. An obvious option is to simply use the sun itself. In locations where there is little cloud this is
a good solution 1 but there are still variations in atmospheric conditions that require correction to compare measurements over time.
The spectrum also changes through out the day and this further limits the time for testing.

The most common solution is to use an artificial light source that simulates the sun. The ideal illumination source would have
following features 2;

1. a spatial non uniformity of less than 1%.

2. a variation in total irradiance with time of less than 1%,
3. filtered for a given reference spectrum to have a spectral mismatch error of less than 1%.
4. These requirements are essential in obtaining an accuracy of better than 2%

Testers are classified according to three criteria:

1. Spectral match
2. Irradiance inhomogeneity - spatial uniformity over the illumination area
3. Temporal Instability - stability over time.

There are three classes within each of these criteria where 'A' is the top rating an 'C' is the lowest rating
Table: Solar simulator classification according to IEC 60904-9 Ed. 2.0.

Irradiance
Class Spectral Match Temporal Instability
inhomogeneity
Long Term Short Term

A 0.75 - 1.25% 2% 0.5% 2%

B 0.6 - 1.4% 5% 2% 5%
C 0.4-2.0% 10% 10% 10%

For instance, a simulator with the designation ABA would have a spectral match < 1.25%, an inhomogeneity 2-5%, and stability <
0.5% in the long term and < 2% in the short term.

Many simulator related artifacts can be eliminated by using a reference cell that has the same spectral response as the cell under test.

The most common light source is a Xenon arc lamp with filters installed to approximate the AM1.5G spectrum. Simple testers often
just use a halogen lamp with a dichroic filter. The lamp filament is much lower than the sun's 6000 K so it produces much more
infrared light and much less UV. The reflector on the bulb is selective so that the visible and UV is reflected towards the cell but most
of the infrared radiation isn't reflected and passes out the back of the bulb. Halogen lamps have the advantage of greater temporal
stability compared to Xenon arc lamps.

ELH lamp with a dichroic reflector so that most of the infrared is not reflected. The ELH is an American Standards Institute (ANSI)
code that describes this type of lamp.
The standard airmass 1.5 spectrum compared with the spectrums from typical solar simulator sources. Click on the graph for a scalable
version.

There are also considerable variations between individual samples of the same type and with age.

Deviations from Air Mass 1.5

deviations from AM1.5 cause errors in Isc

It is difficult to make a light source that exactly matches the AM1.5 spectrum and with the necessary illumination intensity. As the
previous graph shows there is often a considerable amount of variation between the spectrum of the lamp and the required AM 1.5
spectrum. There are two approaches for correcting for the differences between the AM1.5 spectrum and the actual spectrum from a
solar simulator.

Calibration cell with the same spectral response.

The approach taken by most in-house testers is to use a calibration cell that has the same spectral response as the cell under test. The
light intensity of the tester is adjusted so that the cell Isc matches the Isc as measured at an external testing laboratory. However, slight
changes in cell processing (e.g. the doping profile of the emitter, variation of anti-reflection coatings) cause changes in spectral
response and the need for a new calibration standard.

Measure Spectral Response

Primary calibration labs use light sources that are much closer to the standard however differences still occur. To compensate for the
differences, calibration labs measure the spectral response of the device under test and then use that to correct for the known difference
between the spectrum of the light source and the standard spectrum 3. Such corrections are time consuming and prone to error. In the
early 90's an analysis of test error lead to an improved standard and also the down grading of some record efficiencies by up to 1%
absolute 4.

There is an online calculator for determining the level of spectral mismatch from a light source at PV Lighthouse. It includes the ability
to correct for an arbitrary spectrum.

1. W. Keogh and Blakers, A. W., “Natural Sunlight Calibration of Silicon Solar Cells.”, 17th European Photovoltaic Solar
Energy Conference. Munich, Germany, 2001.
2. K. Emery, Myers, D., and Rummel, S., “SOLAR SIMULATION - PROBLEMS AND SOLUTIONS”, 20th IEEE PV
Specialists Conference. p. 1087, 1988.
3. “Test Method for Determination of the Spectral Mismatch Parameter Between a Photovoltaic Device and a Photovoltaic
Reference Cell”. 2016.
4. D. King and Hansen, B., “A Sensitivity Analysis of the Spectral Mismatch Correction Procedure Using Wavelength-
Dependent Error Sources”, 22nd IEEE PV Specialists Conference. 1991.

Temperature Control
Add new comment

One-sun illumination is quite intense so there has to be some mechanism to remove the excess heat. Typically the cell is placed on a
large metal block and water runs through the block to keep it cool. A thermocouple is inserted in the block and the control system is
adjusted to the required 25 °C. The simple arrangement works very well for commercial cells where the entire rear is covered with
metal and makes good contact with the test block.

A more sophisticated arrangement is required for cells where some or all of the contacts are on the cell rear, or for bifacial
illumination. A common method used is flash testing where light is flashed on the cell and measurements are taken very quickly. While
this largely eliminates temperature control problems sophisticated electronics are needed to take measurements quickly and
synchronized with the flash. Flash testing is also commonly used for module testing where it is generally not possible to directly
control the cell temperature due to the encapsulation.

Poor temperature control introduces errors in VOC. The error is dependent on the bandgap of material (see Effect of
Temperature).

Electronics
 Add new comment
1 comment(s)

The solar cell light IV curve can be traced by altering a variable resistor across the cell and recording the voltage and current at the cell
terminals. While this is quite simple, it is time consuming so in practice more sophisticated electronics is used. The most common
method is to use a variable voltage source that is also capable of sinking current. To improve accuracy Isc and Voc are usually
measured separately (by respectively setting the voltage to zero and the current to zero) from the rest of the curve. The IV curve has a
strongly varying slope that causes additional problems. There are various schemes for improving the accuracy but one of the simplest
is to use equally spaced measurement in voltage in two steps. The first section is widely spaced and covers from 0 to 70 % of Voc. The
second section has points more closely spaced and covers from 70% to Voc. The second region contains the maximum power point, the
open circuit voltage and has a much higher slope.

IV measurement. The second sweep has a higher resolution and covers the more interesting parts of the curve.

For cell testing the difficulty is the high current of large area cells. The trend in silicon substrates is to move to larger substrates of 15 x
15 cm² and an Isc approaching 10 amps. For Dark IV measurements the tester also needs to source current and to be varied over several
orders of magnitude, from nanoamps to several amps. There now exist several systems that fulfill these requirements and are integrated
into one package with computer control but the user has to specify the size they want.
Equipment rack for cell tester at NREL (Photo: Jim Yost Photography, NREL Picture Exchange)

Probing
Add new comment

Probing errors primarily cause problems with FF but can also cause Voc and Jsc errors.

Cell testing uses a four point probe to contact the cell. A current and voltage probe on top of the cell and a current and voltage probe on
the bottom of the cell. The most common arrangement is to have the metal of the block act as the rear current probe and then to have a
voltage pin through it. For the top contacts it is usually insufficient to have a single voltage and current probe thus a number of pairs
are used.

Top side probes in a cell tester. At each end of the busbar there is a current voltage probe pair

Ideally, the probes make good contact with the cell and the voltage and current probes are within close proximity but not touching each
other. Spurious measurements occur if the probes don't make good contact with the cell or if the probes are spaced a distance apart.
Depending on the internal arrangement of the measurement electronics, the apparent fill factor can rise above the theoretical maximum
(0.85 for silicon solar cells at one sun illumination). The simple solution to contacting problems is to move the probes on the cell
slightly and remeasure the cell. Large changes in FF when moving the probes are indicative of a contacting problem.

Dark IV Measurements
Add new comment

Since solar cells convert light to electricity it might seem odd to measure the photovoltaic cells in the dark. However, dark IV
measurements are invaluable in examining the diode properties. Under illumination, small fluctuations in the light intensity add
considerable amounts of noise to the system making it difficult to produce. Dark IV measurements use inject carriers into the circuit
with electrical means rather than with light generated carriers. In most cases the two are equivalent and the Dark IV measurements give
extra information about the cell for diagnostic purposes. Even in the absence of noise, there is a wealth of information in comparing the
illuminated and dark IV curves.

A solar cell in the dark is a large flat diode. A simple dark IV measurement produces the exponential curve so characteristic of a diode.
Dark IV curve with a linear scale. One exponential looks much like another.

The linear graph of current vs. voltage reveals very little information about the diode, much more information is revealed from a
semilog plot.

Semilog plot of the same IV curve above revealing much more information about the diode. Different regions of the IV curve are
dominated by different loss mechanisms.
Dark IV measurement of a PERL cell. While the cell had a record Voc of 702 mV the fill factor was only 0.81, lower than the 0.83 for
earlier cells. Additionally, there was no evidence of parasitic series or shunt resistance. The dark IV measurement showed an unusual
curve where the J0 was changing and was explained by a changing recombination velocity at the rear surface 1.

Limitations of Dark IV Measurements

The use of Dark IV curves in solar cell analysis relies on the principle of superposition. That is, in the absence of resistive effects, that
the light IV curve is the dark IV curve shifted by the light generated current. While this is true for most cells it is not always the case.

A second problem is that in dark IV measurements the current is flowing in the opposite direction and the current paths are different.
The change in the current path causes a lower series resistance in the dark measurements to the light measurements.

Comparison of current paths under illumination and in the dark. In both cases the currents are the same. In the dark case the current
flows into the cell and in the illuminated case the current flows out of the cell. Since in the dark case, most of the current crosses the
junction under the contact it has a lower series resistance than for the illuminated case. (move the mouse over the image to see dark
current flows)

1. J. Zhao, A., W., Dai, X., Green, M. A., and Wenham, S. R., “Improvements in Silicon Solar Cell Performance”, 22nd IEEE PV
Specialists Conference. pp. 399-402, 1991.

Jsc-Voc and Suns-Voc

Add new comment

Jsc Voc

A JSC VOC curve1 is a valuable way of looking at an IV curve in the absence of series resistance. To trace a JSC VOC curve the
illumination on a cell is varied and the cell JSC and VOC measured at each illumination level. The series resistance has no effect on the
VOC since no current is drawn from the cell. The series resistance has no effect on JSC so long as the series resistance is less than
10 Ωcm2 since the IV curve is flat around JSC as illustrated in the series resistance page. The curve that is traced is similar to the Dark
IV curve but without the effects of series resistance.

Plot Jsc@1 sun - Jsc

Concentation, 1 suns

0.01 1.5

Rs =

0.1 3

Rshunt =

100 105

The graph on the left is a standard JV plot with the Jsc Voc point highlighted.

SunsVoc

SunsVOC measurements 23 are very similar to the JSC VOC measurements except that the SunsVOC uses a separate solar cell to
monitor the illumination intensity rather than using the JSC of the solar cell. The monitor cell simplifies the measurement electronics
since measuring cell voltage is much simpler than measuring the cell current. By using a flash lamp with a slow decay a SunsVOC can
be taken very quickly and simply.

The SunsVOC measurement provides the IV curve of the diode without the effects of series resistance. Fitting to the SunsVOC curve is
easier than to the dark or illuminated curves since there is no Rs.

1. M. Wolf and Rauschenbach, H., “Series Resistance Effects on Solar Cell Measurements”, Advanced Energy Conversion, vol.
3, 1963.
2. R. A. Sinton and Cuevas, A., “A Quasi-Steady-State Open-Circuit Voltage Method for Solar Cell Characterization”, in 16th
European Photovoltaic Solar Energy Conference, Glasgow, Scotland, 2000, pp. 1152–1155.
3. M. J. Kerr, Cuevas, A., and Sinton, R. A., “Generalized analysis of quasi-steady-state and transient decay open circuit voltage
measurements”, Journal of Applied Physics, vol. 91, p. 399, 2002.

Measurement of Series Resistance

 Add new comment

The series resistance of a solar cell dominates fill factor losses, especially in large area commercial solar cells, so an accurate
measurement is vital in quantifying losses. There are several methods to measure series resistance and the comparisons of the accuracy
for specific cell types.12

Curve Fitting

The simplest way to measure series resistance is to fit the illuminated IV curve with either the ideal diode equation or the double diode
equation. While this is conceptually very simple there are often problems in practice. One of the biggest problems is that the cell series
resistance is a lumped parameter composed of many resistances within the device. A solar cell is a three dimensional device and can be
thought of as a network of resistors and diodes. As the level of current changes so does the apparent series resistance. A Thévenin or
Norton equivalent circuit can only be constructed in the absence of non-linear elements such as diodes.

The blue points are measured data and the blue line is a double diode fit. Curve fitting only works so long as the externally seen Rs is
constant, which is rarely the case in practice.

Effect of Rs on Only Part of the Cell

The simplest way to demonstrate the problems caused by the deviations from the one dimensional model is with the model shown
below3. Here only part of the cell is affected by a series resistance as shown in the figure below:

Solar cell in which only part of the cell is affected by series resistance.

With only part of the cell affected by Rs a variety of curves are produced. In the simulation below try setting the fraction to 1 (i.e. the
simple case where Rs affects the whole cell) and adjusting the internal Rs. It sill follows the simple one dimensional case. The try
setting the fraction to 0.5, it produces a rounding around the maximum power point similar to a high J02. In the final case try setting
the fraction of the cell affected by Rs to 0.1. In this case the IV curve looks very similar to a cell with a low Rshunt.

Rs =

0 20

Fraction with high Rs, fraction = 0.5 units

0 1

Model of a solar cell where only part of the cell is affected by series resistance

Due to the practical limitations of curve fitting extra measurements need to be taken to measure series resistance.

1. A. Mette and et al, “Series resistance characterization of industrial silicon solar cells with screen-printed contacts using
hotmelt paste”, Progress in Photovoltaics: Research and Applications, vol. 15, pp. 493-505, 2007.
2. D. PYSCH, Mette, A., and Glunz, S. W., “A review and comparison of different methods to determine the series resistance of
solar cells”, Solar Energy Materials and Solar Cells, vol. 91, pp. 1698 - 1706, 2007.
3. S. Bowden and Rohatgi, A., “Rapid and Accurate Determination of Series Resistance and Fill Factor Losses in Industrial
Silicon Solar Cells”, in 17th European Photovoltaic Solar Energy Conference, Munich, Germany, 2001.

Double Diode Model

Add new comment

The single diode equation assumes a constant value for the ideality factor n. In reality the ideality factor is a function of voltage across
the device. At high voltage, When the recombination in the device is dominated by the surfaces and the bulk regions the ideality factor
is close to one. However at lower voltages, recombination in the junction dominates and the ideality factor approaches two. The
junction recombination is modeled by adding a second diode in parallel with the first and setting the ideality factor typically to two.
Circuit diagram of the double diode model including the parasitic series and shunt resistances

The equation of the double diode model under illumination is:

Practical measurements of the illuminated equation are difficult as small fluctuations in the light intensity overwhelm the effects of the
second diode. Since the double diode equation is used to characterise the diode it is more common to look at the double diode equation
in the dark.

In both the light and dark cases the -1 terms in the exponential are typically ignored as it makes the analysis far easier.

Under illumination:

In the dark:

The double diode equation in the dark is graphed below:

 Saturation current, I01 = 1e-12 A/cm2?

1e-14 1e-08

Saturation current, I02 = 1e-10 A/cm2?

1e-12 1e-06

Rs =

0.1 3

Rshunt =

100 105

Graph showing the double diode model. The device in gray has no parasitic resistance losses. The device in red has the loss of series
and shunt resistance included.

Limitations of the Double Diode Model

In actual silicon devices, the recombination components are a complex function of the carrier concentration. For example, in high
efficiency PERL solar cells as the number of carriers increase with the applied voltage, the recombination at the rear surface changes
dramatically with voltage. In such cases the analysis is best performed by a single diode, but allowing both the ideality factor and the
saturation current to vary with voltage. In such cases, which are quite common in silicon devices, a double diode fit yields erroneous
values.

Measuring Ideality Factor

Add new comment

The ideality factor is derived from the slope of the dark-IV, Suns-Voc and occasionally the Light-IV curve.

The basic cell equation in the dark is:

,
where I is the current through the diode, V is the voltage across the diode, I0 is the dark saturation current, n is the ideality factor and T
is the temperature in kelvin. q and k are both constants. for V > 50 - 100 mV the -1 term can be ignored and so the above equation
reduces to:

taking the log of both sides of the equation gives:

When plotting the natural log of the current against the voltage, the slope gives q/nkT and the intercept gives ln(I0). In real cells the
ideality factor depends on the voltage across the cell. The ideality factor can either be plotted as a function of voltage or it can be given
as a single value. Since the ideality factor varies with voltage, if given as a single value the voltage range also needs to be given.

Deviations in the ideality factor from one indicate that either there are unusual recombination mechanisms taking place or that the
recombination is changing in magnitude. Thus the ideality factor is a powerful tool for examining the recombination in a device. The
measurement of I0 is only valid when the ideality factor is stable.

There are several practical problems when measuring ideality factors:

At low voltages the shunt resistance (Rshunt)dominates the device performance and causes a large peak. It is usually not possible
in practice to correct for the effects of Rshunt.
At high voltages in a dark-IV curve the series resistance dominates and this causes a large peak in the ideality factor curve at
high voltages. This can be alleviated by using the Suns-Voc curve that as noted earlier gives a curve the same as dark-IV but
without the effects of series resistance.
The ideality factor comes from the differential of a signal so it is very prone to noise. Noise problems are particularly
problematical in Suns-Voc measurements. To reduces noise the slope is usually taken as a fit over several points.
The effects of temperature are a problem particularly if the temperature change during the measurement.

The ideality factor animated graph below shows the effects on a dark IV curve. The curve without resistive effects is quite simple, at
low voltages the ideality factor is two, at high voltages the ideality factor is one. Once the resistive effects are included the curve
becomes quite complicated. The domination by Rshunt at low voltages means that the ideality factor at low voltages is not valid.

Series resistance is: 0.100 Ω cm2

 Minority carrier lifetime: 0.001 seconds

Dark current I01: 1.00e-13 A/cm2

Dark current I02: 1.00e-10 A/cm2

Local Ideality factor determined from the dark IV curve with and without the effects of parasitic resistance.

Dark IV measurements of buried contact cells and the extracted local ideality factor fits. The different curves are from varying the
distance to the cell edge. The extracted ideality factors show that the unusual IV curves were due to the edge recombination (Picture
redrawn from McIntosh 1.

1. K. R. McIntosh and Honsberg, C. B., “The Influence of Edge Recombination on a Solar Cell’s IV Curve”, 16th European
Photovoltaic Solar Energy Conference. 2000.

Reflectance
Add new comment
1 comment(s)

Reflectance

Spectral Response
Add new comment
1 comment(s)

The spectral response is conceptually similar to the quantum efficiency. The quantum efficiency gives the number of electrons output
by the solar cell compared to the number of photons incident on the device, while the spectral response is the ratio of the current
generated by the solar cell to the power incident on the solar cell. A spectral response curve is shown below.
The spectral response of a silicon solar cell under glass. At short wavelengths below 400 nm the glass absorbs most of the light and the
cell response is very low. At intermediate wavelengths, the cell approaches the ideal. At long wavelengths, the response fall back to
zero. Silicon is an indirect band gap semiconductor so there is not a sharp cut off at the wavelength corresponding to the band gap (Eg
= 1.12 eV).

The ideal spectral response is limited at long wavelengths by the inability of the semiconductor to absorb photons with energies below
the band gap. This limit is the same as that encountered in quantum efficiency curves. However, unlike the square shape of QE curves,
the spectral response decreases at small photon wavelengths. At these wavelengths, each photon has a large energy, and hence the ratio
of photons to power is reduced. Any energy above the band gap energy is not utilised by the solar cell and instead goes to heating the
solar cell. The inability to fully utilize the incident energy at high energies, and the inability to absorb low energies of light represents a
significant power loss in solar cells consisting of a single p-n junction.

Spectral response is important since it is the spectral response that is measured from a solar cell, and from this the quantum efficiency
is calculated. The quantum efficiency can be determined from the spectral response by replacing the power of the light at a particular
wavelength with the photon flux for that wavelength. This gives:

Light Beam Induced Current

Add new comment
Module scanning with a laser beam. NREL Picture exchange

Bulk Lifetime
Add new comment

As discussed elsewhere, there are various recombination mechanisms within even a uniformly doped piece of semiconductor. In the
bulk of the material the carriers recombine by either radiative (also known as band-to-band) recombination, Auger recombination or
Shockley-Read-Hall (SRH) via traps within the energy gap.

The lifetime of carriers in the material bulk τb is composed of radiative lifetime τrad, Auger lifetime τA and a SRH lifetime τSRH with
the relation:

For an indirect bandgap semiconductor such as silicon τrad is very large and usually neglected.

The auger lifetime time can be calculated using theoretical models 1, however, the SRH lifetime depends on the level of defects in the
crystal lattice and so is difficult to calculate theoretically. The bulk lifetime for extrinsic silicon can be determined using semi-
empirical models based on lifetime measurements of float-zone silicon with very low defect levels. The lifetime is dependent on the
excess carriers and doped atoms concentrations. The models presented here are based on 2 but a more recent model was developed3.
Most silicon wafers have higher levels of contaminants and so lower lifetimes than calculated here. Further details on silicon properties
are in the appendices.

Bulk Lifetime for p-type Silicon

P Bulk Lifetime Calculator

Excess carriers concentration, Δn = cm-3 Doped

atoms concentration, NA = cm-3
Bulk Lifetime, τbulk = µsec

Bulk Lifetime for n-type Silicon

N Bulk Lifetime Calculator

Excess carriers concentration, Δp = cm-3 Doped

atoms concentration, ND = cm-3
Bulk Lifetime, τbulk = µsec
The dependence of the minority carriers lifetime from their concentration and semiconductor doping level for n- and p-type silicon.
You can copy data from the graph to the clipboard clicking the "Get Data" button.

1. J. Schmidt, Kerr, M. J., and Altermatt, P. P., “Coulomb-enhanced Auger recombination in crystalline silicon at intermediate
and high injection densities”, Journal of Applied Physics, vol. 88, pp. 1494-1497, 2000.
2. M. J. Kerr, Cuevas, A., and Sinton, R. A., “Generalized analysis of quasi-steady-state and transient decay open circuit voltage
measurements”, Journal of Applied Physics, vol. 91, p. 399, 2002.
3. A. Richter, Glunz, S. W., Werner, F., Schmidt, J., and Cuevas, A., “Improved quantitative description of Auger recombination
in crystalline silicon”, Physical Review B, vol. 86, no. 16, 2012.

Separator
Ipso

Surface Recombination
Add new comment
4 comment(s)

The surface also plays an important role in recombination. Typically the surfaces complicate the measurement of the bulk lifetime.
Recombination at the surfaces is typically described by a surface lifetime τs, which includes the fundamental decay mode but ignores
higher decay modes. τs is a function of the surface recombination velocities S1 and S2, the cell width W and the minority carrier
diffusivity D. The precise solution is quite complicated but approximate solutions exist for special cases and are sufficiently accurate
for most purposes 1.

The simplified formulas given below can be used for transient measurements when the effective lifetime is much greater than the

transient surface lifetime ( ). The more exact expression 2 3in transient measurements that works in all cases is:

$$ S = \sqrt{D \left(\frac{1}{\tau_{eff}}- \frac{1}{\tau_b}\right)} \tan \left[\frac{W}{2} \sqrt{\frac{1}{D}\left(\frac{1}{\tau_{eff}}-

\frac{1}{\tau_b}\right)}\right]$$

For quasi-steady-state measurements, the formulas are accurate if .

There is a list of approximate solutions for several important cases below:

Silicon wafer with a thickness of W

The surfaces are identical so S = S1 = S2.

One surface is perfectly passivated so S2 = 0.

Both surfaces are perfectly passivated so S1 = S2 = 0.

Both surface have high recombination so S1 and S2 are large.

One surface has a high recombination and the other has a low recombination so S1 = 0 and S2 = ∞

Effective Lifetime

The combination of the effects of surface and bulk recombination give the effective lifetime of carriers τeff within a sample.

For many measurements, it is possible to reduce surface recombination of the surfaces to a low enough value so that τb ≫ τ eff.
Alternatively, if τb is sufficiently high the lifetime measurement is completely dominated by the surfaces and τs ≫ τeff. An accurate
measurement of surface recombination is then possible.
 Effective Lifetime Calculator

Wafer thickness W = µm Diffusivity D = cm2/s The bulk

lifetime τb s The recombination of the first surface S = cm/s
Surface lifetime τs = sec Effective
lifetime τeff = sec

The second surface has:

The same surface recombination velocity as the first surface
Very low surface recombination velocity

1. A. B. Sproul, “Dimensionless solution of the equation describing the effect of surface recombination on carrier decay in
semiconductors”, Journal of Applied Physics, vol. 76, pp. 2851-2854, 1994.
2. K. L. Luke and Cheng, L. - J., “Analysis of the interaction of a laser pulse with a silicon wafer: Determination of bulk lifetime
and surface recombination velocity”, Journal of Applied Physics, vol. 61, pp. 2282-2293, 1987.
3. J. Brody and Rohatgi, A., “Analytical approximation of effective surface recombination velocity of dielectric-passivated
p-type silicon”, Solid-State Electronics, vol. 45, no. 9, pp. 1549 - 1557, 2001.

Transient Measurements
Add new comment

Transient lifetime measurements rely on the decay of carriers over time. Carriers are generated by a very short pulse of light and the
decay of the carrier density with time is measured. The longer the minority carrier lifetime the more slowly the carriers decay.

The animation shows the decay of carriers after a short light pulse. The time scale is much faster than shown here as
typical carrier lifetimes are at most in the millisecond range for silicon materials.

Illumination time and minority carrier density. The pulse of light is very short and the carrier density typically follows an exponential
decay. Move the mouse over the graph to see that the carrier density decay is a straight line on a log scale.
 carrier lifetime 1.0e-4 seconds

1e-5 1e-3

In the animation above the slider controls the minority carrier lifetime. Increasing the lifetime causes the carriers to decay more slowly.

Minority-Carrier Lifetime (no Auger correction) vs. Carrier Density

Quasi-Steady-State Lifetime Measurements

 Add new comment

Quasi-Steady-State lifetime measurements rely on the number of carriers present when a steady light is shone on a sample. It is
assumed that the intensity of the flash changes sufficiently slowly so that carrier populations in the sample are always at steady state.
This condition is met so long as the sample lifetime is less than the characteristic decay of the flash lamp.

The generation rate is determined by measuring the amount of light falling of the cell and then correcting for the reflectivity and the
absorption coefficient of silicon (ref).

Illumination at Reference Cell and Carrier Density

Minority-Carrier Lifetime (no Auger correction) vs. Carrier Density

QSS lifetime measurement

General Lifetime Measurements

Add new comment

Transient and quasi-steady-state photoconductance decay measurements of lifetime are the limiting cases of the general equation given
as 1.

In the case of transient measurements, the pulse of light is only to excite the carriers and is off during the actual measurement so that
G(t) = 0. For the QSS measurements, it is assumed that the light is varying very slowly so that d(Δn)/dt = 0.

The generalized form of the equation is useful when the flash and wafer have similar time constants.
Illumination at Reference Cell and Carrier Density

Minority-Carrier Lifetime (no Auger correction) vs. Carrier Density

1. H. Nagel, Berge, C., and Aberle, A. G., “Generalized analysis of quasi-steady-state and quasi-transient measurements of
carrier lifetimes in semiconductors”, Journal of Applied Physics, vol. 86, pp. 6218-6221, 1999.

Effect of Trapping on Lifetime Measurements

Add new comment

For energy levels created in the middle of the band gap by impurities (deep level "traps"), the probability of capturing electrons and
hole is similar, so traps at mid-gap operate as recombination centers. For energy levels near the band edges the probability of capturing
one carrier is much higher than the other. For example in the animation below, the probability of capturing electrons is much higher
than for capturing holes. Electrons tend to get trapped but they don't recombine, so shallow traps don't reduce the minority carrier
lifetime. However, the shallow level traps do cause significant problems with lifetime measurements

The trapped carrier has a mobile carrier of the opposite

polarity associated with it to preserve charge
neutrality. In the animation, the trapped electron has a
mobile hole associated with it that increases the wafer
conductivity. Since QSS measurements use the carrier
concentration under steady state conditions as a
measure of carrier lifetime, the extra carriers are
confused with a high lifetime material. The number of
shallow traps is usually quite low but the traps fill first,
so at very low injection levels the traps dominate the
QSS measurements and need to be corrected for. As
the number of carriers generated by light increases, the
percentage of carriers trapped is only a very small
section of the total and so the traps only have a very
small effect. Since the number of trapped carriers is
roughly constant with injection level, the traps can be
corrected for.

For transient measurements, the slow release of Animation showing the effect of trapping on lifetime measurements
carriers out of the traps that causes problems.

Correcting for the Effect of Traps

The presence of traps causes spuriously large lifetimes at low injection levels. The use of a bias light saturates the traps enabling the
measurement of the lifetime. In QSS measurements the bias light can either be an external source of infrared light (Ref) or part of the
same light used for the lifetime measurement 1. The level of bias light is adjusted to that it just saturates the traps 2.

1. R. A. Sinton and Cuevas, A., “Contactless determination of current–voltage characteristics and minority-carrier lifetimes in
semiconductors from quasi-steady-state photoconductance data”, Applied Physics Letters, vol. 69, pp. 2510-2512, 1996.
2. D. Macdonald, Sinton, R. A., and Cuevas, A., “On the use of a bias-light correction for trapping effects in photoconductance-
based lifetime measurements of silicon”, Journal of Applied Physics, vol. 89, pp. 2772-2778, 2001.

Electroluminescence
Add new comment

Electroluminescence relies on the same principle as a light emitting diode (LED). Current is fed into a solar cell (essentially a large
diode) and radiative recombination of carriers causes light emission. As an indirect bandgap semiconductor, most of the recombination
in silicon occurs via defects or Auger recombination. The amount of band-to-band recombination producing radiative emission is
relatively low. However, there is a small amount of radiative recombination that happens even in silicon and this signal can be sensed
using an external detector. The technique requires electrical contact and so can only be used once the metallization has been applied
and the cell is substantially complete. Electroluminsecence provides a wealth of data about the area related uniformity of solar cells
and modules. It is non descructive and relatively fast with measurement times of 1 s possible.
The luminescence signal of silicon peaks at 1150 nm corresponding to the energy of the bandgap 1.

Detectors

Electroluminescence has become increasingly popular with the advent of low cost silicon CCD arrays. They are similar to the ones
used for digital cameras but optimised for sensitivity in the near-infrared and cool to reduce thermal noise. As with digital cameras,
there are detectors with multiple mega-pixels resolutions of 2048 × 4096 pixels enabling high resolution images of entire modules. A
significant drawback of a silicon detector is that they have a poor response beyond 1000 nm due to the low absorption coefficient of
silicon. An alternative detector is arrays of InGaAs photodiodes. It has a much better response over the 1000 to 1300 nm wavelength
range enabling much faster data acquisition but with significantly higher cost. Resolution tends to be in the sub-megapixel range with
640 × 512 pixels common.

Quantum efficiency of silicon CCD detectors and InGaAs photodiode array. The low cost and resolution of the silicon CCD makes up
for the poor response over the 1000 -1200 nm region of interest.

Images

The key advantage as noted above is the ability of electroluminescence imaging an entire solar cell or module in a relatively short
space of time. The light output increases with the local voltage so that regions with poor contact show up as dark.
Electroluminscence image of a monocrystalline silicon wafer. The intensity of the light given off is proportional to the voltage, so
poorly contacted and inactive regions show up as dark areas. The microcrack and printing problem are not detectable with visual
inspection.

1. T. Fuyuki, Kondo, H., Yamazaki, T., Takahashi, Y., and Uraoka, Y., “Photographic surveying of minority carrier diffusion
length in polycrystalline silicon solar cells by electroluminescence”, Applied Physics Letters, vol. 86, p. 262108, 2005.

Introduction to Simulation
Add new comment

The equations that describe solar cell can be solved analytically or numerically. While the analytical equations are easier to solve by
hand and give great insight into cell operation, they become difficult to solve as more factors of cell operation are included. In the past,
it was common to re-write the equations slightly to simplify the solution to solve specific cases, however, such methods are time
consuming. For example. the solar cell modelling program on page XXX accurately models solar cell operation but only for a limited
number cases as noted on the page.

Computer speeds have increased (and memory for 2D cases) so it is now easier to write a general solver that applies to most cases.
There are an enormous number of packages for modelling semiconductor devices. However most of these packages either don't
consider light generation or only partially include light generation effects. Even for packages specifically designed for simulating solar
cells there exist a wide range of solvers both in house and commercially available for simulating solar cell operation. Most of these
packages have fairly similar basic module and it comes down to how fast they are, how easy they are to use and how many effects they
model.

The basic operation of a modelling program consists of setting up the model with user defined parameters, a generation of nodes to
solve, then iterating to produce a solution that is consistent with all the nodes.

Two device modelling programs used commonly within the photovoltaic community: PC1D for one-dimensional modelling and
DESSIS for two dimensional modelling

PC1D
 Add new comment

PC1D is the most commonly used of the commercially available solar cell modelling programs. Its success is based on its speed, user
interface and continual updates to the latest cell models. It is used to simulate new device performance and also for new users to
develop an understanding of device physics.

PC1D is now available free of charge from the University of NSW. The raw source code of PC1D is now available on SourceForge for
programmers and developers. PV Lighthouse has more PC1D resources including a batch file generator

Setup of a problem in PC1D. The user can alter the cell parameters and either use internal models or provide externally measured data.

Once the user has setup the basic model PC1D, generates a number of nodes to solve. The number of nodes is increased in parts of the
cell where the doping is changing and near surfaces.

Doping density as a function of distance from the front for the model described above. The cell is 300 micrometer thick but only the
first few microns are shown to show the extra notes in the emitter and the depletion region. Not shown are the extra nodes at the rear of
the device.

During simulation it is also possible to pause the program and examine the spatial distribution of carriers or field across the device at a
particular bias point.

PC1D works in all windows versions up to Windows 7. However, the help file is no longer supported directly by more recent versions
of Windows. A solution is to download the WinHlp32.exe from Microsoft. The help file containing a wealth of information about
photovoltaic device simulation and operation can also be downloaded (PC1D5_9.pdf)

Tandem Calculations
Add new comment
2 comment(s)

Description of Tandem Calc

TandemCalc is a Windows-standalone application for calculating maximum efficiency of tandem solar cells. The application was developed with MatLab
as the code base and requires installation of a 'MatLab Compiler Runtime' to operate. The MCR installation file is part of the downloadable package. To
run the application, after installing the MCR (which you should be taken through automatically when executing the package file), run the
TandemCalcv1.exe program.

Overview of Application Usage:

Parameters:
- Spectrum Type (dropdown box). The options are Blackbody and 'AM 1.5'. When selecting 'AM 1.5', if 'Concentration' is 1 then AM1.5G is used,
otherwise AM1.5D is used.
- Concentration
- Connection Type (dropdown box). The options are Independent and Series.
- Output Type. Currently hard-coded to download a text file (and display results on screen).
- Enter # of Bandgap Stacks: box that requires integer entry.
- Six columns of boxes for bandgap sweep by stack. Enter the bandgap sweeps in order from lowest bandgaps (Stack # 1) to highest bandgap (Stack # n,
where n=the value entered under 'Enter # of Bandgap Stacks').
After entering your parameters, click 'Calculate Efficiencies'. You will then see a progress bar that includes an estimate of how much time is remaining for
this calculation. When calculating for Independent, you may see a sequence of progress bars and the last progress bar may show a bogus % complete
above 100%, however, when that happens the program is nearly complete.

When the program is complete, you should see values under the fields at the bottom of the window.

Main Data File (data file that includes the efficiency % for each unique bandgap set calculated).

Help Data File (data file that includes the parameters you entered along with the max efficiency % and max efficiency bandgaps).
Max Efficiency % (maximum % efficiency based on the parameters you entered).
Max Efficiency Bandgaps (bandgaps related to the maximum-calculated efficiency %).

Reporting Defects and Enhancements

Please keep in mind that this software should be considered beta release and thus it is possible that despite the functional testing performed heretofore, you
may enter parameters that result in an error I did not see/resolve. When this occurs, please log basic information related to the error at the following shared
Google Doc in the 'Defects' tab.

Also if you have suggestions on how to improve this software application, please log it at the following shared Google Doc in the 'Enhancements' tab.

https://spreadsheets.google.com/spreadsheet/ccc?hl=en_US&key=taDB4wQEqv6-I53t-jBqjSg&hl=en_US#gid=0

2D Modeling
Add new comment

Two dimensional modelling is a far more complex operation than one dimensional modelling. Drawing devices in 2D is often a
challenge and the noding process is far more difficult. Finally, 2D modelling needs computers with a larger amount of memory space.
The requirements for 2D modelling is now being met by personal comuputers as memory costs have fallen and memory sizes
approaching a 1 GByte or more are possible. Historically 2D and 3D programs have been on Unix but they may move to PCs.

SEM and EBIC

Add new comment

SEM

Scanning electron microscope (SEM) pictures are useful for examining the fine structure of solar cells. Even in large area commercial
devices, for example, an SEM photograph can show the depth of the rear surface aluminium alloyed layer. Many of the cell features are
of the order of microns and so not possible to view with an optical microscope. An additional advantage of an electron microscope is
its higher depth of field. With an electron microscope it is possible to have the whole device in focus at once whereas in an optical
microscope at high magnification only parts of the device will be in focus at any time.

EBIC

Electron Beam induced current (EBIC) is useful for showing the electrically active areas of the device. To create an EBIC image an
electron beam is swept across the sample and the output current of the device measured. Those electrons that are collected by the
junction show up as bright regions while electrons uncollected remain dark. It is common to superimpose the EBIC images on SEM
picture to show more clearly show the location of the active areas.

In the figure below the cell has been cleaved to take a picture of the cell edge. The cleaved surface is unpassivated so most carriers will
recombine in a very short distance and only those created right at the junction edge will be collected. Hence the junction shows up as a
very bright line.

EBIC is also used in material characterisation. Altering the energy of the beam changes the depth and volume at of electrical excitation
and this can be used to characterise defects. By sweeping the beam across the surface of the device it is possible to identify the
electrically inactive areas such as grain boundaries. A numerical analysis of the induced current gives a measure of the diffusion length
1.
SEM picture of a cleaved PERL solar cell. Click on the image to change to an EBIC view. In the EBIC view the brightest area is
around the junction between the n-type diffusion and the p-type base. It shows that the emitter diffusion is quite shallow at 1-2 µm and
that it follows the surface topology. It also shows that the diffusion under the finger contact is 15 µm wide and 3 µm deep. (Photo: Tom
Puzzer)

EBIC image superimposed on an SEM image of an aluminium alloyed rear surface layer. Such layers are typically used as back-
surface-fields in commercial solar cells. The bright line of the silicon-aluminium juction shows that the aluminium layer is highly non-
uniform. Optimisation of the processing conditions led to much more uniform layers 2, 3.

1. R. Corkish, Luke, K. L., Altermatt, P. P., and Heiser, G., “Simulating Electron-Beam-Induced Current Profiles Across p-n
Junctions”, 16h European Solar Energy Conference. pp. 1590-1593, 2000.
2. C. B. Honsberg, Anwar, K. K., Mehrvarz, H. R., Cotter, J. E., and Wenham, S. R., “Dependence of aluminium alloying on
 solar cell processing conditions”, 13th Workshop on Crystalline Silicon Solar Cell Materials and Processes. 2003.
3. K. K. Anwar, “Aluminium Back Surface Field in Buried Contact Solar Cells”, University of New South Wales, 2000.

Four Point Probe Resistivity Measurements

Add new comment
2 comment(s)

A four point probe is a simple apparatus for measuring the resistivity of semiconductor samples. By passing a current through two
outer probes and measuring the voltage through the inner probes allows the measurement of the substrate resistivity. The doping
concentration can be calculated from the resistivity using the formulas discussed in the Appendices and the PV Lighthouse Resistivity
Calculator

The Measurement of Sheet Resistivity

The sheet resistivity of the top emitter layer is very easy to measure experimentally using a "four point probe". A current is passed
through the outer probes and induces a voltage in the inner voltage probes. The junction between the n and p -type materials behaves as
an insulating layer and the cell must be kept in the dark.

Use of a four point probe to measure the sheet resistivity of a solar cell.

Using the voltage and current readings from the probe:1

Where:

The typical emitter sheet resistivity of silicon solar cells lies in the range 30-100 Ω/□ .

In typical usage, the current is set to 4.53 mA so that the resistivity is simply the voltage reading in mV.

The Measurement of Bulk Resistivity

The measurement of bulk resistivity is similar to that of sheet resistivity except that a resistivity in cm-3 is reported using the wafer
thickness, t:

Where t is the layer/wafer thickness in cm.

The simple formula above works for when the wafer thickness less than half the probe spacing (t < s/2) 2. For thicker samples the
formula becomes:

Where s is the probe spacing.

The following calculator implements the above equation:

Four Point Probe Resistivity Calculator

Edit
Delete
Clone

Four Point Probe Resistivity Calculator

Voltage across inner probes, V = Volts Current in outer probes, I = Wafer thickness, W = Probe
spacing, s =
Sheet resistivity, ρsquare = Ω/square Wafer resistivity, ρ = Ω-cm Resistivity corrected for thickness, ρ =
Ω-cm

Measurement Problems

While simple in principle, there are experimental issues to take into account when using a four-point probe. In particular, the
application of a metal to a semiconductor forms a schottky diode rather than an ohmic contact. Very high or very low resistivity
samples require adjustment of the drive current to obtain a reliable reading. Samples with as cut or as lapped surfaces are easier to
measure than samples with polished surfaces

High Resistivity Samples

For high resistivity samples, the current is reduced so as not to have an excessively larger voltage at the contacts. It is recommended
that voltage on the inner probes be less than 100 mV/mm ( see reference)

Low Resistivity Samples

Low resistivity samples are usually much easier to measure as the contacts to the silicon are ohmic. For very low resistivity you will
have to increase the current to 45.3 mA and set the voltmeter to a lower scale. For very low resistivity samples the current passing
through the sample causes resistive heating that will, in turn, increase the measured resistivity.

1. F. M. Smits, “Measurement of sheet resistivities with the four-point probe”, Bell System Technical Journal, vol. 34, pp.
711-718, 1958.
2. D. Schroder, Semiconductor material and device characterization, 3rd editionrd ed. Piscataway NJ; Hoboken N.J.: IEEE
Press; Wiley, 2006.

Materials
Add new comment
General Properties of Silicon
Property Value
5 x 1022 cm-3
Atomic Density
5 x 1028 m-3
Atomic Weight 28.09

2.328 g cm-3
Density (ρ)
2328 kg m-3
Energy Bandgap (EG) 1.1242 eV

1 x 1010 cm-3
Intrinsic Carrier Concentration (ni) at 300K*
1 x 1016 m-3

8.6 x 109 cm-3

Intrinsic Carrier Concentration (ni) at 25°C*
8.6 x 1015 m-3

Lattice Constant 0.543095 nm

Melting Point 1415 °C

1.5 Wcm-1K-1
Thermal Conductivity
150 Wm-1K-1

Thermal Expansion Coefficient 2.6 x 10-6 K-1

3 x 1019 cm-3
Effective Density of States in the Conduction Band (NC)
3 x 1025 m-3

1 x 1019 cm-3
Effective Density of States in the Valence Band (NV)
1 x 1025 m-3

Relative Permittivity (εr) 11.7

Electron Affinity 4.05 eV

Electron Diffusion Coefficient (De) kT/q µe

Hole Diffusion Coefficient (Dh) kT/q µh

* updated values given in 1 2.

Properties of Silicon as a Function of Doping (300 K)

Carrier mobility is a function of carrier type and doping level. The values calculated here use the same formula as PC1D to fit values
given in 3 and 45 6. Lifetime as a function of doping is given on bulk lifetime.

Doping Level: cm-3

Electron mobility (µe) Diffusivity (De)

Majority carrier: cm2V-1s-1 cm2s-1

Minority carrier: cm2V-1s-1 cm2s-1

Hole Mobility (µh) Diffusivity (Dh)

Majority carrier: cm2V-1s-1 cm2s-1

Minority carrier: cm2V-1s-1 cm2s-1
 Approx. Resistivity (for doping > 1012)

n-type: ohm cm
p-type: ohm cm

1. A. B. Sproul, Green, M. A., and Zhao, J., “Improved value for the silicon intrinsic carrier concentration at 300 K”, Applied
Physics Letters, vol. 57, p. 255, 1990.
2. A. B. Sproul and Green, M. A., “Improved value for the silicon intrinsic carrier concentration from 275 to 375 K”, Journal of
Applied Physics, vol. 70, pp. 846-854, 1991.
3. C. H. Wang, Misiakos, K., and Neugroschel, A., “Minority-carrier transport parameters in n-type silicon”, IEEE Transactions
on Electron Devices, vol. 37, pp. 1314 - 1322, 1990.
4. W. R. Thurber, Mattis, R. L., Liu, Y. M., and Filliben, J. J., “Resistivity-Dopant Density Relationship for Boron-Doped
Silicon”, Journal of The Electrochemical Society, vol. 127, pp. 2291-2294, 1980.
5. W. R. Thurber, Mattis, R. L., Liu, Y. M., and Filliben, J. J., “Resistivity-Dopant Density Relationship for Phosphorus-Doped
Silicon”, Journal of The Electrochemical Society, vol. 127, pp. 1807-1812, 1980.
6. W. R. Thurber, Mattis,, Liu,, and Filliben, “The Relationship Between Resistivity and Dopant Density for Phosphorus- and
Boron-Doped Silicon”. U.S. Department of Commerce National Bureau of Standards, 1981.

Optical Properties of Silicon

The optical properties of silicon measure at 300K 1. While a wide range of wavelengths is given here, silicon solar cells typical only
operate from 400 to 1100 nm. There is a more up to date set of data in Green 2008 2. It is available in tabulated form from
pvlighthouse as text and in graphical format.

The data on this page is also available as an Excel spreadsheet.

Absorption coefficient of silicon in cm-1 as a function of the wavelength. Silicon is an indirect bandgap semiconductor so there is a
long tail in absorption out to long wavelengths. The data is graphed on a log scale. The drop in absorption at the band gap (around
1100 nm) is sharper than might first appear. See also absorption coefficient.

The absorption depth is the inverse of the absorption coefficient. An absoption depth of, for example, 1 um means that the light
intensity has fallen to 36% (1/e) of its original value. See also absorption coefficient.

Real and (negative) imaginary components of the refractive index for silicon at 300 K.

The reflectivity of a polished silicon wafer is determined from the complex refractive index.
The data for the above graph is given below. As noted above, it is also available in a much more convenient Excel spreadsheet.

W - wavelength
a - absorption coefficient
n - real component of refractive index
k - extinction coefficient
refractive index = n - ik
W(nm) a(/cm) n k
250 1.84E+06 1.694 3.666
260 1.97E+06 1.800 4.072
270 2.18E+06 2.129 4.690
280 2.36E+06 3.052 5.258
290 2.24E+06 4.426 5.160
300 1.73E+06 5.055 4.128
310 1.44E+06 5.074 3.559
320 1.28E+06 5.102 3.269
330 1.17E+06 5.179 3.085
340 1.09E+06 5.293 2.951
350 1.04E+06 5.483 2.904
360 1.02E+06 6.014 2.912
370 6.97E+05 6.863 2.051
380 2.93E+05 6.548 0.885
390 1.50E+05 5.976 0.465
400 9.52E+04 5.587 0.303
410 6.74E+04 5.305 0.220
420 5.00E+04 5.091 0.167
430 3.92E+04 4.925 0.134
440 3.11E+04 4.793 0.109
450 2.55E+04 4.676 0.091
460 2.10E+04 4.577 0.077
470 1.72E+04 4.491 0.064
480 1.48E+04 4.416 0.057
490 1.27E+04 4.348 0.050
500 1.11E+04 4.293 0.045
510 9.70E+03 4.239 0.039
520 8.80E+03 4.192 0.036
530 7.85E+03 4.150 0.033
540 7.05E+03 4.110 0.030
550 6.39E+03 4.077 0.028
560 5.78E+03 4.044 0.026
570 5.32E+03 4.015 0.024
580 4.88E+03 3.986 0.023
590 4.49E+03 3.962 0.021
600 4.14E+03 3.939 0.020
610 3.81E+03 3.916 0.018
620 3.52E+03 3.895 0.017
630 3.27E+03 3.879 0.016
640 3.04E+03 3.861 0.015
650 2.81E+03 3.844 0.015
660 2.58E+03 3.830 0.014
670 2.38E+03 3.815 0.013
680 2.21E+03 3.800 0.012
690 2.05E+03 3.787 0.011
700 1.90E+03 3.774 0.011
710 1.77E+03 3.762 0.011
720 1.66E+03 3.751 0.010
730 1.54E+03 3.741 0.009
740 1.42E+03 3.732 0.008
750 1.30E+03 3.723 0.008
760 1.19E+03 3.714 0.007
770 1.10E+03 3.705 0.007
780 1.01E+03 3.696 0.006
790 9.28E+02 3.688 0.006
800 8.50E+02 3.681 0.005
810 7.75E+02 3.674 0.005
820 7.07E+02 3.668 0.005
830 6.47E+02 3.662 0.004
840 5.91E+02 3.656 0.004
850 5.35E+02 3.650 0.004
860 4.80E+02 3.644 0.003
870 4.32E+02 3.638 0.003
880 3.83E+02 3.632 0.003
890 3.43E+02 3.626 0.002
900 3.06E+02 3.620 0.002
910 2.72E+02 3.614 0.002
920 2.40E+02 3.608 0.002
930 2.10E+02 3.602 0.002
940 1.83E+02 3.597 0.001
950 1.57E+02 3.592 0.001
960 1.34E+02 3.587 0.001
970 1.14E+02 3.582 0.001
980 9.59E+01 3.578 0.001
990 7.92E+01 3.574 0.001
1000 6.40E+01 3.570 0.001
1010 5.11E+01 3.566 -
1020 3.99E+01 3.563 -
1030 3.02E+01 3.560 -
1040 2.26E+01 3.557 -
1050 1.63E+01 3.554 -
1060 1.11E+01 3.551 -
1070 8.00E+00 3.548 -
1080 6.20E+00 3.546 -
1090 4.70E+00 3.544 -
1100 3.50E+00 3.541 -
1110 2.70E+00 3.539 -
1120 2.00E+00 3.537 -
1130 1.50E+00 3.534 -
1140 1.00E+00 3.532 -
1150 6.80E-01 3.530 -
1160 4.20E-01 3.528 -
1170 2.20E-01 3.526 -
1180 6.50E-02 3.524 -
1190 3.60E-02 3.522 -
1200 2.20E-02 3.520 -
1210 1.30E-02 3.528 -
1220 8.20E-03 3.516 -
1230 4.70E-03 3.515 -
1240 2.40E-03 3.513 -
1250 1.00E-03 3.511 -
1260 3.60E-04 3.510 -
1270 2.00E-04 3.508 -
1280 1.20E-04 3.507 -
1290 7.10E-05 3.506 -
1300 4.50E-05 3.504 -
1310 2.70E-05 3.503 -
1320 1.60E-05 3.501 -
1330 8.00E-06 3.500 -
1340 3.50E-06 3.498 -
1350 1.70E-06 3.497 -
1360 1.00E-06 3.496 -
1370 6.70E-07 3.495 -
1380 4.50E-07 3.493 -
1390 2.50E-07 3.492 -
1400 2.00E-07 3.491 -
1410 1.50E-07 3.490 -
1420 8.50E-08 3.489 -
1430 7.70E-08 3.488 -
1440 4.20E-08 3.487 -
1450 3.20E-08 3.486 -

1. M. A. Green and Keevers, M. J., “Optical properties of intrinsic silicon at 300 K”, Progress in Photovoltaics: Research and
Applications, vol. 3, pp. 189 - 192, 1995.
2. M. A. Green, “Self-consistent optical parameters of intrinsic silicon at 300 K including temperature coefficients”, Solar
Energy Materials and Solar Cells, vol. 92, pp. 1305–1310, 2008.

Bismuth trisulfide (Bi2S3)

Add new comment

Basic Info
Bismuth trisulfide (Bi2S3) has an orthorhombic crystal structure with 4 molecules per unit cell. Each molecule contains two bismuth
atoms and 3 sulfide atoms which add up to 20 atoms per unit cell. Bi2S3 occurs naturally in the form of bismuthinite, which has a lead-
gray streaked color with a metallic luster. The mineral is primarily found in Bolvia, Peru, central Europe, Austrailia, and the western
United States. Bismuthinite is mainly processed to obtain pure bismuth metal or other bismuth compounds, but has also been
researched for the use in photovoltaics. 1

Crystal Structure
 Fractional Coordinates Orthogonal Coordinates
Label Elmt x y z xor[Å] yor[Å] zor[Å]
1. T1 Bi 0.5165 0.2500 0.1748 -3.914 2.533 4.141
2. T1 Bi 0.9835 0.7500 0.6748 -6.849 8.734 8.119

3. T1 Bi 0.4835 0.7500 0.8252 -2.532 9.931 4.286

4. T1 Bi 0.0165 0.2500 0.3252 0.403 3.729 0.308

5. T2 Bi 0.6596 0.7500 0.4655 -3.971 6.133 6.169

6. T2 Bi 0.8404 0.2500 0.9655 -6.946 11.534 5.081

7. T2 Bi 0.3404 0.2500 0.5345 -2.475 6.331 2.258

8. T2 Bi 0.1596 0.7500 0.0345 0.501 0.930 3.347

9. T3 S 0.6230 0.7500 0.0575 -3.543 1.609 6.617

10. T3 S 0.8770 0.2500 0.5575 -7.157 7.076 6.051

11. T3 S 0.3770 0.2500 0.9425 -2.902 10.855 1.810

12. T3 S 0.1230 0.7500 0.4425 0.711 5.387 2.376

13. T4 S 0.7153 0.2500 0.3063 -5.681 4.163 5.333

14. T4 S 0.7847 0.7500 0.8063 -5.151 9.999 6.471

15. T4 S 0.2847 0.7500 0.6937 -0.764 8.301 3.095

16. T4 S 0.2153 0.2500 0.1937 -1.294 2.464 1.957

17. T5 S 0.4508 0.7500 0.3730 -2.127 4.923 4.839

18. T5 S 0.0492 0.2500 0.8730 -0.027 9.789 -0.410

19. T5 S 0.5492 0.2500 0.6270 -4.319 7.540 3.589

20. T5 S 0.9508 0.7500 0.1270 -6.418 2.674 8.838
The graph below shows peak intensities for Bi2S3:

PV Applications
Bi2Si3 thin films are prepared from several methods which include: Cathodic electrodeposition, anodic electro deposition, vacuum
evaporation, hotwall method, solution-gas interface, spray deposition, and chemical bath deposition. The most common method for
Bi2Si3 prepared films is chemical bath deposition. This is because it is simple, economic, and well suited for a large area of any
configuration. 2

Bi2Si3 thin films prepared from chemical bath deposition reach fill factors around 46.77% and a conversion efficiency of 0.089%. 3

Basic Parameters at 300 K

Crystal structure: Orthorhombic 4

Group of symmetry: Pnma 4

Number of atoms in 1 cm3: 3.99*1026 4

Unit cell volume: 501.6730 Å3 5

Atoms per unit cell: 20 4

Density: 6.807 g/cm3 4

Dielectric constant: ɛ(0)|| = 120 T=300K , at 1kHz 5

ɛ(0)⊥ = 38 T=300 K, E ⊥ c

ɛ(∞)|| = 13 T=300 K, E || c

ɛ(∞) ⊥ = 9 T=300, 90 K, E ⊥ c

Lattice constants: a = 11.305 Å 4

b = 3.981 Å
 c = 11.147 Å

Band Structure and carrier concentration

Graph of carrier concentration may be found in 6

Donors and Acceptors

Impurities: Pb, Cu, Fe, As, Sb, Se, Te 7

Electrical Properties
Basic Parameters of Electrical Properties

1.3 eV T=300 K, E ⊥ b:
Energy gap: 5
1.45 eV T=77 K, E ⊥ b

Intrinsic carrier concentration: : n = 3 × 1018 cm-3 T = 300 K : 5

Carrier mobility: : μn = 200 cm2/Vs T=300 K : 5

Hole mobility: : μh = 1100 cm2/Vs : 8

Intrinsic resistivity: : ρ = 105 Ω cm T=300K: 2

Electrical Conductivity: : σ = 10-6…10-7 Ω-1 cm-1 T=300 K: 5

Optical properties
Refractive indices:

λ = 589.3 nm 5

nα 1.315

nβ 1.900

nγ 1.670

Absorption coefficient: α = 104 cm-1 (In the order of) 2

Graph of optical transmittance data may be found in:9

Mechanical properties, elastic constants, lattice vibrations Basic Parameters

Hardness: 2-2.5 7

Cleavage planes: Perfect on (010) 7

Imperfect on (100), (110) Ralph2003
Developed at the University of Utah primarily by undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike
Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse, Jash Sayani, and Liz Lund also contributed. The work was fully supported by
the US National Science Foundation under the Materials World Network program award 1008302. These pages are a work in progress
and we solicit input from knowledgeable parties around the world for more accurate or additional information. Contact
earthabundantpv@eng.utah.edu with such suggestions.

1. G. S. Brady, Clauser, H. R., and Vaccari, J. A., Materials handbook: [foundation for the science of metallurgy; compares the
advantages and disadvantages of different manufacturing processes; comprehensive in approach]. New York [u.a.]: {McGraw-
Hill}, 2002.
2. a. b. c. J. D. Desai and Lokhande, C. D., “Chemical deposition of Bi2S3 thin films from thioacetamide bath”, Materials
Chemistry and Physics, vol. 41, no. 2, pp. 98 - 103, 1995.
3. R. S. Mane, Sankapal, B. R., and Lokhande, C. D., “Photoelectrochemical cells based on chemically deposited nanocrystalline
Bi2S3 thin films”, Materials Chemistry and Physics, vol. 60, no. 2, pp. 196 - 203, 1999.
4. a. b. c. d. e. f. R. Caracas and Gonze, X., “First-principles study of the electronic properties of A2B3 minerals, with A=Bi,Sb
and B=S,Se”, Physics and Chemistry of Minerals, vol. 32, no. 4, pp. 295 - 300, 2005.
5. a. b. c. d. e. f. g. O. Madelung, Semiconductors Data Handbook. Berlin: Springer, 2004, p. 691.
6. Citekey Park2011&nbsp; not found
7. a. b. c. J. R. Ralph, “Bismuthinite”. 2003.
8. F. V. Wald, Bullitt, J., and Bell, R. O., “Bi2S3 as a high Z material for γ-ray detectors”, IEEE Transactions on Nuclear Science,
vol. 22, no. 1, pp. 246 - 250, 1975.
9. P. K. Nair, Nair, M. T. S., Pathirana, H. M. K. K., Zingaro, R. A., and Meyers, E. A., “Structure and Composition of
Chemically Deposited Thin Films of Bismuth Sulfide and Copper Sulfide Effect on Optical and Electrical Properties”, Journal
of The Electrochemical Society, vol. 140, pp. 754–759, 1993.

CZTSSe

Authors: Scott McClary, Caleb Miskin, Rakesh Agrawal, Purdue University, Davidson School of Chemical Engineering.

As seen elsewhere on this site, CdTe and Cu(In,Ga)Se2 (CIGSe) solar cells have achieved remarkable efficiencies exceeding 20%.
However, elements such as indium and tellurium are rare, and cadmium is quite toxic, which may limit the implementation of these
devices at the terawatt scale. In recent years, Cu2ZnSnS4 (CZTS), Cu2ZnSnSe4 (CZTSe), and Cu2ZnSn(S,Se)4 (CZTSSe) have
emerged as replacement materials for thin-film PV due to promising optoelectronic properties and the use of nontoxic, earth-abundant
elements.

History of CZTSSe
CZTS was first investigated as a solar cell material in 1988 at Shinshu University in Japan. The researchers deposited thin films of
CZTS using sputtering and measured p-type conductivity, a direct band gap of 1.45 eV, and absorption coefficients exceeding 104 cm-1
in the visible range.1 They reported a solar cell the next year with a Voc of 0.165 V, though the Jsc was very low. In 1996, the first solar
cells based on CZTS and CZTSe absorbers using cadmium sulfide/zinc oxide window layers were reported with efficiencies exceeding
0.6%.23 Further research continued over the next decade, with improvements to processing conditions and window layers, resulting in
efficiencies nearing 7%.4 Around this time, research into solution processing of CZTSSe cells began.56 In 2013, IBM’s Watson
Research Center was able to push the efficiency of a lab-scale solar cell to 12.6% using a hydrazine-based solution process; this stands
as the world record for a CZTSSe-based solar cell today.7

Structures of CZTSSe
CZTS may crystallize in a zincblende-derived structure similar to other semiconductor materials such as silicon and CIGSe.
Conceptually, this kesterite CZTS cell is formed by substituting half of the In/Ga sites in CIGSe with Zn and the other half with Sn.
However, CZTS can have two other forms. The stannite structure has the same tetragonal coordination, but different symmetry due to
alternate placement of cations in the crystal lattice. The wurtzite-derived structure is a hexagonal close-packed array. All three have the
nominal Cu2ZnSnS4 stoichiometry, though the kesterite structure is used most often in solar cells.8 Note that the CZTSSe and CZTSe
compounds adopt analogous structures to those of CZTS.

Fabrication of CZTSSe Solar Cells

Thin-film solar cells with CZTSSe absorber layers are typically fabricated using a substrate configuration, as shown above. The layers
are all deposited on a conductive substrate (typically molybdenum-coated soda lime glass), which serves as the back contact. Some
 Crystal structures of semiconductor materials. (Courtesy of Dr. Bryce Walker)

cells are also fabricated using a superstrate configuration in which all of

the layers are sequentially deposited on a transparent conducting
surface, often tin-doped indium oxide (ITO) or fluorine-doped tin oxide
(FTO) coated on glass. The cell is then flipped over and illuminated
through the superstrate for operation.

Fabrication methods for CZTSSe solar cell absorber layers can be

broadly classified into two different categories: those that are vacuum-
based and those that are solution-based (i.e. non-vacuum based).
Vacuum-based processes include (but are not limited to) sputtering,
evaporation, and atomic layer deposition, while solution-based
processes include nanocrystal “inks”, molecular precursors,
electrodeposition, sol-gel, and spray pyrolysis.

To fabricate a typical substrate CZTS solar cell, one starts with a

substrate with a metallic back contact, such as Mo. Then, the CZTS Typical architecture of a CZTSSe-based solar cell.
material is deposited on a substrate using one of the processes above. It
is often annealed at high temperatures (~500-600 °C) in gaseous
atmospheres such as sulfur, selenium, hydrogen sulfide, and argon.
Then, a thin n-type window layer (typically CdS) is deposited by chemical bath deposition or sputtering, followed by sputtering of a
hole-blocking layer of intrinsic ZnO and a transparent conducting layer of ITO. Patterned metal grids (e.g. nickel/aluminum) are
evaporated to complete the device. Individual cells are often defined by mechanical scribing down to (but not through) the Mo-back
contact. The solar cell can then be tested under simulated sunlight to evaluate its performance.

CZTSSe Limitations and Future Research Directions

CZTSSe is an inherently challenging material to work with due to its complexity. Forming a phase-pure material is difficult, as its
thermodynamic stability window is quite narrow; additionally, Sn-compounds are volatile, and CZTS breaks down at high
temperatures. Many secondary phases (particularly Cu2SnS3 and ZnS) may coexist in the absence of carefully controlled reaction
conditions.9 Additionally, the stannite and kesterite phases have similar formation energies, so it is likely that a mixture of both can
form during synthesis. To complicate matters further, it is difficult to distinguish between these unwanted phases through traditional
characterization techniques such as X-ray diffraction.

Crystal structure defects can form quite easily in CZTS. One of the most common defects is an antisite defect – Cu cations can occupy
Zn sites and vice versa due to their similar ionic radii. These defects are partially compensated by targeting a Cu-poor, Zn-rich
composition during materials synthesis. However, these defects are not eliminated completely and can give rise to band tail states –
 Sample schematic of CZTS solar cell production: Nanocrystal-based processing

essentially, these are trap states that arise due to electrostatic potential fluctuations in the material. These states lower the Voc
considerably and severely limit the cell’s efficiency.1011

Current research is focused primarily on understanding and mitigating the defects within the CZTSSe crystal structure. One potential
method is to replace either the Cu or Zn with another atom of a significantly different size (e.g. Cu with Ag or Zn with Ba) to prevent
formation of antisite defects, though this could change other material properties such as the conductivity type or crystal structure.1213
Another critical area of research focuses on optimizing the solar cell structure, particularly through use of alternative n-type layers that
may have energy bands that align more favorably with CZTS.14

Further Reading
Copper Zinc Tin Sulfide-Based Thin Film Solar Cells, 1st ed.; Ito, K., Ed.; John Wiley & Sons, Ltd., 2015.

Mitzi, D. B.; Gunawan, O.; Todorov, T. K.; Wang, K.; Guha, S. The Path towards a High-Performance Solution-Processed Kesterite
Solar Cell. Sol. Energy Mater. Sol. Cells 2011, 95 (6), 1421–1436.

Zhou, H.; Hsu, W.-C.; Duan, H.-S.; Bob, B.; Yang, W.; Song, T.-B.; Hsu, C.-J.; Yang, Y. CZTS Nanocrystals: A Promising Approach
for next Generation Thin Film Photovoltaics. Energy Environ. Sci. 2013, 6 (10), 2822–2838

Acknowledgement
The material on this page was developed as part of the 2016 Hands-On PV Experience (HOPE) Workshop at NREL.

http://www.nrel.gov/pv/hands-on-photovoltaic-experience.html

1. K. Ito and Nakazawa, T., “Electrical and Optical Properties of Stannite-Type Quaternary Semiconductor Thin Films”,
Japanese Journal of Applied Physics, vol. 27, no. Part 1, No. 11, pp. 2094 - 2097, 1988.
2. H. Katagiri, Sasaguchi, N., Hando, S., Hoshino, S., Ohashi, J., and Yokota, T., “Preparation and evaluation of Cu2ZnSnS4 thin
films by sulfurization of E-B evaporated precursors”, Solar Energy Materials and Solar Cells, vol. 49, pp. 407 - 414, 1997.
3. T. Magorian Friedlmeier, Wieser, N., Walter, T., Dittrich, H., and Schock, H. W., “Heterojunctions based on Cu2ZnSnS4 and
Cu2ZnSnSe4 thin films”, in 14th European PVSEC, 1997.
4. H. Katagiri et al., “Development of CZTS-based thin film solar cells”, Thin Solid Films, vol. 517, pp. 2455–2460, 2009.
5. Q. Guo, Hillhouse, H. W., and Agrawal, R., “Synthesis of Cu2ZnSnS4 Nanocrystal Ink and Its Use for Solar Cells”, Journal of
the American Chemical Society, vol. 131, no. 33, pp. 11672 - 11673, 2009.
6. C. Steinhagen, Panthani, M. G., Akhavan, V., Goodfellow, B., Koo, B., and Korgel, B. A., “Synthesis of Cu2ZnSnS4
nanocrystals for use in low-cost photovoltaics”, Journal of the American Chemical Society, vol. 131, pp. 12554–12555, 2009.
 7. W. Wang et al., “Device characteristics of CZTSSe thin-film solar cells with 12.6% efficiency”, Advanced Energy Materials,
vol. 4, 2014.
8. S. Chen, Gong, X. G., Walsh, A., and Wei, S. - H., “Crystal and electronic band structure of Cu2ZnSnX4 (X= S and Se)
photovoltaic absorbers: first-principles insights”, Applied Physics Letters, vol. 94, p. 41903, 2009.
9. M. Kumar, Dubey, A., Adhikari, N., Venkatesan, S., and Qiao, Q., “Strategic review of secondary phases, defects and defect-
complexes in kesterite CZTS–Se solar cells”, Energy Environ. Sci., vol. 8, no. 11, pp. 3134 - 3159, 2015.
10. T. Gokmen, Gunawan, O., Todorov, T. K., and Mitzi, D. B., “Band tailing and efficiency limitation in kesterite solar cells”,
Applied Physics Letters, vol. 103, p. 103506, 2013.
11. C. J. Hages, Carter, N. J., and Agrawal, R., “Generalized quantum efficiency analysis for non-ideal solar cells: Case of Cu 2
ZnSnSe 4”, Journal of Applied Physics, vol. 119, no. 1, p. 014505, 2016.
12. C. J. Hages, Koeper, M. J., and Agrawal, R., “Optoelectronic and material properties of nanocrystal-based \{CZTSe\}
absorbers with Ag-alloying”, Solar Energy Materials and Solar Cells, vol. 145, Part 3, pp. 342 - 348, 2016.
13. F. Hong, Lin, W., Meng, W., and Yan, Y., “Trigonal Cu 2 -II-Sn-VI 4 (II = Ba, Sr and VI = S, Se) quaternary compounds for
earth-abundant photovoltaics”, Phys. Chem. Chem. Phys., vol. 18, no. 6, pp. 4828 - 4834, 2016.
14. K. Sun et al., “Over 9% Efficient Kesterite Cu 2 ZnSnS 4 Solar Cell Fabricated by Using Zn 1- x Cd x S Buffer Layer”,
Advanced Energy Materials, vol. 6, no. 12, p. 1600046, 2016.

CdS
Add new comment

Basic Information:
Molecular Weight: 144.48 gm (77.81% Cd, 22.19% S)

CdS is a naturally occurring mineral that shows up with two different crystal structures as greenockite and hawleyite.

Greenockite [1] [2]:

Named after Lord Greenock (1783-1859), the owner of the land where it was discovered.
Discovered in 1840 in Greenock, Scotland.
Found in Traprock cavities and ore veins.
An uncommon mineral of sulfide deposits

Density (calculated): 4.824 g/cm3

Member of: Wurtzite Group

Lustre: Adamantine, Resinous

Color: Yellow to red

Streak: Orange-yellow to brick red

Hardness (Mohs): 3-3 ½

Tenacity: Brittle

Cleavage: Distinct/Good

Distinct on {1122}, imperfect on {0001}

Fracture: Conchoidal

Hawleyite [4]:

Named after Professor James Edwin Hawley (1897-1965), a Canadian mineralogist at Queen's University.
Discovered in the Hector-Calumet mine in Canada.
Found as a coating on fine-grained sphalerite and siderite.

Density (calculated): 4.87 g/cm3

Member of: Sphalerite Group

Lustre: Metallic

Color: Bright Yellow

Streak: Light Yellow

Hardness (Mohs): 2½ - 3

Diaphaneity: Opaque

Cleavage: None

Common Uses [3]:

Color for soaps, textiles, paper and rubber.

Used in printing inks, ceramic glazes, fireworks, x-ray fluorescent screens and in body temperature detectors.
Provides stability against oxidation and UV radiation in some industrial products.
Used to create pigments with colors ranging from yellow to deep red that have high tolerance to heat and light. These pigments
are mostly used in coloring plastics, ceramics and paints.
Used in small amounts in electric batteries and other electric components.

Crystal Structure
Greenockite [2]: Dihexagonal Pyramidal

Crystal System: Hexagonal

Cell Dimensions: a = 4.136Å, c = 6.713Å, Z = 2; Den(Calc)= 6.06

Axial Ratios: a:c = 1 : 1.623

Unit Cell Volume: 99.45 Å3

Space Group: P63mc

Morphology: Crystals hemi hemimorphic pyramidal. Earthy coatings (sphalerite).

Twinning: Twin plane {1122} rare forming trillings.

X-ray Diffraction [6]:

X-RAY WAVELENGTH: 1.541838

MAX. ABS. INTENSITY / VOLUME**2: 122.0494781

2-THETA INTENSITY D-SPACING H L K

24.82 59.21 3.5824 1 0 0
26.52 42.68 3.3574 0 0 2

28.20 100.00 3.1607 1 0 1

36.64 25.83 2.4497 1 0 2

43.72 45.25 2.0683 1 1 0

47.87 45.29 1.8982 1 0 3

50.93 6.79 1.7912 2 0 0

51.87 33.85 1.7610 1 1 2

52.85 15.27 1.7307 2 0 1

54.62 2.33 1.6787 0 0 4

 58.33 5.62 1.5803 2 0 2
60.88 2.38 1.5201 1 0 4

66.83 14.14 1.3985 2 0 3

69.33 4.60 1.3540 2 1 0

70.94 10.80 1.3273 2 1 1
72.44 5.12 1.3034 1 1 4

75.54 9.29 1.2575 1 0 5

75.66 4.52 1.2557 2 1 2
80.33 5.66 1.1941 3 0 0
83.34 13.69 1.1585 2 1 3

86.40 6.04 1.1251 3 0 2

Hawleyite [5]: Hextetrahedral

Crystal System: Isometric

Cell Dimensions: a = 5.818 Å, Z = 4

Unit Cell Volume: 196.93Å3

Space Group: F4 3m

Morphology: Fine grained powdery coatings

X-ray Diffraction [7]:

X-RAY WAVELENGTH: 1.541838

MAX. ABS. INTENSITY / VOLUME**2: 213.1595047

2-THETA INTENSITY D-SPACING H L K Multiplicity

26.54 100.00 3.3590 1 1 1 4

30.74 23.16 2.9090 2 0 0 6

44.02 51.93 2.0570 2 2 0 12

52.14 39.02 1.7542 3 1 1 12

54.65 5.46 1.6795 2 2 2 4

64.01 7.15 1.4545 4 0 0 6

70.56 13.46 1.3347 3 3 1 12

72.68 5.95 1.3009 4 2 0 24

80.95 12.96 1.1876 4 2 2 12

87.03 6.95 1.1197 5 1 1 12

87.03 2.32 1.1197 3 3 3 4

PV Applications [10]:

Polycrystalline CdS thin films have properties that are suitable for solar cell fabrication. They have good optical transmittance, a wide
band-gap and good electrical properties. CdS thin films also have a high absorption coefficient, electron affinity, low resistivity, and
easy ohmic contact, also making them suitable for solar cell applications. Because of these qualities, a lot of research is being done on
direct band-gap thin films, especially because of its intermediate band-gap. CdS thin films are usually grown by chemical bath
deposition because it is a more simple technique that produces quality films. Other techniques are sometimes used such as
electrodeposition, screen printing, sputtering, and spray pyrolysis. The thin films with a and b phase depending on the deposition
conditions. a-CdS grows with columnar structure along the c-axis perpendicular to the substrate so there are no grain boundaries
parallel to the junction. CdS is used as a window electrode because of its stability, reasonable conversion efficiency, and the deposition
technique is low-cost.

A diagram of CdS acts as a window layer can be found from reference [9].

For Cu(In,Ga)Se2 (CIGS) solar cells which contains chemical-bath-deposited CdS have attained a record efficiency of 20.3%. In this
case, the CdS is the buffer layer. This instance had the following parameters [11]:

Open-circuit voltage: 730 mV

Fill Factor: 77.7%

Shunt resistance: 880 Ω cm2

Short-circuit current density: 35.7 mA/cm2

Electron current density: 4.2E-11 A/cm2

Specific contact resistance: 0.23 Ω cm2

Electrode polarization: 880 Ω cm2

Photocurrent density: 35.6 mA/cm2

Diode ideality factor: 1.38

Cell Area: 0.50 cm2

Cell setup:

soda-lime glass (3 mm)

sputtered molybdenum (500–900 nm)
CIGS (2.5–3.0 µm)
chemical bath deposited CdS buffer layer (40–50 nm)
sputtered undoped ZnO (50–100 nm)
sputtered aluminium doped ZnO (150–200 nm)
nickel/aluminium-grid

Basic Parameters at 300 K [14]

Heat Capacity: 53.97 + 3.77.10-3T J mol-1 K-1

Melting Temperature: 1750 K, 1405(10) ◦C

Volume Compressibility: 1.586.10-7 bar-1

Band Structure:

Band Gap[12]: 2.42 eV

CdS is a direct gap semiconductor with the smallest energy gap in the center of the Brillouin Zone. The topmost valence band is split
due to crystal field and spin-orbit coupling into three spin-degenerate states. Exciton states formed with holes in these valence band
states are denoted A, B and C exciton, respectively.

Calculated energies of symmetry points of the band structure (relative to the top of the valence band E(Γ5V)):

A graph of the band structure from an empirical tight binding model [18K] compared to the pseudopotential band structure of [67B]
can be found from reference [14] (Energy bands corresponding to the Cd 4d states and spin-orbit coupling are not considered in this
calculation).

Carrier concentration of undoped CdS [13]: 1.19 × 1019 cm−3

Temperature Dependences:

A graph of the energy gap of the A gap vs. temperature, experimental points can be found from reference [14].

Effective Masses and Density of States [14]

conduction band, effective masses

From experiments the conduction band mass can beassumed to be nearly isotropic.
mn 0.25 m0 T= 300 K thermoelectric power
0.2...0.16 m0 T= 25...700 K mobility analysis, OMS, PPS
valence band, effective masses
0.7 (1) m0 T= 1.6 K exciton magneto-absorption
5 m0

Basic Parameters of Electrical Properties [14]

electromechanical coupling factors

k31 0.119 T= 298 K ultrasound resonance
k33 0.262
k15 0.188
kt 0.154

Mobility and Hall Effect [14]

≥10000 cm2/Vs T= 30...40 K peak mobilities in ultrapure crystals

160 cm2/Vs T= 300 K In-doped, n=5•1019 cm-3

Optical properties [14]

refractive index and birefringence

Δn (=nǁc - n ‫ ﻠ‬c)
n‫ﻠ‬c nǁc Δn λ [μm] T [K]
2.573 2.586 0.55 293 prism
2.479 2.496 0.61
2.417 2.434 0.69
2.358 2.375 0.85
2.296 2.312 1.50
2.281 1.678 2 293 interference
2.258 1.662 6
2.187 1.408 14
2.051 18
1.880 24

temperature dependence of the refractive index

T [°C] λ [μm]
(1/n ‫)ﻠ‬dn ‫ﻠ‬/dT 26.8(3)٠10-6 K-1 35...80 10.3 interference
(1/nǁ)dnǁ/dT 27.8(2)٠10-6 K-1
d(Δn)/dT 3.07(12)٠10-6 K-1

Graphs of numerically calculated spectral dependence of the absorption coefficient α and normal-incidence reflectivity R for E T c (a)
and E ǁ c (b) at 300 K can be found from refernece [14].
Elastic Constants [15]: sE11, sE12, sD33, sE55, sD33, sD55

Phonon Frequencies [14]

References

[1] WebMineral, “Greenockite Mineral Data”, http://webmineral.com/data/Greenockite.shtml.

[2] J. Ralph, I. Chau, mindat Available at: <http://mindat.org> (2012).

[3] United States Department of Labor, “Chemical Identification, Production, and use

of Cadmium,” Occupational Safety and Health Administration, April 23rd 1993 . http://

www.osha.gov/pls/oshaweb/owadisp.show_document?p_id=818&p_table=PREAMBLES

[4] WebMineral, “Hawleyite Mineral Data”, http://webmineral.com/data/Hawleyite.shtml.

[5] J. Ralph, I. Chau, mindat, (2012). http://www.mindat.org/min-1834.html

[6] Downs R T (2006) The RRUFF Project: an integrated study of the chemistry,

crystallography, Raman and infrared spectroscopy of minerals. Program and Abstracts of the

19th General Meeting of the International Mineralogical Association in Kobe, Japan. O03-13

http://rruff.info/Greenockite/R090045

[7] Downs R T (2006) The RRUFF Project: an integrated study of the chemistry,

crystallography, Raman and infrared spectroscopy of minerals. Program and Abstracts of the

19th General Meeting of the International Mineralogical Association in Kobe, Japan. O03-13

http://rruff.geo.arizona.edu/AMS/result.php?mineral=Hawleyite

[8] S. Schorn, “Hawleyite,” Mineralienatlas, 2012. http://www.mineralatlas.eu/lexikon/

index.php/MineralData?mineral=Hawleyite

[9] Y. Wakchaure, “Solar Cells,” www.nd.edu/~gsnider/EE698A/Yogesh_Solar_cells.ppt

[10] G. Sasikala, P. Thilakan, C. Subramanian, “Modifcation in the chemical bath deposition

apparatus, growth and characterization of CdS semiconducting thin films for photovoltaic

applications,” Solar Energy Materials & Solar Cells, vol. 62, no. 2000, pp. 275-293, October

1999. http://144.206.159.178/ft/957/11986/243188.pdf

[11] P. Jackson, et al., “New world record efficiency for Cu(In,Ga)Se2 thin-film solar cells

beyond 20%,” Progress in Photovoltaics: Research and Applications,vol. 19, no. 7, pp. 894-897,

November 2011.

http://onlinelibrary.wiley.com/doi/10.1002/pip.1078/full

[12] J. Britt, Thin‐film CdS/CdTe solar cell with 15.8% efficiency. Applied Physics

Letters, 62, 2851-2852 (1993). http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?

arnumber=4880703&abstractAccess=no&userType=inst

[13] S. Kose, et al., ‘Optical characterization and determination of carrier density of

ultrasonically sprayed CdS:Cu films,” Applied Surface Science, vol. 256. No. 13, April 2010.

http://www.sciencedirect.com/science/article/pii/S0169433210001820

[14] Madelung, O. (2004). Semiconductors: Data handbook. (3rd ed.). Springer.

[15] I.B. Kobiakov, “Elastic, piezoelectric and dielectric properties of ZnO and CdS single

crystals in a wide range of temperatures,” Solid State Communications, vol. 53, no. 3, July 1980.

http://www.sciencedirect.com/science/article/pii/0038109880905025

The development of these pages on photovoltaic materials’ properties was carried out at the University of Utah primarily by
undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse,
Jash Sayani, and Liz Lund also contributed. The work was fully supported by the US National Science Foundation under the Materials
World Network program award 1008302. These pages are a work in progress and we solicit input from knowledgeable parties around
the world for more accurate or additional information. Contact earthabundantpv@eng.utah.edu with such suggestions. Neither the
University of Utah nor the NSF guarantee the accuracy of these values.

CdSe (wurtzite)
Add new comment

Crystal Structure
 Fractional Coordinates Orthogonal Coordinates
Label Elmt x y z xor[Å] yor[Å] zor[Å]
1. T1 Cd 0.6667 0.3333 0.0000 -1.738 -0.109 1.768

2. T1 Cd 0.3333 0.6667 0.5000 0.824 3.199 2.747

3. T1 Cd 0.6667 0.3333 1.0000 -1.415 6.859 2.513

4. T2 Se 0.6667 0.3333 0.3750 -1.617 2.504 2.048

5. T2 Se 0.3333 0.6667 0.8750 0.945 5.812 3.027

Theoretical diffraction data using a Cu Kα monochromatic source.

(m is the multiplicity and N is the maximum number of flexions)

ref no. h k l d(hkl) 2-Theta Intensity I/Imax m N

[1] 0 1 0 3.72261 23.8828 2.01941e-001 100.0 6 1
[2] 0 0 2 3.50760 25.3704 1.19461e-001 59.2 2 4

[3] 0 1 1 3.28831 27.0935 1.53355e-001 75.9 12 2

[4] 0 1 2 2.55288 35.1217 7.60123e-002 37.6 12 5

[5] 1 1 0 2.14925 42.0016 1.76915e-001 87.6 6 2

[6] 0 1 3 1.98014 45.7830 1.76516e-001 87.4 12 10

[7] 0 2 0 1.86131 48.8904 2.81154e-002 13.9 6 4

[8] 1 1 2 1.83259 49.7080 1.10330e-001 54.6 12 6

[9] 0 2 1 1.79906 50.6993 2.62001e-002 13.0 12 5

[10] 0 0 4 1.75380 52.1043 9.71422e-004 0.5 2 16

[11] 0 2 2 1.64416 55.8710 1.92128e-002 9.5 12 8

[12] 0 1 4 1.58655 58.0890 1.00406e-003 0.5 12 17

[13] 0 2 3 1.45629 63.8640 6.32987e-002 31.3 12 13

[14] 1 2 0 1.40701 66.3825 2.18411e-002 10.8 12 5

[15] 1 2 1 1.37954 67.8819 2.09435e-002 10.4 24 6

[16] 1 1 4 1.35881 69.0623 2.22018e-003 1.1 12 18

[17] 0 1 5 1.31289 71.8441 4.43706e-002 22.0 12 26
[18] 1 2 2 1.30587 72.2909 1.74150e-002 8.6 24 9
[19] 0 2 4 1.27644 74.2329 4.40427e-004 0.2 12 20

[20] 0 3 0 1.24087 76.7395 2.86434e-002 14.2 6 9

[21] 1 2 3 1.20560 79.4187 6.73141e-002 33.3 24 14

[22] 0 3 2 1.16983 82.3607 2.45689e-002 12.2 12 13
[23] 0 0 6 1.16920 82.4144 4.08848e-003 2.0 2 36

[24] 0 2 5 1.12039 86.8627 2.73303e-002 13.5 12 29

[25] 0 1 6 1.11547 87.3415 5.39674e-003 2.7 12 37
[26] 1 2 4 1.09748 89.1489 5.54168e-004 0.3 24 21
[27] 2 2 0 1.07463 91.5755 1.92663e-002 9.5 6 8

[28] 1 3 0 1.03247 96.4951 8.94575e-003 4.4 12 10

[29] 2 2 2 1.02749 97.1194 1.82529e-002 9.0 12 12

[30] 1 1 6 1.02706 97.1730 1.82418e-002 9.0 12 38

[31] 1 3 1 1.02146 97.8876 8.99105e-003 4.5 24 11

[32] 0 3 4 1.01296 98.9977 9.71735e-004 0.5 12 25

PV Applications

Basic Parameters at 300 K

Crystal structure: Wurtzite 1

Group of symmetry: P6_3mc 1
 1
Number of atoms in 1 cm3: 3.56*1026

Unit cell volume: 112.2547 Å3 1

Atoms per unit cell: 4 1

Auger recombination coefficient C:

Debye temperature: 181.7 K T=0K 2

Density: 5.81 g/cm3 T = 300 K 2

Dielectric constants: ɛ(0)|| = 10.16 T = 300 K 2

9.29 T = 100 K

ɛ(0)⊥ = 9.29 T = 300 K

9.15 T = 100 K

ɛ(∞)|| = 6.30 T = 300 K

6.30 T = 100 K

ɛ(∞) ⊥ = 6.20 T = 300 K

6.20 T = 100 K

Effective electron density:

Effective electron masses: mn = 0.12 m0 2

Effective hole density:

Effective hole masses: mhh = 2.14 m0 [111] direction 2

0.9 m0 [100] direction

1.7 m0 [110] direction

mlh = 0.16 m0 [111] direction

0.18 m0 [100] direction

0..16 m0 [110] direction

mso = 0.12 m0 split-off band

Lattice constants: a=b= 4.2985 Å 1

c = 7.0152 Å

Band structure and carrier concentration

Temperature Dependences

Temperature dependence on annealed CdSe thin films 3

Activation energy Ea
Crystalite size Band Gap 'Eg' Electrical resistivity
Thin Films (eV)
(Å) (eV) (Ωcm)
HR LR
As-
deposited
40 2.3 3.25 × 105 0.86 0.34

373 K 60 2.0 9.58 × 104 0.79 0.31

 473 K 80 1.8 5.38 × 104 0.72 0.27

573 K 120 1.8 8.23 × 103 0.69 0.18

673 K 180 1.7 1.17 × 103 0.65 0.16

HR: High temperature region LR: Low temperature region

Plot of absorbance vs. wavelength for CdSe thin films can be found in:3

Plot of log ρ vs. (1000/T) of CdSe thin films can be found in: 3

Electrical Properties
Basic Parameters of Electrical Properties

Energy gap: 1.74 eV T = 300 K

2

Energy spin-orbital splitting: 470 meV 2

Intrinsic carrier concentration: 6*1013 cm-3 T = 800 K 2

6*1016 cm-3 T = 1300 K

Carrier mobility: μn = 660 cm2/Vs T = 300 K 2

5000 cm2/Vs T = 80 K

200 cm2/Vs T = 800 K

Electron drift mobility: μdr,n = 720 cm2/Vs T = 300 K 2

Hole Hall mobility: μH,p = 40 cm2/Vs T = 300 K 2

Thermal properties
Heat capacity: Cp [J mol-1 K-1] = 48.46 + 5.87-3 *T – 58154*T-2 2

Thermal conductivity: κL = 0.09 W cm-1 K-1 2

Mechanical properties, elastic constants, lattice vibrations

Basic Parameters

Density: 5.81 g/cm3 T = 300 K

2

Elastic Constants

Elastic Constants: c11 = 74.6 Gpa T = 300 K 2

 c12 = 46.1 Gpa

c33 = 81.7 Gpa

c44 = 13.0 Gpa

c66 = 14.3 Gpa

Developed at the University of Utah primarily by undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike
Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse, Jash Sayani, and Liz Lund also contributed. The work was fully supported by
the US National Science Foundation under the Materials World Network program award 1008302. These pages are a work in progress
and we solicit input from knowledgeable parties around the world for more accurate or additional information. Contact
earthabundantpv@eng.utah.edu with such suggestions.

1. a. b. c. d. e. f. Y. - N. Xu and Ching, W., “Electronic, optical, and structural properties of some wurtzite crystals”, Physical
Review B, vol. 48, no. 7, pp. 4335 - 4351, 1993.
2. a. b. c. d. e. f. g. h. i. j. k. l. m. n. o. O. Madelung, Semiconductors Data Handbook. Berlin: Springer, 2004, pp. 815-835.
3. a. b. c. R. B. Kale and Lokhande, C. D., “Influence of air annealing on the structural, optical and electrical properties of
chemically deposited CdSe nano-crystallites”, Applied Surface Science, vol. 223, no. 4, pp. 343 - 351, 2004.

CuInSe2
Basic Info
Melting temperature 1: 1260K

Crystal Structure 1: Chalcopyrite lattice

Space Group: I42d

Z=4

a = 5.781 Å

c = 11.552 Å

c/a = 2

A graph showing peak intensities (intensity vs. 2-theta) can be found at reference 2

The unit cell can be seen in reference 3.

PV Applications:
Conversion efficiencies of 17.8% have been reached for vapor-deposited CIS cells

The chalcopyrite structure of ternary I-III-VI2 compounds have high absorption coefficients making CIS well-suited for solar cells

CuInSe2 is the absorber layer

Basic Parameters at 300 K:

Density 1: 5.77g/cm3

Dielectric Constants 1: ɛ(0) = 15.2 E ǁ c

ɛ(0) = 16.0 E⊥c

ɛ(∞) = 8.5 E ǁ c
ɛ(∞) = 9.5 E⊥c

Band Structure and carrier concentration 1:

Single Crystals:

p-type

μn (6±3) cm2/V-s

μp (3.1±0.15) cm2/V-s

p 5x1015 cm-3

Bandgap of 1.02 eV

A graph of the band structure of CuInSe2 can be seen at reference 1.

Temperature Dependences:

Energy Gap Narrowing at High Doping Levels 1:

Energy Gap

Eg.dir (1.010 ± 0.001) eV single crystal

dEg/dT -1.1x10-4 eV/K T= 100…300K

Effective Masses and Density of States

Effective Masses 1:

The Fermi surface of electrons is spherical and located at the center of the Brillouin zone.

Cyclotron mass value m*c 0.0822 m0

Eff. Mass at minimum of conduction band m*0 0.0772 m0

mn = mp 0.087 m0

Donors and Acceptors 4:

The acceptor levels are at 0.020 eV and 0.028 eV above the valence band and the donor levels are at 0.012 eV and 0.18 eV below the
conduction band.

Electrical Properties
Exciton binding energy 1: 4.3meV T = 2K

Splitting energies (at Γ)

Δso 0.8 eV

Δcf +0.006 eV T = 77 K

Δso 0.23 eV T = 77 K

Basic Parameters of Electrical Properties:

Mobility and Hall Effect 1:

A graph showing Electron mobility vs. temperature for four n-type samples can be found at reference 1.
A graph of the Resistivity and Hall coefficient vs. reciprocal temperature for four p-type samples can be found at reference 1.

Optical properties 1:

Refractive index: n = 2.5-2.7 hv = 0.5-0.9 eV

Absorption coefficient 5: 104 cm-1

Thermal properties:

Coefficient of thermal expansion 1:

X-ray studies on powder prepared from single crystals

Parallel: 7.90x10-6 1/K along c – axis

Perpendicular: 11.23x10-6 along a – axis

Debye temperature 1: 243.7 K

A graph of electron mobility vs. temperature can be found at reference 1.

Mechanical properties:

Bulk modulus 1: 4.82x1012 Pa

Elastic Constants

Second order elastic moduli 1: T=300 K

C11 9.70x1010 N/m

C33 10.89x1010 N/m

C44 3.62x1010 N/m

C66 3.16x1010 N/m

C12 5.97x1010 N/m

C13 8.60x1010 N/m

Acoustic Wave Speeds:

Sound velocity [1]:

υL 3.77x105 cm/s

υT 2.10x105 cm/s

Phonon Frequencies

A graph of the phonon dispersion curves of TA[100], LA[100], TA1[110], TA2[110], LA[1110], TO2[110], LA[001], TA[001] and
TO[001] modes can be seen at reference 1.

Developed at the University of Utah primarily by undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike
Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse, Jash Sayani, and Liz Lund also contributed. The work was fully supported by
the US National Science Foundation under the Materials World Network program award 1008302. These pages are a work in progress
and we solicit input from knowledgeable parties around the world for more accurate or additional information.
Contact earthabundantpv@eng.utah.edu with such suggestions.

1. a. b. c. d. e. f. g. h. i. j. k. l. m. n. o. p. O. Madelung, Semiconductors Data Handbook. Berlin: Springer, 2004, p. 691.

2. L. Al Juhaiman, Scoles, L., Kingston, D., Patarachao, B., Wang, D., and Bensebaa, F., “Green synthesis of tunable
 Cu(In1−xGax)Se2 nanoparticles using non-organic solvents”, Green Chemistry, vol. 12, no. 7, p. 1248, 2010.
3. C. Rincón and Ramírez, F. J., “Lattice vibrations of CuInSe2 and CuGaSe2 by Raman microspectrometry”, Journal of Applied
Physics, vol. 72, 1992.
4. T. Irie, Endo, S., and Kimura, S., “Electrical Properties of p- and n-Type CuInSe 2 Single Crystals”, Japanese Journal of
Applied Physics, vol. 18, no. 7, pp. 1303 - 1310, 1979.
5. S. Prabahar, Balasubramanian, V., Suryanarayanan, N., and Muthukumarasamy, N., “Optical properties of copper indium
diselenide thin films”, Chalcogenide Letters, vol. 7, pp. 49–58, 2010.

CuO
Add new comment

Basic Information
Alternate names:

copper (ii) oxide

cupric oxide
black copper
Tenorite – named after Professor Michele Tenore, an Italian botanist at the University of Naples, Italy [1]

CuO is a secondary copper mineral, a rare earth metal, and the most stable form of oxidized copper

Found in the oxidized zone of hydrothermal copper deposits, a volcanic sublimate [1]

CuO is a p-type semiconductor

Crystal Structure
X-ray Diffraction data 1:

X-RAY WAVELENGTH: 1.541838

MAX. ABS. INTENSITY / VOLUME**2: 94.00384821

2θ Intensity D-Spacing H K L Multiplicity

32.72 5.84 2.7372 1 1 0 4
35.64 30.43 2.5191 0 0 2 2

35.76 80.26 2.5108 -1 1 1 4

38.96 100 2.3118 1 1 1 4

39.26 23.3 2.2947 2 0 0 2

46.54 1.59 1.9516 -1 1 2 4

49.1 28.24 1.8553 -2 0 2 2

51.67 1.03 1.769 1 1 2 4

53.76 11.47 1.705 0 2 0 2

58.72 14.81 1.5724 2 0 2 2

61.92 20.59 1.4986 -1 1 3 4

66.18 16.04 1.412 0 2 2 4

66.8 15.56 1.4004 -3 1 1 4

68.34 9.91 1.3726 1 1 3 4

68.57 14.08 1.3686 2 2 0 4

 73.01 7.87 1.296 3 1 1 4
75.48 5.82 1.2596 0 0 4 2

75.77 6.97 1.2554 -2 2 2 4

80.76 2.32 1.1899 -2 0 4 2

83.06 5.17 1.1628 -3 1 3 4
83.66 5.08 1.1559 2 2 2 4

84.43 4 1.1474 4 0 0 2

A graph of XRD patterns of copper oxide thin films as deposited and annealed at various temperatures can be found from reference 2.

A graph of X-ray powder diffraction pattern of CuO nanoparticles (sample no. 2) (a) before calcination, (b) after calcination (PEG
template), (c) after calcination (PVA template) and (d) after calcination (PPG template) can be found from reference 3.

PV Applications
CuO has been used in solar cell research. At the University of Shiga, CuO layers were spin-coated at 100 nm thick on FTO substrate. It
was concluded that the formation of higher quality CuO thin films may improve future CuO cell efficiency.

A diagram of the structure of a FTO/CuO//Al heterojunction solar cell can be found from reference 4.

The solar cell of structure CuO(300°C)/ (spin.) gave a power conversion efficiency (ɳ) of 1.5E-4%, a fill factor (FF) of 0.25, short
circuit current density (JSC) of 13 μAcm^-2 and open-circuit voltage (VOC) of 45mV. The solar cell CuO(450°C)/ (eva.) showed
a similar photovoltaic performance. This table can be found from reference 4.

At Chiang Mai University, ZnO dye-sensitized solar cells (DSSCs) with different photoelectrodes were studied on the effect of CuO
layer as a barrier layer toward power conversion characteristics.

Schematic diagram of DSSC structures with different photoelectrodes for ZnO/CuO layer can be found from reference 5.

Semiconductor oxides are a promising alternative to silicon-based solar cells because they possess high optical absorption and are
composed of low cost materials. 6

Potential applications: photoconductive, photothermal, catalysis and gas sensor

CuO has been employed in photo-electrochemical cells 7<>

CuO has been used as a hole transfer layer and barrier layer for dye-sensitized solar cells 5, active layer in various types of solar cells
8, passive layer in solar-selective surfaces.

It would make a good selective absorbing layer because of its high solar absorbance and low thermal emittance.

Prepared by: spraying, chemical conversion, chemical brightening, etching, electrodeposition, electronbeam evaporation, reactive DC
sputtering and chemical vapor deposition

25°C 100°C 200°C 300°C

αs 0.73 0.73 0.73 0.73

CuO ε 0.040 0.038 0.044 0.052

300°C αs/ε 18.2 19.1 16.4 13.9

αs 0.90 0.90 0.90 0.90

CuO ε 0.52 0.55 0.59 0.65

500°C αs/ε 1.73 1.63 1.51 1.39

[10] Solar absorptance αs, thermal emittance ε, and selectivity αs/ε of CuO films deposited on gold-coated glass substrate

CuO film prepared at 500°C shows smaller values of selectivity, indicating that the well-crystallized CuO film is not useful as a solar
selective surface.

Basic Parameters at 300 K

Crystal Structure Monoclinic [11]

Group of Symmetry – C 2/c [11], [12]
Unit Cell Volume: V 80.63 ų
Density 6.31 g/cm³

Dielectric constant 18.1 [13]

Effective electron mass 0.4-0.95 9
7.9 mo 9

Electron affinity 4.07 eV

a= 4.652 Å
Lattice constants b= 3.410Å [11], [12]
c= 5.108 Å

Energy band-gap 1.35 eV 10

Band Structure and Carrier Concentration

A diagram showing the Band gap: 1.3 – 1.7 eV with a black color and a partial transparency in the visible range can be found from
reference 11.

Temperature Dependences:

A graph of the temperature dependence of conductivity plotted as ln(s) vs. 10^3/T in a temperature range 125–365 K and a graph of the
temperature dependence of conductivity plotted as ln(σT1/2) vs. 103/T can be found from reference [17].

Effective Mass 12: 7.9 m0

Electrical Properties
Limited data available

CuO is antiferromagnetic

Mobility holes [18]: 0.1 cm2 V−1s−1

Electric dipole moment [19]: 4.500 .5 (Debye)

A graph of the device current density voltage (J-V) curves both in the dark and light (under AM 1.5 100 mW cm−2 illumination) of a
bi-layer cell with a ~40 nm thick CuO layer can be found from reference 7.

A graph of the measured J-V characteristics of CuO/ thin films in the dark and under AM1.5 illumination can be found from reference
4.

Optical Properties
Basic Information [20]:

Type: Anisotropic
 Anisotropism: Strong, blue to grey
Bireflectance: Strong

Color in reflected light: light gray with golden tint

Pleochromism: Weak
Comments: Distinct, light to dark brown
Absorption coefficient [21]: α=0

A graph of the optical transmittance (T%) spectra of a copper oxide thin film as-deposited and annealed at various temperatures can be
found on reference 13Serin2005.

Refractive Index: n=2.65498 and the Extinction coefficient: k=0 can be found from reference [21].

Thermal Properties

Enthalpy of formation (298.15 K) [19]: 306.27 kJ/mol (Uncertainty: 41.8 kJ/mol)

Entropy (298.15 K) [19]: 234.62 J/mol*K

Integrated heat capacity (0-298.15 K) [19]: -9.75 kJ/mol

Heat capacity (298.15 K) [19]: 35.69 J/mol*K

A graph of the glancing angle XRD patterns of copper oxide thin film on n-Si wafer at various deposition temperatures, a graph of Cu
2p X-ray photoemission spectra of copper oxide film at various deposition temperatures, and a graph of Spectrophotometric
transmittance for copper oxide film deposited at various temperatures can be found from reference 6.

Mechanical Properties

Vibrational zero-point energy [19]: 320.1

A: 0
Rotational Constants [19]: B: .44454
C: .44454

37.92152 amu Å
Product of moments of inertia [19]:
6.29711E-39 gm cm²
Young’s modulus [23]: 81.6 GPa

References

[11] O. Madelung, Semiconductors: Basic Data (Springer-Verlag, Germany, 1996) pg. 11

[12] Landolt-Börnstein, Numerical data and functional relationships in science and technology.
Semiconductors 17, (1983)

[13] K-Tek, “Dielectric Constants Chart”, Available at: <http://www.asiinstr.com/technical/

Dielectric%20Constants.htm>

[19] NIST Computational Chemistry Comparison and Benchmark DataBase 101, “Listing of experimental
data for CuO (Copper Monoxide)”. NIST (2011)Available at: http://cccbdb.nist.gov

[20] J. Ralph, I. Chau, mindat Available at: <http://mindat.org> (2012)

[21] M. Polyanskiy, Refractive Index Database Available at: <http://refractiveindexdatabase.info/?
group=CRYSTAL&material=CuO>(2008)

The development of these pages on photovoltaic materials’ properties was carried out at the University of Utah primarily by
undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse,
Jash Sayani, and Liz Lund also contributed. The work was fully supported by the US National Science Foundation under the Materials
World Network program award 1008302. These pages are a work in progress and we solicit input from knowledgeable parties around
the world for more accurate or additional information. Contact earthabundantpv@eng.utah.edu with such suggestions. Neither the
University of Utah nor the NSF guarantee the accuracy of these values.

1. R. T. Downs and Hall-Wallace, M., “The American Mineralogist Crystal Structure Database”, American Mineralogist, vol. 88,
2003.
2. Citekey Serin2010 not found
3. R. Ranjbar-Karimi, Bazmandegan-Shamili, A., Aslani, A., and Kaviani, K., “Sonochemical synthesis, characterization and
thermal and optical analysis of CuO nanoparticles”, Physica B: Condensed Matter, vol. 405, no. 15, pp. 3096 - 3100, 2010.
4. a. b. c. R. MOTOYOSHI et al., “Structure and photovoltaic activity of cupric oxide-based thin film solar cells”, Journal of the
Ceramic Society of Japan, vol. 118, no. 1383, pp. 1021 - 1023, 2010.
5. a. b. P. Raksa, Nilphai, S., Gardchareon, A., and Choopun, S., “Copper oxide thin film and nanowire as a barrier in ZnO dye-
sensitized solar cells”, Thin Solid Films, vol. 517, no. 17, pp. 4741 - 4744, 2009.
6. a. b. K. Hyun Yoon, Choi, W. Jin, and Kang, D. Heon, “Photoelectrochemical properties of copper oxide thin films coated on
an n-Si substrate”, Thin Solid Films, vol. 372, pp. 250 - 256, 2000.
7. a. b. Y. - F. Lim, Choi, J. J., and Hanrath, T., “Facile Synthesis of Colloidal CuO Nanocrystals for Light-Harvesting
Applications”, Journal of Nanomaterials, vol. 20123312825910287, no. 710315291, pp. 1 - 6, 2012.
8. E. P. S. Tan et al., “Crystallinity and surface effects on Young’s modulus of CuO nanowires”, Applied Physics Letters, vol. 90,
no. 16, p. 163112, 2007.
9. a. b. M. Parhizkar et al., “Nanocrystalline CuO films prepared by pyrolysis of Cu-arachidate LB multilayers”, Colloids and
Surfaces A: Physicochemical and Engineering Aspects, vol. 257-258, pp. 277 - 282, 2005.
10. J. - W. Park, Baeg, K. - J., Ghim, J., Kang, S. - J., Park, J. - H., and Kim, D. - Y., “Effects of Copper Oxide/Gold Electrode as
the Source-Drain Electrodes in Organic Thin-Film Transistors”, Electrochemical and Solid-State Letters, vol. 10, no. 11, p.
H340, 2007.
11. D. Chauhan, Satsangi, V. R., Dass, S., and Shrivastav, R., “Preparation and characterization of nanostructured CuO thin films
for photoelectrochemical splitting of water”, Bulletin of Materials Science, vol. 29, p. 709, 2006.
12. Citekey Seri2011 not found
13. Citekey not found

CuS
Material: CuS
Earth abundance [1]: Good crystals are extremely rare

Basic Info
Occurs in nature as the mineral Covellite [1]
Usually as indigo-blue massive metallic material [1]
Discovered in 1832 [1]
Found in copper deposits [1]
Named after the Italian mineralogist, N. Covelli (1790-1829) [2]
Molecular Weight = 95.61 gm [2]
Physical Properties [1]:

Luster: Sub-Metallic

Diaphaneity
Opaque
(Transparency):

Color: Indigo-blue or darker, inclining towards blue-black, often iridescent with purplish, deep red, and brassy-yellow reflections.

Streak: Shiny metallic, lead-grey to black

Hardness (Mohs): 1½ - 2

Hardness (Vickers): VHN100=128 - 138 kg/mm2

Hardness Data: Measured

Tenacity: Flexible

Perfect
Cleavage:
Perfect on {0001}.

Fracture: Irregular/Uneven, Hackly

Density (measured): 4.6 - 4.76 g/cm3

Density (calculated): 4.602 g/cm3

Calculated Properties [2]:

Bulk Density (Electron Density)=4.41 gm/cc

Electron Density:
note: Specific Gravity of Covellite =4.68 gm/cc.

Fermion Index = 0.0025153223

Fermion Index:
Boson Index = 0.9974846777

Photoelectric: PECovellite = 31.55 barns/electron

Crystal Structure [1]:

Crystal System: Hexagonal

Class (H-M): 6/mmm (6/m 2/m 2/m) - Dihexagonal Dipyramidal

Space Group: P63/mmc {P63/m 2/m 2/c}

Cell Parameters: a = 3.7938Å, c = 16.341Å

Ratio: a:c = 1 : 4.307

Unit Cell Volume: V 203.7 ų

Z: 6

Hexagonal plates {001}, with pyramidal faces striated horizontally and hexagonal striations on the base. Common forms:
Morphology: {001}, {104}, {103}, {308}, {102}, {9/0/16}, {5.08}, {101} and {201}. Less common to rare: {1.0.16}, {1.0.12}, {3.0.32
{108}, {106}, {3.0.16}, {105}, {205}, {203}, {304}, {15.0.16} and {908}.

Twinning: None reported.

d-spacing Intensity

3.22 (30)

3.05 (70)

X-Ray Powder 2.81 (100)

Diffraction Data [1]:

2.72 (60)

2.32 (10)

1.90 (80)
 1.74 (40)

1.57 (20)

1.56 (40)

PV Applications:
CuS thin films are deposited by a chemical bath on top of other thin films such as Sb2S3 [3] and SnS [4]. Then the films are heat
treated with Nitrogen and used as an absorber layer [3].

Layer structures were: CuS, SnS, CdS, SnO2 [4] and Sb2S2-CuS, Sb2S3, CdS, TCO [3]

Band structure and carrier concentration

Band Gap [5]: 1.55 eV

Carrier Concentration [6]: +2.5×1021 cm-3

Temperature Dependences

A graph showing density of states for CuS vs (E-Ef) (Ryd) can be found at reference 7.

Electrical Properties
Conductivity [6]: 2×103 S cm-1

DC resistance [8]: 7.5 × 103 Ω

Resistivity [8]: 2.67 Ω cm

Basic Parameters of Electrical Properties

Electrical conductivity of CuS thin film [5]: 3.6 x 103 Ω-1 cm-1

The Hall mobility [6]: 7.24cm2 V-1 s-1

Optical properties [1] :

Type: Uniaxial (+)

RI values: nω = 1.450 nε = 2.620

Maximum δ = 1.170

Surface Relief: Very High

Type: Anisotropic

Anisotropism: Strong

Dispersion: Strong

Pleochroism: Visible

Comments: Deep blue to blue-white

Thermal properties

Mechanical properties [8]

Scratch resistance [8]: 3 mN

References:
[1] J. Ralph, I. Chau, mindat Available at: <http://mindat.org> (2012).

[2] WebMineral, “Covellite Mineral Data”, http://webmineral.com/data/Covellite.shtml#.UmK-dvlvNlJ

[3] http://www.sciencedirect.com.ezproxy.lib.utah.edu/science/article/pii/S0040609006016129

[4] http://www.sciencedirect.com.ezproxy.lib.utah.edu/science/article/pii/S004060900601604X

[5] http://iopscience.iop.org.ezproxy.lib.utah.edu/0268-1242/13/10/019/pdf/0268-1242_13_10_019.pdf

[6] http://scitation.aip.org.ezproxy.lib.utah.edu/content/aip/journal/apl/93/13/10.1063/1.2991441

[7] http://iopscience.iop.org.ezproxy.lib.utah.edu/0953-8984/4/50/034/pdf/0953-8984_4_50_034.pdf

[8] http://www.sciencedirect.com.ezproxy.lib.utah.edu/science/article/pii/S0025540810004046

The development of these pages on photovoltaic materials’ properties was carried out at the University of Utah primarily
by undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike Scarpulla. Caitlin Arndt, Christian
Robert, Katie Furse, Jash Sayani, and Liz Lund also contributed. The work was fully supported by the US National Science
Foundation under the Materials World Network program award 1008302. These pages are a work in progress and we
solicit input from knowledgeable parties around the world for more accurate or additional information.
Contact earthabundantpv@eng.utah.edu with such suggestions. Neither the University of Utah nor the NSF guarantee the
accuracy of these values.

FeS2 (Pyrite)
Add new comment

Basic Info
Pyrite, formally known as Iron disulfide, is the most abundant naturally occurring of the sulfide minerals. It has a crystal structure that
resembles the fluorite structure. Iron disulfide has a yellow-brass, metallic luster that is sometimes incorrectly recognized as gold. Due
to this mistaken identity it is often referred to as “fool’s gold”. 1

As the result of sparks generated when struck against metal, pyrite was used as a source of ignition for early firearms. Pyrite is also
used for commercial production of sulfur dioxide, which is used in the paper industry as well as in the manufacture of sulfuric acid.
Fool’s gold also has applications in jewelry, mineral detection in radio receivers, and photovoltaics. 1

Crystal Structure
 Fractional Coordinates Orthogonal Coordinates
Label Elmt x y z xor[Å] yor[Å] zor[Å]
1. Fe 0.0000 0.0000 0.0000 0.000 0.000 0.000

2. Fe 0.5000 0.0000 0.5000 2.862 0.058 2.544

3. Fe 0.5000 0.5000 0.0000 2.646 2.765 -0.157

4. Fe 0.0000 0.5000 0.5000 0.102 2.709 2.705

5. Fe 1.0000 0.0000 0.0000 5.405 0.114 -0.318

6. Fe 1.0000 0.5000 0.5000 5.507 2.823 2.387

7. Fe 0.0000 1.0000 0.0000 -0.114 5.415 0.004

8. Fe 0.5000 1.0000 0.5000 2.748 5.473 2.548

9. Fe 1.0000 1.0000 0.0000 5.291 5.529 -0.314

10. Fe 0.0000 0.0000 1.0000 0.318 0.003 5.407

11. Fe 0.5000 0.5000 1.0000 2.964 2.767 5.250

12. Fe 1.0000 0.0000 1.0000 5.723 0.117 5.089

13. Fe 0.0000 1.0000 1.0000 0.204 5.417 5.411

14. Fe 1.0000 1.0000 1.0000 5.609 5.532 5.093

15. S 0.3849 0.3849 0.3849 2.159 2.129 1.960

16. S 0.8849 0.3849 0.1151 4.776 2.185 0.343

17. S 0.1151 0.8849 0.3849 0.644 4.806 2.048

18. S 0.6151 0.8849 0.1151 3.261 4.862 0.430

19. S 0.6151 0.6151 0.6151 3.450 3.403 3.133

20. S 0.1151 0.6151 0.8849 0.833 3.346 4.750

 21. S 0.8849 0.1151 0.6151 4.966 0.726 3.045
22. S 0.3849 0.1151 0.8849 2.349 0.670 4.662

Theoretical diffraction data using a Cu Kα monochromatic source.

(m is the multiplicity and N is the maximum number of flexions)

ref no. h k l d(hkl) 2-Theta Intensity I/Imax m N

[ 1] 1 1 1 3.12693 28.5207 3.45405e-002 37.7 8 3
[ 2] 0 0 2 2.70800 33.0502 8.38185e-002 91.5 6 4

[ 3] 0 2 1 2.42211 37.0850 4.90083e-002 53.5 12 5

[ 4] 1 1 2 2.21107 40.7741 4.07521e-002 44.5 24 6

[ 5] 0 2 2 1.91485 47.4381 4.57922e-002 50.0 12 8

[ 6] 1 2 2 1.80533 50.5106 5.56486e-004 0.6 24 9

[ 7] 1 1 3 1.63299 56.2871 9.15808e-002 100.0 24 11

[ 8] 2 2 2 1.56346 59.0306 1.45073e-002 15.8 8 12

[ 9] 0 2 3 1.50213 61.6974 1.49215e-002 16.3 12 13

[10] 1 2 3 1.44749 64.2989 2.01434e-002 22.0 48 14

[11] 0 0 4 1.35400 69.3429 3.85639e-005 0.0 6 16

[12] 0 4 1 1.31357 71.8008 6.67362e-004 0.7 36 17

[13] 1 1 4 1.27656 74.2244 3.85281e-004 0.4 24 18

[14] 1 3 3 1.24252 76.6194 9.79268e-003 10.7 24 19

[15] 0 2 4 1.21105 78.9908 1.25237e-002 13.7 24 20

[16] 1 2 4 1.18187 81.3432 8.38154e-003 9.2 48 21

[17] 2 3 3 1.15470 83.6810 4.14859e-003 4.5 24 22

[18] 2 2 4 1.10554 88.3290 1.21383e-002 13.3 24 24

[19] 0 4 3 1.08320 90.6469 3.45885e-004 0.4 12 25

[20] 1 3 4 1.06217 92.9660 5.11313e-004 0.6 48 26

[21] 1 1 5 1.04231 95.2899 3.31754e-002 36.2 32 27

[22] 0 2 5 1.00573 99.9680 1.10694e-002 12.1 60 29

Photovoltaic Applications
The primary method of pyrite for photovoltaic applications is that of thin-films. It demonstrates extremely promising results for the use
as the active layer in solar photovoltaic and photoelectrochemical cells. Pyrite has a suitable band gap (Eg = 0.95 eV), effective light
absorption (R > 105 cm-1 for hν > 1.3 eV), an adequate minority carrier diffusion length (100-1000 nm), and for all intensive purposes
is in infinite elemental abundance. In principle, all of U.S. primary power demand (∼3.5 TW) could be met with 10% of the pyrite that
is disposed annually as mining waste in six U.S. states alone (assuming 10% cell efficiency and a conservative 5 μm thick pyrite active
layer). 2

Basic Parameters at 300 K

Crystal structure: Fluorite 3

Group of symmetry: Pa(-3) 3

Number of atoms in 1 cm3: 7.55*1022 3

Unit cell volume: 158.8678 Å3 3

Atoms per unit cell: 12 3

Auger recombination coefficient C: 10-26 cm6 s-1 3

Debye temperature: 6*106 K 4

Density: 5.0159 g/cm3 3

Dielectric constant: ɛ=10.9 5

Effective electron density: Nc = 3*1018 cm-3 5

Effective electron masses: me* = 0.25me 5

Effective hole density: Nv = 3*1019 ± 5*1019 cm-3 5

Effective hole masses: mh* = (2.2 ± 0.7) me 5

Lattice constant: 5.416 Å 6

 Optical phonon energy: 1.048 ± 0.005 eV 4

Band structure and carrier concentration

Graph on the amount of photo-generated carriers as a function of thickness of the planar pyrite and silicon film may be found in Pietro
P. Altermatt et al. 5

Graph of Majority carrier mobility as a function of majority carrier density of natural and synthetic pyrite crystals and of pyrite thin
films can be found in Pietro P. Altermatt, Tobias Kiesewetter, Klaus Ellmer, Helmut Tributsch, Specifying targets of future research in
photovoltaic devices containing pyrite (FeS2) by numerical modelling, Solar Energy Materials and Solar Cells, Volume 71, Issue 2, 1
February 2002, Pages 181-195, ISSN 0927-0248, 10.1016/S0927-0248(01)00053-8.5

Graph of the lifetime of excess carriers as a function of majority carrier density for various Auger coefficients C can be found in Pietro
P. Altermatt et al. 5

Temperature Dependency

Graph of optical absorption edge as a function of temperature may be found in C de las Heras et al 4

Donors and Acceptors

Donors: Ni, Co 1

Acceptors: As 1

Electrical Properties

Energy gap: 0.95 eV 6

Energy spin-orbital splitting: 1.2 eV 7

Intrinsic carrier concentration: 2.78*1012 cm-3 5

Carrier mobility: 120 cm2 V-1 s-1 4

Intrinsic resistivity: 0.18 Ω·cm 4

Basic Parameters of Electrical Properties

Mobility and Hall Effect

Mobility parameters: µmax = 300 cm2 / V s 5

µmin = 0.02 cm2 / V s 5

cref = 6*1017 cm-3 5

β = 1.3 5

Optical properties

Refractive index nref = 4.5 on average 5

Absorption coefficient 5*105 cm-1 (λ<750nm) 5

Thermal properties
Mechanical properties, elastic constants, lattice vibrations
Basic Parameters

Bulk modulus: 143 GPa 6

Density: 5.0159 g/cm3

Hardness: 6.3 on the Mohs scale

792 kg/mm2 @ 100 Gms load

Surface microhardness (using Knoop's pyramid test):
577 kg/mm2 @ 300 Gms load

Cleavage planes: (1 1 0), (1 1 1), (0 0 1)

Elastic Constants

C11 = 3.46-3.818 Mbars

C12 = -0.529-0.34 Mbars

C44 = 0.68-1.187 Mbars

Data on Raman spectrum of polycrystalline thin film may be found in C de las Heras et al 4

The development of these pages on photovoltaic materials’ properties was carried out at the University of Utah primarily by
undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse,
Jash Sayani, and Liz Lund also contributed. The work was fully supported by the US National Science Foundation under the Materials
World Network program award 1008302. These pages are a work in progress and we solicit input from knowledgeable parties around
the world for more accurate or additional information. Contact earthabundantpv@eng.utah.edu with such suggestions.

1. a. b. c. d. S. Lehner, Ciobanu, M., Savage, K., and Cliffel, D. E., “Electrochemical Impedance Spectroscopy of Synthetic
Pyrite Doped with As, Co, and Ni”, Journal of The Electrochemical Society, vol. 155, no. 5, p. P61, 2008.
2. J. Puthussery, Seefeld, S., Berry, N., Gibbs, M., and Law, M., “Colloidal Iron Pyrite (FeS 2 ) Nanocrystal Inks for Thin-Film
Photovoltaics”, Journal of the American Chemical Society, vol. 133, no. 4, pp. 716 - 719, 2011.
3. a. b. c. d. e. f. g. M. Rieder, Crelling, J. C., Šustai, O., Drábek, M., Weiss, Z., and Klementová, M., “Arsenic in iron disulfides
in a brown coal from the North Bohemian Basin, Czech Republic”, International Journal of Coal Geology, vol. 71, no. 2-3, pp.
115 - 121, 2007.
4. a. b. c. d. e. f. delas C Heras, Ferrer, I. J., and Sanchez, C., “Temperature dependence of the optical absorption edge of pyrite
FeS 2 thin films”, Journal of Physics: Condensed Matter, vol. 6, no. 46, pp. 10177 - 10183, 1994.
5. a. b. c. d. e. f. g. h. i. j. k. l. m. n. o. P. P. Altermatt, Kiesewetter, T., Ellmer, K., and Tributsch, H., “Specifying targets of future
research in photovoltaic devices containing pyrite (FeS2) by numerical modelling”, Solar Energy Materials and Solar Cells, vol.
71, no. 2, pp. 181 - 195, 2002.
6. a. b. c. J. Muscat, Hung, A., Russo, S., and Yarovsky, I., “First-principles studies of the structural and electronic properties of
pyrite”, Physical Review B, vol. 65, no. 5, 2002.
7. C. Wadia, Wu, Y., Gul, S., Volkman, S. K., Guo, J., and A. Alivisatos, P., “Surfactant-Assisted Hydrothermal Synthesis of
Single phase Pyrite FeS 2 Nanocrystals”, Chemistry of Materials, vol. 21, no. 13, pp. 2568 - 2570, 2009.

Mg2Si
Add new comment

Material: Magnesium Silicide (Mg2Si)

Basic Info
Magnesium silicide is a compound that is dark blue or slightly purple in color. When crystallized, Mg2Si displays a face-centered cubic
lattice arrangement. This arrangement possesses the antifluorite structure. Mg2Si is used as an additive for aluminum alloys, as a
negative electrode material for lithium-ion batteries, and in photovoltaic applications due to its semiconducting properties.

PV Applications
Magnesium silicide is primarily only used in thin film applications due to its difficulties in crystal growth. The low condensation
coefficient of magnesium and the barrier behavior of Mg2Si make it difficult to grow Mg2Si of substantial thickness. Successful
methods used to produce magnesium silicide films include: molecular-beam epitaxy, solid-phase growth, and ion-beam synthesis.

Basic Parameters at 300 K

Crystal structure: Antifluorite [1]

Debye temperature: 417 K T = 300 K [1]

Density: 1.88 g cm-3 [1]

Dielectric constants: ɛ(0) 20 [1]

ɛ(∞) 13.3

Effective masses: mn 0.46 m0 [1]

mp 0.87 m0 [1]

Electron affinity: 3.59 eV [5]

Lattice constant: a = 6.338 Å [1]

Band structure and carrier concentration

Band Structure and carrier concentration

Information on band structure may be found in Madelung, O. (2004). Semiconductors: Data handbook. (3rd ed., pp. 1595-1605).
Springer. [1]

Temperature Dependences

Graph of electrical conductivity vs. temperature for three n-type samples in the range of mixed conduction may be found in Madelung,
O. (2004). Semiconductors: Data handbook. (3rd ed., pp. 1595-1605). Springer. [1].

Graph of thermal conductivity vs. temperature for Mg2Si, Mg2Ge and Mg2Sn may be found in Madelung, O. (2004). Semiconductors:
Data handbook. (3rd ed., pp. 1595-1605). Springer.b [1].
 Energy Gap Narrowing at High Doping Levels

Effective Masses and Density of States

Donors and Acceptors

Donors: Bi [3]

Impurity formation energy data may be found in Zwolenski, P., Tobola, J., Kaprzyk, S., A theoretical sarch for efficient dopants in
Mg2X ( X = Si, Ge, Sn) Thermoelectrical materials. Springer Boston, Volume 40, Issure 5, 1 May 2011, Pages 889-897, ISSN
0361-5235, 10.1007/s11664-011-1624-y. [4]

Electrical Properties
Basic Parameters of Electrical Properties

Energy gap: 0.77 eV Indirect [7]

Conduction type: n-type [3]

Energy spin-orbital splitting: 0.03 eV [6]

Intrinsic carrier concentration: ni = 1*1014 cm-3 T= 300 K [1]

Electron Carrier mobility: 370 cm2/V s [7]

Hole mobility 70 cm2/V s [7]

Mobility and Hall Effect

Graph of Mg2Si hall mobility vs. temperature may be found in Madelung, O. (2004). Semiconductors: Data handbook. (3rd ed., pp.
1595-1605). Springer [1].

Optical properties
Refractive index 3.591 [2]

Thermal properties
Lattice heat capacity: Cv = 67.87 J mol-1 K-1 T= 300 K [1]

Mechanical properties, elastic constants, lattice vibrations

Basic Parameters

Density: 1.88 g cm-3 [1]

Elastic Constants

Elastic Constants:
 c11 12.1(2)*1011 dyn cm-2 T=300 K [1]

c12 2.2(2)*1011 dyn cm-2

c44 4.64(5)*1011 dyn cm-2

Graph of Mg2Si elastic moduli vs. temperature calculated from sound velocity data can be found in Madelung, O. (2004).
Semiconductors: Data handbook. (3rd ed., pp. 1595-1605). Springer. [1]

Acoustic Wave Speeds

Sound velocities:

νLA 7.68*105 cm s-1 T= 300 K [110] -direction [1]

νTA,I 4.83*105 cm s-1 [110] -direction, lower branch

νTA,II 4.97*105 cm s-1 [110] -direction, upper branch

νLA 7.65*105 cm s-1 [111] –direction

νTA 4.95*105 cm s-1 [111] -direction

Phonon Frequencies

νTO(Г15) 8.0*1012s-1 T=300 K [1]

νLO(Г15) 9.8*1012s-1

10.56*1012s-1 T= 77 K

ν(Г25’) 7.75*1012s-1 T=300 K

7.86*1012s-1 T=77 K

References

[1] Madelung, O. (2004). Semiconductors: Data handbook. (3rd ed., pp. 1595-1605). Springer.

[2] Benhelal, O., Chahed, A., Laksari, S., Abbar, B., Bouhafs, B. and Aourag, H. (2005), First-principles calculations of the structural,
electronic and optical properties of IIA–IV antifluorite compounds. Phys. Status Solidi B, 242: 2022–2032.
doi: 10.1002/pssb.200540063
(http://onlinelibrary.wiley.com/doi/10.1002/pssb.200540063/abstract)

[3] Jun-ichi Tani, Hiroyasu Kido, Thermoelectric properties of Bi-doped Mg2Si semiconductors, Physica B: Condensed Matter,

(http://www.sciencedirect.com/science/article/pii/S092145260500709X)

[4] Zwolenski, P., Tobola, J., Kaprzyk, S., A theoretical search for efficient dopants in Mg2X ( X = Si, Ge, Sn) Thermoelectrical
materials. Springer Boston, Volume 40, Issue 5, 1 May 2011, Pages 889-897, ISSN 0361-5235, 10.1007/s11664-011-1624-y.

(http://dx.doi.org/10.1007/s11664-011-1624-y)

[5] A. Atanassov, M. Baleva, On the band diagram of Mg2Si/Si heterojunction as deduced from optical constants dispersions,

(http://www.sciencedirect.com/science/article/pii/S0040609006010054)

[6] F. Vazquez*, Richard A. Forman, and Manuel Cardona. Electroreflectance Measurements on Mg2Si, Mg2Ge, and Mg2Sn.
Department of Physics, Brown University, Providence, Rhode Island 02912. 10.1103/PhysRev.176.905

(http://prola.aps.org/abstract/PR/v176/i3/p905_1)

[7] Yoji Imai, Akio Watanabe, Masakazu Mukaida, Electronic structures of semiconducting alkaline-earth metal silicides, Jo

(http://www.sciencedirect.com/science/article/pii/S0925838803000379)

The development of these pages on photovoltaic materials’ properties was carried out at the University of Utah primarily by
undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse,
Jash Sayani, and Liz Lund also contributed. The work was fully supported by the US National Science Foundation under the Materials
World Network program award 1008302. These pages are a work in progress and we solicit input from knowledgeable parties around
the world for more accurate or additional information. Contact earthabundantpv@eng.utah.edu with such suggestions. Neither the
University of Utah nor the NSF guarantee the accuracy of these values.

MnS
Basic Info [1]:
A naturally occurring mineral, Alabandite.
Named after the location of discovery, Alabanda, Aïdin, Turkey.
Discovered in 1832.
 Found in epithermal sulfide vein deposits.

Molecular Weight: 87.00 gm

Composition: 63.14% Mn, 36.86 % S

Empirical Formula: Mn2+S

Cleavage: {100} Perfect, {010} Perfect, {001} Perfect

Color: Black, Lead gray, Brownish gray.

Diaphaneity: Opaque

Fracture: Irregular/uneven

Hardness: 3.5-4 - Copper Penny-Fluorite

Luminescence: Non-flourecent

Luster: Sub Metallic

Magnetism: Nonmagnetic

Streak: Dark green, brown.

Tenacity: Brittle

Used for 1:

solar selective coatings

sensors
photoconductors
optical mass memories

MnS thin films can be found in several forms: the cubic modification having the rock-salt type structure (α-MnS), the zinc-blende-type
(β-MnS), or wurtzite-type structure (γ-MnS) 1.

Crystal Structure [2]:

Crystal System: Isometric

Class (H-M): m3m (4/m 3 2/m) - Hexoctahedral

Space Group: Fm3m {F4/m 3 2/m}

Space Group Setting: Fm3m

Cell Parameters: a = 5.2236Å

Unit Cell Volume: V 142.53 ų (Calculated from Unit Cell)

Z: 4

Morphology: Crystals cubic or octahedral, to 1 cm. Commonly massive, granular.

Twinning: Lamellar parallel to {111}

[3] X-RAY WAVELENGTH: 1.541838

MAX. ABS. INTENSITY / VOLUME**2: 84.46014982

2-THETA INTENSITY D-SPACING H K L Multiplicity
29.62 13.40 3.0164 1 1 1 8

34.33 100.00 2.6123 2 0 0 6

49.34 61.53 1.8471 2 2 0 12
58.60 5.40 1.5752 3 1 1 24

61.48 18.79 1.5082 2 2 2 8

72.35 7.90 1.3061 4 0 0 6
80.06 1.60 1.1986 3 3 1 24

82.58 20.55 1.1682 4 2 0 24

X-ray diffraction pattern for MnS showing peak intensities can be found in reference 4.

Find and list the crystal structure, atom positions, and create a Crystal Maker Model

PV Applications:

MnS is a dilute magnetic semiconductor that is used in solar cells as a window/ buffer layer [5]

Basic Parameters at 300 K [1]:

Density: 3.95 - 4.04, Average = 3.99

Band Structure:

Band Gap [5]: 3.02 eV

Temperature Dependences 1:

β and γ types MnS can transform irreversibly to the stable α type MnS at 100–400 °C

MnS thin films can be deposited at room temperature by chemical bath deposition

Donors and Acceptors [8]: Donors are substitutional iodine and acceptors are manganese vacancies for p-type α-MnS.

Electrical Properties

Basic Parameters of Electrical Properties [8]: For p-type α-MnS, conductivity occurs from holes in a 3d-band (Mn3+). The mobility
of these holes are not thermally activated.

Mobility and Hall Effect [6]:

Hall Mobility: 10 cm2/V s T = 625 K

Hall coefficient: 102 cm3/C

Optical properties [2]:

Type: Isotropic

RI values: n = 2.70

Maximum Birefringence: δ = 0.000 - Isotropic minerals have no birefringence

Surface Relief: Very High

Color in reflected light: Gray-white

A graph of the optical transmission spectra of MnS showing the peak maximum points and the peak minimum points can be seen at
reference 7.

A graph of the variation of refractive index n versus wavelength for a MnS thin film can be seen at reference 7.

A graph of α 2 (absorption coefficient) vs. photon energy can be seen at reference 7. The optical energy gap Eg = 3.88 eV was found.

Thermal properties [9]: Linear thermal expansion coefficients β for α-MnS are

β = 16. 3x10-6 ◦C-1 (225-591◦C) and β = 17. 4x10-6 ◦C-1 (591-928◦C).

References:

[1] WebMineral, “Alabandite Mineral Data”, http://webmineral.com/data/Alabandite.shtml#.Uj9224ZvPew

[2] J. Ralph, I. Chau, mindat Available at: <http://mindat.org> (2012).

http://www.mindat.org/min-89.html

[3] Downs R T (2006) The RRUFF Project: an integrated study of the chemistry, crystallography, Raman and infrared spectroscopy of
minerals. Program and Abstracts of the 19th General Meeting of the International Mineralogical Association in Kobe, Japan. O03-13

http://rruff.geo.arizona.edu/AMS/minerals/Alabandite

[4] Downs R T (2006) The RRUFF Project: an integrated study of the chemistry, crystallography, Raman and infrared spectroscopy of
minerals. Program and Abstracts of the 19th General Meeting of the International Mineralogical Association in Kobe, Japan. O03-13

http://rruff.info/alabandite/display=default/

[5] C. D. Lokhande et al. “Process and Characterization of Chemical Bath Deposited Manganese Sulphide (MnS) Thin Films,” Thin
Solid Films, vol. 330, No. 2, PP. 70-75. September 1998.

[6] Madelung, O. (2004). Semiconductors: Data handbook. (3rd ed.). Springer.

1 C. Gumus, C. Ulutas, Y. Ufuktepe. “Optical and Structural Properties of Manganese Sulfide Thin Films,” Optical Materials, vol. 29,
No. 9, pp. 1183-1187. May 2007.

[8] H.H. Heikens, C.F. Van Bruggen, C. Haas. “Electrical Properties of α-MnS,” Journal of Physics and Chemistry of Solids, vol. 39
No. 8, pp. 833-840, 1978.

[9] S. Furuseth and A. Kjekshus. “On the Properties of α-MnS and MnS2,” ACTA Chemica Scandinavica, Vol. 19, pp. 1405-1410.
1965.
Developed at the University of Utah primarily by undergraduate students Jeff Provost and Carina Hahn working with Prof.
Mike Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse, Jash Sayani, and Liz Lund also contributed. The work was
fully supported by the US National Science Foundation under the Materials World Network program award 1008302.
These pages are a work in progress and we solicit input from knowledgeable parties around the world for more accurate or
additional information. Contact earthabundantpv@eng.utah.edu with such suggestions.

1. a. b. c. d. C. Gümüş, Ulutaş, C., and Ufuktepe, Y., “Optical and structural properties of manganese sulfide thin films”, Optical
Materials, vol. 29, no. 9, pp. 1183 - 1187, 2007.

SnS (Herzenbergite)
Add new comment

Basic Info

Tin(II) Sulfide (SnS) is a brown solid and is insoluble in water. Its occurrence in nature takes the form a herzenbergite, which has an
orthorhombic crystal structure. SnS also has hexoganol (wurtzite) and cubic (Sphalerite) crystal forms which are less stable than
herzenbergite at room temperature.

Crystal Structure

Fractional Coordinates Orthogonal Coordinates

Label Elmt x y z xor[Å] yor[Å] zor[Å]
1. T2 S 0.4780 0.8500 0.2500 1.865 -4.067 8.693

2. T2 S 0.9780 0.6500 0.7500 4.885 -3.169 6.764

3. T2 S 0.5220 0.1500 0.7500 3.585 -1.595 1.198

4. T2 S 0.0220 0.3500 0.2500 0.565 -2.493 3.127

5. T1 Sn 0.1150 0.1180 0.2500 0.971 -1.035 0.982

6. T1 Sn 0.6150 0.3820 0.7500 3.795 -2.608 3.610
 7. T1 Sn 0.8850 0.8820 0.7500 4.479 -4.627 8.908
8. T1 Sn 0.3850 0.6180 0.2500 1.655 -3.053 6.281

The graph below shows peak intensities for SnS:

PV Applications

SnS is a p-type semiconducting material with a layered orthorhombic structure. It has little toxicity to humans and the environment as
well as a band gap of 1.0-1.5 eV. This makes SnS a suitable option as the absorption layer in n-type solar cells with a wide band gap.
Various techniques of fabricating SnS thin films include: vacuum evaporation, chemical bath deposition, and spray pyrolysis. 1

Basic Parameters at 300 K

Crystal structure: Orthorhombic 2

Group of symmetry: Pbnm 2

Number of atoms in 1 cm3: 4.15*1026 2

Unit cell volume: 192.6 Å3 [6]

Atoms per unit cell: 4 [7]

Debye temperature: 270 K [12]

Density: 5.08 g/cm3 [12]

Dielectric constants:

ε(0) 32(4) E||a T=300 K [12]

48(5) E||b
32(4) E||c

ε(∞) 14(2) E||a

 16(2) E||b
16(2) E||c

Effective electron density: 6.3*1014-1.2*1015 cm-3 Devika2006

Effective hole density: 1018 cm-3 Devika2006

Effective Masses:
Mp(||a) 0.2 m0 T= 300 K [12]

Mp(||b) 0.2 m0

Mp(||c) 1.0 m0

Mds,p 0.95 m0 T = 300 K [12]

Mds,p 0.45 m0

Lattice constants:
a= 4.33 Å

b = 11.18 Å

c = 3.98 Å

Band structure and carrier concentration

Band Structure and carrier concentration

Data on variation of optical bandgap of SnS films with thickness may be found in Thickness Effect on the Physical
Properties of Evaporated SnS Films M. Devika, N. Koteeswara Reddy, K. Ramesh, R. Ganesan, K. R. Gunasekhar, E. S.
R. Gopal, and K. T. Ramakrishna Reddy, J. Electrochem. Soc. 154, H67 (2007), DOI:10.1149/1.2398816 [10]

Temperature Dependence

Graphs of conductivity and Grain size vs. temperature of SnS may be found in M. Devika, K. T. Ramakrishna Reddy, N.
Koteeswara Reddy, K. Ramesh, R Ganesan et al., Microstructure dependent physical properties of evaporated tin sulfide films, J. Appl.
Phys. 100, 023518 (2006_; doi: 10.1063/1.2216790 3

Donors and Acceptor

Donors: Cl, Al

Acceptors: Cu, Ag

Donar and acceptor ionization energies [12]

Donor Ionization Acceptor Ionization

Compound Impurity
energy, eV energy, eV
ZnS Cl 0.25, 0.37
Cu 0.95
Ag 0.55

Electrical Properties
Basic Parameters of Electrical Properties

Energy gap: Direct:1.2-1.5 eV

 1

Indirect: 1.0-1.2 ev
1

Binding Energies: S 2p3/2 = 161.4 eV

3

Sn 3d5/2 = 485.3 eV
3

Intrinsic carrier concentration: 109 cm-3 (300 K)

[8]

Hole mobility (Hall Mobility): 90 cm2/Vs (resistivity = 0.06 Ω cm @ 300 K)

3

Intrinsic resistivity: 13-20 Ω

3

Mobility and Hall Effect

Graph of hall mobility and carrier density as a function of substrate temperature may be found in N. Koteeswara Reddy, K.T.
Ramakrishna Reddy, Electrical properties of spray pyrolytic tin sulfide films, Solid-State Electronics, Volume 49, Issue 6, June 2005,
Pages 902-906, ISSN 0038-1101, 10.1016/j.sse.2005.03.003. [9]

Optical properties
Refractive index 3.52 3

Absorption coefficient 104 cm-1 3

Graph of Transmittance as a function of wavelength for SnS grown thin films can be found in Thickness Effect on the Physical
Properties of Evaporated SnS Films M. Devika, N. Koteeswara Reddy, K. Ramesh, R. Ganesan, K. R. Gunasekhar, E. S. R. Gopal, and
K. T. Ramakrishna Reddy, J. Electrochem. Soc. 154, H67 (2007), DOI:10.1149/1.2398816 [10]

Plots of absorption spectroscopy and absorption coefficient vs photon energy may be found in Wei Guang-Pu; Zhang Zhi-Lin; Zhao
Wei-Ming; Gao Xiang-Hong; Chen Wei-Qun; Tanamura, H.; Yamaguchi, M.; Noguchi, H.; Nagatomo, T.; Omoto, O.; , "Investigation
on SnS film by RF sputtering for photovoltaic application," Photovoltaic Energy Conversion, 1994., Conference Record of the Twenty
Fourth. IEEE Photovoltaic Specialists Conference - 1994, 1994 IEEE First World Conference on , vol.1, no., pp.365-368 vol.1, 5-9
Dec 1994 doi: 10.1109/WCPEC.1994.519977 [5]

Thermal properties
Linear thermal expansion coefficient: α = 2.8*10-7 K-1 T= 300K [12]

Heat capacity: CP = 45 J mol-1 K-1 T=300 K [12]

CP = 29.3 J mol-1 K-1 T= 80 K [12]

Mechanical properties, elastic constants, lattice vibrations

Basic Parameters

Bulk modulus: 36.6 GPa [6]

Density: 5.08 g/cm3 [12]

Hardness: 2 on Mohs scale [7]

Phonon Frequencies
 VTO(B1u) 6.64*1012 s-1 T = 300 K

VLO(B1u) 8.28*1012 s-1

VTO(B1u) 5.32*1012 s-1
VTO(B1u) 6.43*1012 s-1
VTO(B1u) 2.96*1012 s-1
VLO(B1u) 3.2*1012 s-1

[12]
[2] Naoya Sato, Masaya Ichimura, Eisuke Arai, Yoshihisa Yamazaki, Characterization of electrical properties and photosensi

(http://www.sciencedirect.com/science/article/pii/S0927024804002077)

[4] M. Calixto-Rodriguez, H. Martinez, A. Sanchez-Juarez, J. Campos-Alvarez, A. Tiburcio-Silver, M.E. Calixto, Structural,

(http://www.sciencedirect.com/science/article/pii/S0040609008014077)

[5] Wei Guang-Pu; Zhang Zhi-Lin; Zhao Wei-Ming; Gao Xiang-Hong; Chen Wei-Qun; Tanamura, H.; Yamaguchi, M.; Noguchi, H.;
Nagatomo, T.; Omoto, O.; , "Investigation on SnS film by RF sputtering for photovoltaic application," Photovoltaic Energy
Conversion, 1994., Conference Record of the Twenty Fourth. IEEE Photovoltaic Specialists Conference - 1994, 1994 IEEE First World
Conference on , vol.1, no., pp.365-368 vol.1, 5-9 Dec 1994 doi: 10.1109/WCPEC.1994.519977
(http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=519977&isnumber=11458)

[6] L ehm, K knorr, P dera, A krimmel, P bouvier, and M Meqouar, Pressure-induced structural phase transistion in the IV-VI
semiconductor SnS, J. Phys.: Condens. Matter 16 3545 doi:10.1088/0953-8984/16/21/004

(http://iopscience.iop.org/0953-8984/16/21/004)

[7] Wolframalpha, Herzenbergite: fomula, density, Mohs hardness, refractive index,…

(http://preview.wolframalpha.com/entities/minerals/herzenbergite/ii/w2/73/)

[8] Polymorphic Tin Sulfide Thin Films of Zinc Blende and Orthorhombic Structures by Chemical Deposition David Avellaneda, M. T.
S. Nair, and P. K. Nair, J. Electrochem. Soc. 155, D517 (2008), DOI:10.1149/1.2917198

(http://scitation.aip.org/getpdf/servlet/GetPDFServlet?filetype=pdf&id=JE...)

[9] N. Koteeswara Reddy, K.T. Ramakrishna Reddy, Electrical properties of spray pyrolytic tin sulfide films, Solid-State Electronics,
Volume 49, Issue 6, June 2005, Pages 902-906, ISSN 0038-1101, 10.1016/j.sse.2005.03.003.

(http://www.sciencedirect.com/science/article/pii/S0038110105000845)
[10] Thickness Effect on the Physical Properties of Evaporated SnS Films

M. Devika, N. Koteeswara Reddy, K. Ramesh, R. Ganesan, K. R. Gunasekhar, E. S. R. Gopal, and K. T. Ramakrishna Reddy, J.
Electrochem. Soc. 154, H67 (2007), DOI:10.1149/1.2398816

(http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=JES...)

[11] Richard H. Bube and Edward L. Lind, Photoconductivity of Zinc selenide crystals and a correlation of donor and acceptor levels
in II-VI photoconductors. DOI: 10.1103/PhysRev.110.1040 (http://prola.aps.org/abstract/PR/v110/i5/p1040_1)

[12] Madelung, O. (2004). Semiconductors: Data handbook. (3rd ed., pp. 1981-1989). Springer.

Developed at the University of Utah primarily by undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike
Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse, Jash Sayani, and Liz Lund also contributed. The work was fully supported by
the US National Science Foundation under the Materials World Network program award 1008302. These pages are a work in progress
and we solicit input from knowledgeable parties around the world for more accurate or additional information. Contact
earthabundantpv@eng.utah.edu with such suggestions.
 1. a. b. c. N. Sato, Ichimura, M., Arai, E., and Yamazaki, Y., “Characterization of electrical properties and photosensitivity of
SnS thin films prepared by the electrochemical deposition method”, Solar Energy Materials and Solar Cells, vol. 85, no. 2, pp.
153 - 165, 2005.
2. a. b. c. R. W. G. Wyckoff, Crystal Structures 1, 2ndnd ed., vol. 1. New York, New York: Interscience Publishers, 1963.
3. a. b. c. d. e. f. g. M. Devika et al., “Microstructure dependent physical properties of evaporated tin sulfide films”, Journal of
Applied Physics, vol. 100, no. 2, p. 023518, 2006.

TiS2
Basic Info:
General Description 1: A yellow or gray powder that has an unpleasant odor. It is used as a solid lubricant. TiS2 is very toxic for
humans. If it comes in contact with water, hydrogen sulfide is produces, which is a toxic flammable gas. Inhaling the gas can cause
severe injury or death.

Fire Hazard. TiS2 is moisture sensitive and is very flammable when it comes into contact with moisture or air.

Thin films of TiS2 are used in Lithium Ion Batteries 2

Belongs to the transition metal dichalcogenides 3

TiS2 nanotubes can be used as a hydrogen storage material 3

Crystal Structure:
Lattice Parameters 4:

a = 3.40 Å

c = 5.69 Å

c/a = 1.67

hexagonal, space group P3m1,

Basic Parameters at 300 K:

Band structure and carrier concentration:

Carrier Concentration: 9 x1020 cm-34

A graph showing band structure for TiS2 can be seen at reference 3

Temperature Dependencies

A graph showing the temperature dependence of resistivity can be seen at reference 3

Electrical Properties:

Room temperature conductivity 5: 7.6 x10-1 ohm-1 cm-1

A graph of electrical conductivity vs temperature can be seen at reference 4

Basic Parameters of Electrical Properties:

TiS2 is a Semimetallic 3

Mobility and Hall Effect 5:

A graph of Hall Mobility vs Temperature can be seen at reference 4

Optical properties:

A graph of the measured reflectivity spectrum of TiS2 vs calculated spectra can be seen at reference 6

Absorption coefficient vs. phonon energy, etc.

Thermal properties

Lattice parameters with temperature coefficients 7

Developed at the University of Utah primarily by undergraduate students Jeff Provost and Carina Hahn working with Prof.
Mike Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse, Jash Sayani, and Liz Lund also contributed. The work was
fully supported by the US National Science Foundation under the Materials World Network program award 1008302.
These pages are a work in progress and we solicit input from knowledgeable parties around the world for more accurate or
additional information. Contact earthabundantpv@eng.utah.edu with such suggestions.

1. “TITANIUM(IV) SULFIDE CAS No. 12039-13-3”. .

2. T. S. Lewkebandara and Winter, C. H., “CVD routes to titanium disulfide films”, Advanced Materials, vol. 6, no. 3, pp. 237 -
239, 1994.
3. a. b. c. d. e. B. Liu et al., “Electronic structure of TiS2 and its electric transport properties under high pressure”, Journal of
Applied Physics, vol. 109, no. 5, p. 053717, 2011.
4. a. b. c. d. L. E. Conroy and Park, K. Chang, “Electrical properties of the Group IV disulfides, titanium disulfide, zirconium
disulfide, hafnium disulfide and tin disulfide”, Inorganic Chemistry, vol. 7, no. 3, pp. 459 - 463, 1968.
5. a. b. E. M. Logothetis, Kaiser, W. J., Kukkonen, C. A., Faile, S. P., Colella, R., and Gambold, J., “Hall coefficient and
reflectivity evidence that TiS 2 is a semiconductor”, Journal of Physics C: Solid State Physics, vol. 12, p. L521, 1979.
6. Citekey Vaterlaus1985 not found
7. S. M. Whittingham and Thompson, A. H., “Intercalation and lattice expansion in titanium disulfide”, The Journal of Chemical
Physics, vol. 62, no. 4, p. 1588, 1975.

ZnSe
Add new comment

Basic Info

ZnSe is a light yellow binary solid compound that is rarely occurs in nature. It can be made in hexagonal and cubic crystal structures.
Zinc selenide is used as a blue light source in light-emitting diodes and diode lasers. It is also used as infrared laser gain medium and
as an infrared optical material. ZnSe activated with tellurium is used in x-ray and gamma ray detectors as a scintillator. [9]

Crystal Structure
 Fractional Coordinates Orthogonal Coordinates
Label Elmt x y z xor[Å] yor[Å] zor[Å]
1. T2 Se 0.2500 0.2500 0.2500 -0.863 -1.331 -1.873

2. T2 Se 0.2500 0.7500 0.7500 0.100 -3.980 -4.722

3. T2 Se 0.7500 0.7500 0.2500 -2.657 -1.164 -5.453

4. T2 Se 0.7500 0.2500 0.7500 -3.482 -4.171 -2.935

5. T1 Zn 0.0000 0.0000 0.0000 0.000 0.000 -0.000

6. T1 Zn 0.0000 0.5000 0.5000 0.963 -2.649 -2.849

7. T1 Zn 0.5000 0.5000 0.0000 -1.794 0.167 -3.580

8. T1 Zn 0.5000 0.0000 0.5000 -2.620 -2.841 -1.063

9. T1 Zn 1.0000 0.0000 0.0000 -5.376 -0.025 -1.793

10. T1 Zn 1.0000 0.5000 0.5000 -4.414 -2.674 -4.642

11. T1 Zn 0.0000 1.0000 0.0000 1.788 0.358 -5.366

12. T1 Zn 0.5000 1.0000 0.5000 -0.831 -2.482 -6.429

13. T1 Zn 1.0000 1.0000 0.0000 -3.588 0.333 -7.159

14. T1 Zn 0.0000 0.0000 1.0000 0.137 -5.656 -0.332

15. T1 Zn 0.5000 0.5000 1.0000 -1.657 -5.490 -3.912

16. T1 Zn 1.0000 0.0000 1.0000 -5.239 -5.681 -2.125

17. T1 Zn 0.0000 1.0000 1.0000 1.925 -5.298 -5.698

18. T1 Zn 1.0000 1.0000 1.0000 -3.451 -5.323 -7.491

Theoretical diffraction data using a Cu Kα monochromatic source.

(m is the multiplicity and N is the maximum number of flexions)

ref no. h k l d(hkl) 2-Theta Intensity I/Imax m N

 [ 1] 1 1 1 3.27219 27.2296 3.56799e-001 100.0 8 3
[ 2] 0 0 2 2.83380 31.5438 1.03393e-003 0.3 6 4

[ 3] 0 2 2 2.00380 45.2123 2.62259e-001 73.5 12 8

[ 4] 1 1 3 1.70885 53.5827 1.57080e-001 44.0 24 11

[ 5] 2 2 2 1.63610 56.1706 3.14514e-004 0.1 8 12
[ 6] 0 0 4 1.41690 65.8603 4.10933e-002 11.5 6 16

[ 7] 1 3 3 1.30024 72.6539 6.12994e-002 17.2 24 19

[ 8] 0 2 4 1.26731 74.8585 6.43795e-004 0.2 24 20
[ 9] 2 2 4 1.15689 83.4861 8.34838e-002 23.4 24 24
[ 10] 1 1 5 1.09073 89.8496 4.77530e-002 13.4 32 27

[ 11] 0 4 4 1.00190 100.4906 3.01446e-002 8.4 12 32

PV Applications
ZnSe thin films are prepared primarily by molecular beam epitaxy, chemical vapor deposition, chemical vapor deposition, and vacuum
evaporation. Electrodeposition is a simple, low cost and a viable method for producing good quality films.1

ZeSe films prepared by chemical vapor deposition have a buffer layer that has reached total area efficiencies of up to 9.6% (under AM
1.5 illumination), an open circuit voltage of 482 mV, a short circuit current of 31.0 mA/cm2 and a fill factor reaching 64%. 2

Basic Parameters at 300 K

Crystal structure: Sphalerite 3

Group of symmetry: F-43m 3

Number of atoms in 1 cm3: 4.39*1026 3

Unit cell volume: 182.05 Å3 3

 Atoms per unit cell: 8 3
Debye temperature: 339(2) K 4

Density: 5.266 g/cm3 3

Dielectric constant (static): 8.6 5

Dielectric constant (high frequency): 5.7 5

Effective electron masses: (0.16 ± 0.01)me 6

Effective hole masses: 0.75 mo 4

Lattice constant: a = 5.667 Å 3

Optical phonon energy (longitudinal): 0.0314 eV 7
Conductivity: n-type 6

Temperature Dependences

Graph of electron concentration vs temperature may be found in M. Aven, High Electron Mobility in Zinc Selenide Through Low
Temperature Annealing. J. Appl. Phys. 42, 1204 (1971); doi: 10.1063/1.1660167 7

Donors and Acceptors

Donors: Al, Cl, Ga, In, F, Br 6,8

Acceptors: Cu, Ag, Sb 8

Ionization energy of shallow donors 4

Ed(LiI) 15(1) meV T= 4.2 K

Ed(NaI) 16(1) meV

Ed(Al) 26.3 meV

Ed(Ga) 27 meV

Ed(In) 28.1 meV

Ed(F) 29.3 meV

Ed(Cl) 26.1 meV

Ed(I) 30.4 meV

Ionization evergy of shallow acceptors

Ea(Li) 118(2) meV T = 4.2 K

Ea(Na) 98(2) meV

Ea(K) 94(2) meV

Ea(N) 112 meV T = 4.2 K

Ea(P) 80…92 meV T = 4.2 K

Ea(As) 125 meV T = 77 K

Ea(Sb) 69 meV T = 30 K

Ea(Rb) 89(2) meV T = 4.2 K

Ea(Cs) 74(2) meV

Ea(O) 80 meV T=4K

Ea(VZn) 218 meV T = 4 K

Electrical Properties
Basic Parameters of Electrical Properties

Energy gap: 2.81 eV

1

Energy spin-orbital splitting: ∆0 (Γ8v- Γ7v) 0.42 eV T=295 K 4

∆1 (Γ4,5v- Γ6v) 0.20 eV T=300 K 4

Donor concentration: 1016 cm-3 7

Carrier mobility: μn = up to 400 cm2/Vs T=300K 4

μp = 110 cm2/Vs T=300K 4

Intrinsic resistivity: ~1012 Ω cm 9

Mobility and Hall Effect

Hall mobility: 530 cm2/ V*s (T=300 K) 9

12,000 cm2/ V*s (T=60 K) 9

Absorption coefficient: 104 cm-1 1

Mobilities and mobility ratios as well as a graph of electron hall mobility vs temperature may be found in:

M. Aven, High Electron Mobility in Zinc Selenide Through Low Temperature Annealing. J. Appl. Phys. 42, 1204 (1971); doi:
10.1063/1.1660167 7

Optical properties
Refractive index: Graph of refractive index and absorption index vs. photon energy may be found in 4

Thermal properties
Coefficient of linear thermal expansion: α = 7.4*10-6 K-1 4

Heat capacity: Cp = 51.88 J/Mol*K 4

Thermal conductivity κ = 0.19 W K-1 cm-1 T=300K 4

Graphs of ZnSe's thermal properties may be found in Madelung 4

Mechanical properties, elastic constants, lattice vibrations

 Basic Parameters

Bulk modulus: 62.4(7) GPa 4

Density: 5.266 g/cm3 4

Elastic Constants

Elastic Constants: C11 = 90.3(19) GPa 4

C12 = 53.6(23) Gpa 4

C44 = 39.4(12) Gpa 4

Phonon Frequencies

VLO(Γ) 7.59 THz T=300 K 4

VTO(Γ) 6.39 THz T=300 K

Phonon Energies

HvLO(Γ1) 30.99 meV 4

HvTO(Γ15) 25.17 meV

HvLA(Γ) 19.8 meV

HvTA(Γ) 8.0 meV

HvLO(X) 27.64 meV

HvTO 25.54 meV

HvLA 23.55 meV

HvLO(L) 27.77 meV

HvTO 25.54 meV

Hv(W3) 24.9 meV

Hv(W1) 18.59 meV

Hv(W’2) 11.53 meV

Hv(W”2) 26.53. meV

Hv(W’4) 14.26 meV

Hv(W”4) 24.61 meV

Developed at the University of Utah primarily by undergraduate students Jeff Provost and Carina Hahn working with Prof. Mike
Scarpulla. Caitlin Arndt, Christian Robert, Katie Furse, Jash Sayani, and Liz Lund also contributed. The work was fully supported by
the US National Science Foundation under the Materials World Network program award 1008302. These pages are a work in progress
and we solicit input from knowledgeable parties around the world for more accurate or additional information. Contact
earthabundantpv@eng.utah.edu with such suggestions.
 1. a. b. c. R. Chandramohan, Sanjeeviraja, C., and Mahalingam, T., “Preparation of Zinc Selenide Thin Films by
Electrodeposition Technique for Solar Cell Applications”, physica status solidi (a), vol. 163, no. 2, pp. R11 - R12, 1997.
2. A. Rumberg, Sommerhalter, C., Toplak, M., Jäger-Waldau, A., and Lux-Steiner, M. C., “ZnSe thin films grown by chemical
vapour deposition for application as buffer layer in CIGSS solar cells”, Thin Solid Films, vol. 361-362, pp. 172 - 176, 2000.
3. a. b. c. d. e. f. g. R. W. G. Wyckoff, Crystal Structures 1, 2ndnd ed., vol. 1. New York, New York: Interscience Publishers,
1963.
4. a. b. c. d. e. f. g. h. i. j. k. l. m. n. o. p. q. r. s. Citekey Madelung2004c not found
5. a. b. S. Adachi and Taguchi, T., “Optical properties of ZnSe”, Physical Review B, vol. 43, no. 12, pp. 9569 - 9577, 1991.
6. a. b. c. J. L. Merz, Kukimoto, H., Nassau, K., and Shiever, J. W., “Optical Properties of Substitutional Donors in ZnSe”,
Physical Review B, vol. 6, no. 2, pp. 545 - 556, 1972.
7. a. b. c. d. M. Aven, “High Electron Mobility in Zinc Selenide Through Low-Temperature Annealing”, Journal of Applied
Physics, vol. 42, no. 3, p. 1204, 1971.
8. a. b. R. H. Bube and Lind, E. L., “Photoconductivity of Zinc Selenide Crystals and a Correlation of Donor and Acceptor
Levels in II-VI Photoconductors”, Phys. Rev., vol. 110, pp. 1040–1049, 1958.
9. a. b. c. G. Jones and Woods, J., “The electrical properties of zinc selenide”, Journal of Physics D: Applied Physics, vol. 9, no.
5, pp. 799 - 810, 1976.

Appendices
appendix module

Solar Cell Efficiency Records

Classification Efficiency Device Description Area (cm²) Voc (V) Jsc(mA/cm²) FF (%)
Ga0.35In0.65P/Ga0.83In0.17As/Ge
Isc = 380.5
Cell (concentrator) 41.1 % triple-junction, 454x, AM1.5d. 0.0509 2.867 87.2
mA
(Fraunhofer-ISE 1)

Cell at one-sun 35.8 % InGaP/GaAs/InGaAs (Sharp 2) 3.989 (t) 2.622 14.37 85

Silicon cell (one -sun) 25 % UNSW PERL 3 4.00 (da) 0.706 42.7 82.8

Module (concentrator) 27 % Entech 4 34 (ap)

Module (one-sun) 22.7 % UNSW PERL 5 778 (da) 5.6 3.93* 80.3
Module (production) 20.4 %** Sunpower6 16300 (t) 68.6 6.12* 79.1

* Isc (A) for module

** Best cell efficiency 24.2%

More detailed solar cell efficiency records are published every six months in Progress in Photovoltaics 7

The conditions for cell measurement are standardised for comparison purposes but may not reflect actual operating conditions.
Standard cell test conditions are 1000 W m-2, 25°C. Concentrator cells are measured using the direct beam AM1.5 spectrum while
other terrestrial cells use the global AM1.5 spectrum that also includes diffuse light. All the cells are tested on a temperature controlled
block and heating effects are ignored. In actual installations the cell temperature rises leading to an efficiency reduction.

The number of records reflects the large variety of technology options within the photovoltaic industry and the need for fundamental as
well as applied research. The highest efficiency devices (often on very small areas) demonstrate the practical limits to efficiency
without regard to cost or manufacturing considerations. Other records (particularly the large area cells and modules) give an indication
of the progress of technologies that are aimed at commercialisation.

Definitions of Area

The defined area of a solar cell can have a large effect on the quoted efficiency, especially for small cells.

1. Total area (t). The total area of the device including the frame.
 2. Aperture area (ap). The device is masked so that the illuminated area is smaller than the total cell or module area, but all
essential components of the device such as busbars, fingers and interconnects lie within the masked area.

3. Designated illumination area (da). In this case, the cell or module is masked to an area smaller than the total device area, but
major cell or module components lie outside the masked area. For example, for a concentrator cell, the cell busbars would lie
outside of the area designed for illumination and this area classification would be the most appropriate.

While most solar cells are remarkably stable, some cell technologies such as amorphous initially degrade under illumination and the
initial or stabilised efficiency is noted.

Solar cell efficiency records as a function of cell area. Larger cells and modules tend to have a lower efficiency.

1. F. Dimroth et al., “METAMORPHIC GaInP/GaInAs/Ge TRIPLE-JUNCTION SOLAR CELLS WITH > 41 %

EFFICIENCY”, 34th IEEE Photovoltaic Specialists Conference. 2009.
2. T. Takamoto et al., “World’s Highest Efficiency Triple-junction Solar Cells Fabricated by Inverted Layers Transfer Process”,
35 IEEE Photovoltaic Specialist Conference. Honolulu HI, USA, 2010.
3. M. A. Green, “The path to 25% silicon solar cell efficiency: History of silicon cell evolution”, Progress in Photovoltaics:
Research and Applications, vol. 17, pp. 183-189, 2009.
4. M. J. O’Neil and McDanal, A. J., “Outdoor measurement of 28% efficiency for a mini-concentrator module”, National Center
for Photovoltaics Program Review Meeting. Denver, USA, 2000.
5. J. Zhao et al., “20,000 PERL silicon cells for the "1996 World Solar Challenge" solar car race”, Progress in Photovoltaics:
Research and Applications, vol. 5, pp. 269–276, 1997.
6. P. J. Cousins et al., “Gen III: Improved Performance at Lower Cost”, in 35th IEEE Photovoltaic Specialists Conference,
Honolulu, Hawaii, 2010.
7. M. A. Green, Emery, K., Hishikawa, Y., and Warta, W., “Solar cell efficiency tables (version 35)”, Progress in Photovoltaics:
Research and Applications, vol. 18, pp. 144–150, 2010.

Standard Solar Spectra

The solar spectrum changes throughout the day and with location. Standard reference spectra are defined to allow the performance
comparison of photovoltaic devices from different manufacturers and research laboratories. The standard spectra were refined in the
early 2000's to increase the resolution and to co-ordinate the standards internationally. The previous solar spectrum, ASTMG159, was
withdrawn from use in 2005. In most cases, the difference between the spectrum has little effect on device performance and the newer
spectra are easier to use.

Further details on solar spectra are available at: http://www.nrel.gov/solar_radiation/

ASTM E-490
The standard spectrum for space applications is referred to as AM0. It has an integrated power of 1366.1 W/m2

ASTM G-173-03 (International standard ISO 9845-1, 1992)

Two standards are defined for terrestrial use. The AM1.5 Global spectrum is designed for flat plate modules and has an integrated
power of 1000 W/m2 (100 mW/cm2). The AM1.5 Direct (+circumsolar) spectrum is defined for solar concentrator work. It includes
the the direct beam from the sun plus the circumsolar component in a disk 2.5 degrees around the sun. The direct plus circumsolar
spectrum has an integrated power density of 900 W/m2. The SMARTS (Simple Model of the Atmospheric Radiative Transfer of
Sunshine) program is used to generate the standard spectra and can also be used to generate other spectra as required.

Standard Solar Spectra for space and terrestrial use.

Download the solar spectra in an Excel file.

The international measurement laboratories are in the processes of ratifying the new standards. In the meantime, concentrator cells are
measured using a low aerosol depth spectrum (Lo-AOD) that is appropriate for typical concentrator locations such as the southwest of
USA. It is essentially the same as the ASTM G-173-03 direct data given on this page.

Periodic Table
The Periodic Table of elements. Click on the table for a scalable version.

Units and Conversions

See also Constants

Energy and Power Conversions

1kWh 3.6 × 106 J
1 hp (horsepower) 746 W

1 Btu 1.055 kJ

Time Conversions
1 year 8765.8 hours
1 hour 3600 sec

1 year 3.157 x 107 sec

Solar Radiation Conversions

1 kWh/m2 1 Peak Sun Hour

1 kWh/m2 3.6 MJ/m2

1 kWh/m2 0.0116 Langley

1 kWh/m2 860 cal/m2

1 MJ/m2/day 0.01157 kW/m2

1 kW/m2 100 mW/cm2

 Miscellaneous
1 ft2 0.093 m2

1 m2 is approximately 10 ft2

Standard SI prefixes

Symbol Prefix Factor

T tera 1012

G giga 109

M mega 106

k kilo 103

c centi 10-2

m milli 10-3

μ* micron 10-6

n nano 10-9

p pico 10-12

* occasionally um is used instead of μm for microns

Physical Constants
see also Units and Conversions

Symbol Value Description

q 1.602 × 10-19 coulomb electronic charge

q 1.602 × 10-19 conversion from joules to eV

m0 9.108 × 10-31 kg electron rest mass

c 2.99792458 × 108 m/s speed of light in vacuum

8.85418 × 10-14 farad/cm

ε0 permittivity of free space
8.85418 × 10-12 farad/m

6.62606957 × 10-27 erg·s

h Planck's constant
6.62606957 × 10-34 joule·s

1.3806488 × 10-16 erg/K

k Boltzmann's constant
1.3806488 × 10-23 joule/K

σ 5.67 × 10-8 J/m2s K4 Stefan-Boltzmann constant

kT/q 0.02586 V thermal voltage at 300 K

λ0 wavelength of 1 eV photon 1.24 μm
Equations for Photovoltaics
Basic Equations

Density of States in Conduction and Valence Band

Fermi function:

Carrier Concentration in Equilibrium

Law of mass action:

Carrier concentrations:

n-type material:

p-type material:

Carrier Concentration Under Bias

Generation

Number of photons:

Generation rate:

Generation, homogeneous semiconductor: G = const:

P-type:
N-type:

Recombination

General SRH recombination rate:

Under low injection conditions:

For electrons:

For holes:

Basic PN Junction Equation Set

1. Poisson's equaion:

2. Transport equations:

3. Continuity equations:

General solution for no electric eifled, constant generation

Equations for PN Junctions

Built-in voltage pn homojunction:

General ideal diode equation:

I0 for wide base diode:

I0 for narrow base diode:

Full diode saturation currrent equation:

Depletion region recombination:

Solar Cell Equations

for constant G, wide base

Material Constants and Commont Units

Intrinsic carrier concentration:

Effective density of states:

Intrinsic energy level:

Diffusivity

Minority carrier diffusion length:

Resistivity and conductivity:

Resistance, homogeneous:

Permittivity:

Radiant Energy

Wavelength and energy of a photon:

If E is in eV and λ is in μm:

Spectral irradiance for black body:

Power density of a non-ideal black body:

Photon flux and power density:

Material mn*/m0 mp*/m0 EG (eV) ni (cm-3) c (eV) eS

Si 1.18 0.81 1.12 1.0 × 1010 4.03 11.8

Ge 0.55 0.36 0.66 2.0 × 1013 4.13 16.0

GaAs 0.066 0.52 1.42 1.8 × 106 4.07 13.1

InP 0.08 0.6 1.34 1.3 × 107 4.38 12.5

Equations for Photovoltaics 2

Better rendering in newer browsers using MathML

General equations for testing the browser

Quadratic Formula

Cosine

Properties of Sunlight

Batteries
Add new comment

Batteries module

Storage in PV Systems
Add new comment
Español

A fundamental characteristic of a photovoltaic system is that power is produced only while sunlight is available. For systems in which
the photovoltaics is the sole generation source, storage is typically needed since an exact match between available sunlight and the load
is limited to a few types of systems - for example powering a cooling fan. In hybrid or grid connect systems, where batteries are not
inherently required, they may be beneficially included for load matching or power conditioning.

By far the most common type of storage is chemical storage, in the form of a battery, although in some cases other forms of storage can
be used. For example, for small, short term storage a flywheel or capacitor can be used for storage, or for specific, single-purpose
photovoltaic systems, such as water pumping or refrigeration, storage can be in the form of water or ice.
In any photovoltaic system that includes batteries, the batteries become a central
component of the overall system which significantly affect the cost, maintenance
requirements, reliability, and design of the photovoltaic system. Because of large
impact of batteries in a stand-alone photovoltaic system, understanding the
properties of batteries is critical in understanding the operation of photovoltaic
systems. The important battery parameters that affect the photovoltaic system
operation and performance are the battery maintenance requirements, lifetime of
the battery, available power and efficiency. An ideal battery would be able to be
charged and discharged indefinitely under arbitrary charging/discharging regimes,
would have high efficiency, high energy density, low-self discharge and be low
cost. These are controlled not only by the initial choice of the battery but also by
how it is used in the system, particularly how it is charged and discharged and its
temperature. However, in practice, no battery can achieve the above set of
requirements, even if normally dominant requirement for low-cost is not
considered. This chapter provides an overview of battery operation and use for Energy storage represents a
photovoltaic systems. critical part of any energy system, and
chemical storage is the most frequently
Battery Basics employed method for long term storage.

Add new comment

Español

A battery converts energy stored in the chemical bonds of a material into electrical energy via a set of oxidation/reduction (commonly
abbreviated to redox) reactions. Redox reactions are chemical reactions in which an electron is either required or produced by the
chemical reaction. For primary batteries, this is a one-way process – the chemical energy is converted to electrical energy, but the
process is not reversible and electrical energy cannot be converted to chemical energy. This means that a primary battery cannot be
recharged. Examples of primary batteries are alkaline consumer batteries used in flashlights, etc. In a secondary battery, the conversion
process between electrical and chemical energy is reversible, – chemical energy is converted to electrical energy, and electrical energy
can be converted to chemical energy, allowing the battery to be recharged. For photovoltaic systems, all batteries used must be
rechargeable or secondary batteries. Common examples of secondary batteries are lead acid batteries and lithium-ion batteries used in
higher power consumer electronic equipment such as computer laptops, camcorders, mobile phones, and some digital cameras.

Basic Battery Operation

The basis for a battery operation is the exchange of electrons between two chemical reactions, an oxidation reaction and a reduction
reaction. The key aspect of a battery which differentiates it from other oxidation/reduction reactions (such as rusting processes, etc) is
that the oxidation and reduction reaction are physically separated. When the reactions are physically separated, a load can be inserted
between the two reactions. The electrochemical potential difference between the two batteries corresponds to the voltage of the battery
which drives the load, and the exchange of electrons between the two reactions corresponds to the current that passes through the load.
The components of a battery, which are shown in the figure below, and consist of an electrode and electrolyte for both the reduction
and oxidation reaction, a means to transfer electrons between the reduction and oxidation reaction (usually this is accomplished by a
wire connected to each electrode) and a means to exchange charged ions between the two reactions.

Schematic of a battery in which (a) the electrolyte of the reduction and oxidation reaction are different and (b) the electrolyte is the
 same for both reactions.

The key components which determines many of the basic properties of the battery are the materials used for the electrode and
electrolyte for both the oxidation and reduction reactions. The electrode is the physical location where the core of the redox reaction –
the transfer of electrons – takes place. In many battery systems, including lead acid and alkaline batteries, the electrode is not only
where the electron transfer takes places, but is also a component in the chemical reaction that either uses or produces the electron.
However, in other battery systems (such as fuel cells) the electrode material is itself inert and is only the site for the electron transfer
from one reactant to another. For a discharging battery, the electrode at which the oxidation reaction occurs is called the anode and by
definition has a positive voltage, and the electrode at which the reduction reaction occurs is the cathode and is at a negative voltage.

The electrode alone is not sufficient for a redox reaction to take place, since a redox reaction involves the interaction of more than a
single component. The other chemical components of the reaction are contained in the electrolyte. For many practical battery systems,
the electrolyte is an aqueous solution. One reasons for having an aqueous solution is the oxidized or reduced form of the electrode
exists in an aqueous solution. Further, it is important that the chemical species in the electrolyte be mobile in order that they can move
to the site on the electrode where the chemical reaction takes places, and also such that ion species can travel from one electrode to the
other.

The current in the battery arises from the transfer of electrons from one electrode to the other. During discharging, the oxidation
reaction at the anode generates electrons and reduction reaction at the cathode uses these electrons, and therefore during discharging,
electrons flow from the anode to the cathode. The electrons generated or used in the redox reaction can easily be transported between
the electrodes via a conventional electrical connection, such as a wire attached to the anode and cathode. However, unlike a
conventional electrical circuit, electrons are not the only charge carrier in the circuit. Electrons travel from the anode to the cathode,
but do not return from the cathode to the anode. Instead, electrical neutrality is maintained by the movement of ions in the electrolyte.
If each redox reaction has a different electrolyte, a salt bridge joins the two electrolyte solutions. The direction of the ion movement
acts to prevent a charge build-up at either the anode or the cathode. In most practical battery systems, the same electrolyte is used for
both the anode and the cathode, and ion transport can take place via the electrolyte itself, eliminating the need for a salt bridge.
However, in this case a separator is also inserted between the anode and the cathode. The separator prevents the anode and cathode
from physically touching each other since they are usually in very close physical proximity to one another, and if they were to touch it
would short out the battery as the electrons can be transferred directly without flowing through the external circuit and load.

The redox reactions which comprise a particular battery system define many fundamental parameters about the battery system. Other
key battery properties, including as battery capacity, charging/discharging performance and other practical considerations are also
influenced by the physical configuration of the battery, for example the amount of material in the battery or the geometry of the
electrodes. The following pages describe how battery characteristics – voltage behavior, battery efficiency, battery non-idealities (self-
discharge, degradation of battery capacity, etc) – are dependent on the operation of the redox reactions and the battery configuration.

Oxidation/Reduction Reaction
Reduction/oxidation (redox) reactions are an important class of chemical reactions since they are the driving force behind a vast range
of process, both desirable (for example breathing in mammals) and undesirable (for example rusting of iron). A redox reaction is
characterized by the fact that electrons are produced (in an oxidation reaction) or are used by the reaction (in a reduction reaction). An
oxidation reaction must always be paired with a reduction reaction, as the oxidation reaction produces the electrons required by the
reduction reaction.

The electrons transferred in a redox reaction arise from the change of the valence state of materials in the redox reaction. If a material
gives up or loses an electron, then its valance state becomes more positive (since an electron has a negative charge) and the reaction is
called an oxidation reaction. Since an oxidation reaction gives up electrons, it will always have electrons as one of its products. By
definition, the oxidation reaction occurs at the anode. The chemical reaction shown below is an oxidation reaction where zinc metal
(with a neutral valance state or valance charge = 0) is oxidized to give a zinc ion, which has a 2+ valence charge. The two electrons lost
by the zinc metal are products of the oxidation reaction. The zinc ion does not exist as separate entity, and therefore must for either a
solid salt (in which case its mobility and availability is not useful for redox reactions) as a dissolved salt in a solution. The (aq) after
the zinc ion indicates that it is aqueous. Note that since the overall aqueous solution must be electrically neutral, there must also be ions
with positive charge in the solution. In examining only the behavior of the battery reaction, these may not be specified. However, they
will play a role in the solubility of the Zn water (or an alternate solvent).

Zinc Oxidation with Valence Charge

Edit
Delete
 Clone

Read more about Zinc Oxidation with Valence Charge

Reaction: Oxidation reaction (the valence state of the reactant increases) of zinc metal to a zinc ion. The (s) after the zinc indicates
that it is in solid form. The zinc ion has (aq) after to indicate that it is aqueous, (ie in solution).

If a material gains an electron then its valance state decreases or reduces due to the negative charge of the electrons and the reaction is
a reduction reaction. The reaction below is a reduction reaction in which a copper ion with a valance state of 2+ is reduced to copper
metal, with a valence state of zero. Since a reduction reaction requires electrons, it will always have electrons as one of the reactants.
The reduction reaction occurs at the cathode.

The total redox Copper Reduction with Valence Charge

reaction consists
of both of the Edit
two reactions Delete
together. For the Clone
example of
copper and zinc
above, the total
reaction is
shown below. Read more about Copper Reduction with Valence Charge
Since the
reaction with
zinc metal (ie Reaction: Reduction reaction of Cu ions to form copper metal. The valence state of copper is reduced from 2+ to 0.
the reactant of
the oxidation reaction) is providing the electron required to reduce the copper, the zinc is the reducing agent and the zinc itself is
oxidized. Copper ions in this case are the oxidizing agent - they oxidize the zinc and are themselves reduced. Note that since the
electrons appear on both sides of the chemical equation, they may be omitted when writing the redox reaction. Further note that for
redox reaction, it is important to balance not only the elements in the chemical reactions, but also the electrons.

Zinc and Copper Redox with Electrons

Electrochemical Potential
Edit
Add new comment Delete
Clone
The voltage or potential difference between an oxidation and
reduction reaction arises from the different electrochemical
potentials of the reduction and oxidation reactions in the battery. The
electrochemical potential is a measure of the difference between the Read more about Zinc and Copper Redox with Electrons
average energy of the outer most electrons of the molecule or
element in its two valence states. (For those familiar with
semiconductor theory, the electrochemical potential difference
between an oxidation and reduction reaction is analogous to the Zinc and Copper Redox Balanced
Fermi level difference between two sides diode). As the
electrochemical potential is a measure of the energy of the outer Edit
most electrons, examination of the electronic configuration of the Delete
outer shell of the material will give an indication of the magnitude Clone
and sign of the electrochemical potential between the reactants and
products of a reduction or oxidation reaction.

Read more about Zinc and Copper Redox Balanced

 Reaction: Overall redox reaction for copper zinc reaction.

Figure: Outer electron configuration showing an element with one electron in its outer shell. When it loses its electrons, the outer
shell has eight electrons, and hence is a stable, low energy configuration. The core electrons are not shown.

The lowest energy configuration for materials is for their outer shell to be fully occupied by electrons. Hence, an element (say lithium,
Li) with one electron in its outer shell will have a higher energy than the element with the electron removed. Thus in the reaction

Lithium Oxidation
Edit
Delete
Clone

Read more about Lithium Oxidation

the lithium metal has a higher energy than Li+, and the oxidation reaction has a large positive electrochemical potential, with a value of
3.04V. The positive sign is defined such that the reaction proceeds spontaneously, if the electron is used by another reaction.

By convention, half reactions (a half reaction is either the oxidation or reduction reaction of an oxidation/reduction reaction, ie half of
the overall redox reaction) are written as reduction reactions, and hence the above reaction is typically tabulated as:

Lithium Reduction
Edit
Delete
Clone

Read more about Lithium Reduction

which has a negative standard potential. For all half reactions, changing the direction of the reaction changes the sign of the standard
potential, such that the reduction reaction of lithium above has a standard potential of E0 = -3.04 V.

Similarly, an element with 7 electrons in its outer shell (say Cl), will have a higher electrochemical potential than Cl which has gained
an electron and hence has a full outer shell.

Thus, in the reaction

Chlorine Reduction
 Figure: Outer electron configuration showing an element with one electron in its outer shell. When it loses its electrons, the outer
shell has eight electrons, and hence is a stable, low energy configuration. The core electrons are not shown.

Edit
Delete
Clone

Read more about Chlorine Reduction

the chlorine ion Cl- has a lower energy than the neutral Cl, and the reduction reaction has a large positive electrochemical potential of
1.36V.

Mnemonic Devices: Remembering which reaction occurs where and does what

There are several mnemonic devices to assist in remembering the terms in a battery. To remember the difference between a reduction
and an oxidation reaction, you can use any of the below methods. The first method it just based on the words describing reduction. In a
reduction reaction, the valance state is reduced, meaning that it gains a negative charge, ie an electron, and therefore the electron must
be a reactant in a reduction reaction. Thus the word itself – reduction – identifies the process that occurs there. An imaginative
mnemonic to remember which reaction is the oxidation and the reduction is "LEO (the Lion goes) GER (grrr)" in which LEO is short
for "Loss of Electrons - Oxidation" and GER is short for "Gain of Electrons - Reduction". Another way is to remember the difference
between oxidation and reduction reactions, (which is not quite is the spirit of renewable energy), is "OIL RIG": Oxidation Involves
Loss of electrons, Reduction Involves Gain of electrons."

There are also a few mnemonics to help remember at which reaction occurs at the anode (the oxidation) and which at the cathode (the
reduction)."The RED CAT and AN OX", groups the abbreviations for REDuction and CAThode together and also the abbreviations for
ANode and OXidation. Another way to remember which reactions occur at the anode and cathode is to note that the letter "O" (for
oxidation) occurs before the letter "R" (for reduction) and similarly that the letter "A" (for anode) occurs before the letter "C" (for
cathode). Thus, the oxidation occurs at the anode and the reduction occurs at the cathode. Finally, you may also remember that both
"anode" and "oxidation" begin with vowels and both "cathode" and "reduction" begins with consonants.

A final piece of information to remember is that the anode is negative and the cathode positive. The mnemonic for this is similar to that
of the anode and cathode. The anode is negative, the cathode is positive, "A" comes before "C" and "N" comes before "P". In a battery,
the situation is slightly more complex, as the physical location of the oxidation and reduction reaction changes between when the
battery is charging and discharging. By convention, the terms anode and cathode are defined according to conventional rules when the
battery is discharging and retain the same names when the battery is being charged.

Standard Potential
Add new comment

The overall energy of the outer electrons for a material depends on the temperature of the systems and also, for a material that is not a
pure solid, also depends on the concentration of the reacting species. Thus, in measuring the electrochemical potentials, both the
temperature and the concentration must be specified. The standard potential, given the symbol E0, is the electrochemical potential at
25 °C, with gasses having a pressure of 1 atm and solutions a concentration of 1 Molar (1 M or 1 mol/liter).

When measuring standard potentials, as with all voltages, only a difference in potential or voltage can be measured and hence only a
potential between an oxidation and reduction reaction can be measured. To determine the potential of a single oxidation or reduction
reaction, the potential of one reaction is measured compared to the potential of a specified reactions, which is oxidation of H. The
standard potential of this reaction is by definition set to zero.
Hydrogen Reduction
Edit
Delete
Clone

Read more about Hydrogen Reduction

The standard potential of the reduction /oxidation reaction pair is the sum of the standard potentials for the half reactions (to avoid
continually having to say the oxidation and reduction reaction, they are more generally termed a half reactions, since they make up half
of the overall redox reaction). The values of common half reactions are listed in Tables of Standard Potentials.

The standard potential of a redox reaction is used to determine if a redox reaction will occur spontaneously (ie if it will generate a
voltage between the reduction and oxidation reaction). If the difference between the standard potentials is positive, then the reaction
will proceed spontaneously. If the standard potential is negative, a voltage needs to be applied in order for the reaction to proceed.

From the table of standard potentials for reduction reactions for copper and zinc are:

Zinc Reduction with Standard Potential

Edit
Delete
Clone
 Read more about Zinc Reduction with Standard Potential

Copper Reduction with Standard Potential

Edit
Delete
Clone

Read more about Copper Reduction with Standard Potential

The zinc undergoes an oxidation reaction, so the direction of the reaction and the sign of the standard potential must be reversed to
give the oxidation/reduction reaction pair is:

Zinc Oxidation with Standard Potential

Edit
Delete
Clone

Read more about Zinc Oxidation with Standard Potential

Copper Reduction with Standard Potential

Edit
Delete
Clone

Read more about Copper Reduction with Standard Potential

and the standard potential for the reaction is 1.1V.

Nernst Equation
Add new comment

The standard electrochemical potential defines the voltage between a redox reaction under standard conditions (1 molar for aqueous
solutions, 1 atmosphere for gases). The voltage also depends on the concentration (for liquids) or pressure (for gaseous reactants) of
the components in the redox reaction since the average energy of the components will be altered by changes in concentration or
pressure. The voltage under conditions other than 1 M for solutions and 1 atm for gasses is given by the Nernst equation.

Nernst Equation
Edit
Delete
Clone
 Read more about Nernst Equation

where R is the universal gas constant (8.3145 J/mol K), T is temperature, and F is Faraday’s constant ( 96,484.6 C mol-1), and n is the
number of moles of electrons exchanged between the redox and oxidation reaction. Q is the reaction coefficient and for a reaction of
the form

Standard Redox
Edit
Delete
Clone

Read more about Standard Redox

is given by:

Reaction Coefficient
Edit
Delete
Clone

Read more about Reaction Coefficient

where [X] refers to the concentration or pressure of the material X (ie [D] is the concentration of the compound D in the reaction). For
a solid, this corresponds to 1, which means that battery systems which have all solid components do not have a concentration
dependence of their voltage.

A key implication of the Nernst equation is that the voltage of a battery is not necessarily constant, but varies as it is charged or
discharged as the concentration of the components of the electrolyte changes.

The concentration dependence of the potential means that for battery systems in which the components are not all solids and change
their concentration, the potential changes as the battery charges or discharges. This is shown below for a lead acid battery. However,
for a battery in which all the components of the redox reactions are solids and hence do not change their concentration, their ideal
battery voltage calculated from equilibrium conditions is constant.

Ideal battery capacity

The battery capacity is a measure of the amount of charge
or energy stored in the battery. The fundamental units of
battery capacity is coulombs (C), although a more
common and useful unit is Amp-hrs (Ah) (amps = C/time,
so Ah = C/time(sec) x time (hrs)). The battery capacity in
Ah can be ideally calculated from the weight/volume or
number of moles of materials of the electrode and
electrolyte (if it is an active component in the redox
reactions) in the battery. The ideal battery capacity under
equilibrium conditions (which can differ substantially
from the “real” battery capacity under load) is calculated
by from the moles of available reactants, from which the
moles of electrons can be determined. Using Faraday’s
constant, which gives the number of Coulombs for a mole
of electrons (F = 96,484.56 C/mol), the total available Equilibrium or open circuit battery voltage as a function of concentration.
coulombs (charge) can be determined for the battery.
Since the battery capacity, when described in Ah is a
measure of the total stored coulombs (Amps are C/sec), then the battery capacity can be determined by the equation:

Battery Capacity
Edit
Delete
Clone

Read more about Battery Capacity

Since the primary function of a battery is to store electrical energy rather than electrical charge, the energy storage of a battery is also
an essential parameter. A simple way to determine the energy storage capacity of the battery is to multiply the Ah capacity by the
nominal battery voltage, such that:

Battery Energy Capacity

Edit
Delete
Clone

Read more about Battery Energy Capacity

Battery voltage and capacity in non-equilibrium

The battery voltage described by the Nernst Equation and battery capacity assumes that the battery is in equilibrium. Since a battery
under load is not in equilibrium, the measured voltage and battery capacity may differ significantly from the equilibrium values, and
the further from equilibrium (ie the high the charge or discharge currents), the larger the deviation between the battery voltage and
capacity equilibrium and the realistic battery voltage may be. The difference between the voltage under equilibrium and that with a
current flow is termed polarization. Polarization effects have a significant impact on battery operation, both beneficial and detrimental.
For example, polarization effects mean that under normal operation of lead acid batteries the electrolysis of water proceeds slowly and
to first order can be neglected during discharge (but not charging since the voltage is higher). However, polarization effects also have
detrimental effects on performance, by, for example, reducing efficiency and making the battery capacity sensitive to charging and
discharging conditions.

The polarization is comprised of three basic mechanisms, relating to resistive drops in the battery, and to two effects relating to the
rates at which a reaction can proceed. These two effects are due to kinetic effects caused by the inherent rates of the chemical reaction
(called kinetic overvoltage or activation overvoltage), and by the effects related to the movement of reactants to the electrode (called
mass transport overvoltage). The overvoltage causes a deviation of the voltage and capacity from the equilibrium values calculated
earlier. As shown below, during discharging, the battery voltage is lower than that in equilibrium, while during charging, a higher
voltage than the Nernst voltage is required. Polarization effects have significant impact on the battery efficiency and how the battery
can be charged and discharged.
Cell potential away from equilibrium and including polarization effects.
Reaction Rates and Polarization

The equilibrium electrochemical potentials only take into account the initial and final potentials of the materials in the reaction,
without considering the rates or kinetics of the reactions themselves. The chemical reaction rates play an important role in determining
the operation of a battery and in the processes that control battery behavior. For example, if multiple reactions can occur, then a
reaction with a reaction rate significantly lower than all other reaction rates will not proceed to a significant extend and may potentially
be ignored.

A single chemical reaction is typically composed of multiple steps, and each of these steps has a particular rate. The reaction rate is
controlled by two processes. First, in order for the reaction to proceed, all the reactants must be physically present in one location,
which for a battery is the electrode. The processes which involve the transport of the reactants in their appropriate form to the site of
the chemical reaction are called mass transport or concentration overpotential. The steps in getting the reactants to the electrode are
shown below: all the reactants must be present in their appropriate form (ie in solution or as a solid), those in solution must diffuse to
the site of the reaction, the reactant species must absorb on the surface of the electrode (if the electrode is part of the chemical
reaction), and finally the electron transfer must occur. Applied to a lead acid battery, this means that both the lead metal and the lead
ion must be present. This involve the dissolution of the metallic ion (if it is present in solid form, as in the lead acid case shown
below), the transport of the reactants from the electrolyte to the electrode surface, and the adsorption of the necessary components on
the electrode surface.

In addition to the transport of the reactant species to the site of the reaction, a second possible rate limiting step for the reaction is the
rate at which the chemical reaction proceeds due to the kinetics of the chemical reaction. In many chemical reactions, the reacting
species form short-lived intermediate products, and then these intermediate products react to form the final products. If the rate of
formation of the intermediate species is slower than the remaining steps, then these intermediate steps control the reaction rate. Further,
the energy required to form these intermediate products may be higher than the average energy of the reactants. As the reactants have a
distribution of kinetic energy, and only those with higher energy can form the intermediate products. In this case, only a fraction of the
initial reactants have sufficient energy to allow the reaction to proceed, thus limiting the reaction rate. The higher energy of the
intermediate species gives rise to an activation energy, as shown in the figure below. In order for the reaction to proceed at a rapid rate,
the reactants must be given energy greater than the activation energy. As the kinetic energy of the reactants is determined by their
temperature, increasing the temperature of the reactants is a simple (but for batteries often impractical or accompanied by numerous
other negative aspects) way to increase the reaction rate and decrease the overpotential.

Another way to decrease the activation energy may be reduced for some reactions by the use of a catalyst. In some chemical reactions,
the reactant atoms must interact or collide in a particular way, such that a new material forms. For example, the interaction may require
that the reactants a physically oriented in a particular way, as shown in the figure below. For such reactions, the addition of other
chemical species that tend to orient the molecules in a specific orientation increase the probability of the reaction proceeding. Materials
that have such an effect are called catalysts. This effect makes reaction rates sensitive to the presence of a small number of other
species, which do not appear in the formula of the chemical reaction.

Mass Transport Overvoltage in Batteries

The mass transport overvoltage has a significant impact on batteries, particularly at high rates of charge and discharge. As the battery
discharges, it depletes the region around the electrode of some of the reactants. The concentration gradient between the region
surrounding the electrode and further away in the electrolyte causes reactants to diffuse towards the electrode. However, if the
discharge rate of the battery causes the reactants to be used at a greater rate than they can diffuse towards electrode, then the
concentration near the electrode will continue to drop as the battery discharges. This drop in concentration is greater than that expected
voltage drop if the reactants were uniformly distributed through the electrolyte and therefore, according to the Nernst equation, the
battery voltage decreases more rapidly than that calculated by equilibrium. The more rapidly a battery is discharged, the more rapid the
fall in voltage compared to that from equilibrium. Rapid discharging affects not only the battery voltage, but also battery capacity.
Since some of the reactants are not used in the reaction before the voltage drops below the minimum voltage, then the available battery
capacity is also reduced.

During charging, a similar process occurs, except that charging increases the concentration surrounding the electrode. Consequently, a
higher voltage is required to charge the battery than expected by equilibrium calculations. The mass transport overvoltage has a
significant effect on the battery parameters relevant to a photovoltaic system. The lower voltages during discharge and higher voltages
during charging reduce the battery efficiency. Further, mass transport effects alter the available battery capacity, as the battery capacity
is reduced under high discharge rates. Because of these effects, mass transport has a significant impact on the optimal use of a battery,
limiting both the charge and discharge currents.

Activation Overvoltage in Batteries

The kinetic or activation overvoltage of the reduction and oxidation reactions of the battery should be as small as possible, since during
charging the voltage required will greater than the equilibrium voltage by activation energy. The difference in the charging voltage and
the discharging voltage (i.e., the overvoltage) reduces the battery efficiency.

If there are secondary or side reactions in the battery, then the kinetic overpotential has different effects between charging and
discharging. During discharging, the battery voltage is lower, and therefore there is less possibility that the voltage is sufficient to
overcome the activation energy of secondary battery reactions. During charging, the battery voltage is higher, and hence there is the
possibility that additional reactions can occur. This effect can give rise to beneficial properties. The hydrolysis of water consists of the
redox reaction shown below, which has an electrochemical potential of 1.23 V.

Consequently, if a voltage of more than 1.23V is applied to a battery which has water as a component of the electrolyte, then the
electrolysis of water occurs, producing hydrogen and oxygen instead of the charging reaction for the battery. Since most batteries
operate at about 2V, this would then make water-based electrolytes unsuitable for batteries. However, the overvoltage of the redox
reactions in the electrolysis of water are high enough such that during discharging, gas evolution from the electrolysis of water (or
either one of the half reaction involved in the electrolysis of water) is not a dominant consideration. However, during charging, the
higher voltage experienced by the battery causes first the hydrogen and then the oxygen half reactions to proceed. In lead acid battery
systems, the presence of these two reactions gives rise to gassing. In many battery configurations, gassing leads to numerous
undesirable side-effects, including water loss from the electrolyte and physical damage to the electrolyte.

Resistive Drops in Batteries

A final contribution to the overvoltage in a battery is the resistive drops that occur in a battery. There are several components to the
total battery resistance. Part of the overall resistance is due to resistance of the components in the path of the electron flow, including
the electrode and the connections between the two electrodes. Other components of the resistive polarization include the surface of the
electrodes. The resistive polarization may not be linear with applied voltage. Other components of the resistive polarization include the
resistance of the surface of the electrode. For example, in a lead acid battery, as the discharge reaction proceeds, lead sulfate builds up
on the surface of the electrode, which is non-conductive. The resistive overpolarization has several practical impacts on battery
performance and operation. Similar to the concentration polarization, it reduces the efficiency and places limits on how much the
battery can be charged or discharged.

Secondary Reactions
In addition to the central reduction and oxidation reaction which comprise a battery, secondary or side reactions may occur. In most
cases, these side reactions give rise to unwanted or detrimental effects. In all cases, these secondary reactions reduce the coulombic
efficiency of the battery. If the secondary reaction occurs during discharging, some of the charge (current that would normally flow to
the load is used by the secondary reaction). Similarly, during charging, the secondary reactions use charge intended to drive the main
battery reactions, thus reducing the couloumbic efficiency. The electrolysis of water described in the activation overpotential is an
example of an unwanted secondary reaction. Secondary reactions give rise to several unwanted effects, such a gassing, self-discharge
and corrosion of the electrodes.

Physical State of the Electrodes

The physical state of the electrodes plays an important part in the practical operation of a battery. The key characteristic of a battery
electrode is that its surface area should be large. This lowers the series resistance, increases the area over which the chemical reaction
can take place (hence also reducing the mass transport overvoltage). In addition, a large surface area helps ensure that the reactants are
not completely covered by the products of the chemical reaction. A complete, uniform coverage of the electrode by the product
reaction would prevent the redox reactions from proceeding, since the reactant species could no longer reach the electrode. Moreover,
even in the reaction products allow the reactant species through, the reaction products are often not conductive, and therefore electrons
evolved or required by the redox reactions could not pass through the reaction productions. A large surface area is typically achieved
by using porous materials. The figure below shows the porous lead used in a lead acid battery.

During charging and discharging, several processes can occur that change the structure or shape of the electrode. In most battery
reactions, the electrode materials undergo a physical change during the discharge/charge cycle. The changes to the electrode, both
physical changes as the original electrode material is re-formed and chemical changes of the materials on the electrodes give rise to
numerous non-idealities. A key non-ideality is that the material may change its morphology, potentially during deposition of the
reaction products on the electrode, but more commonly when the electrode material remains unchanged for long periods of time. For
example, in lead-acid batteries, lead sulfate, which forms as the battery is discharged, may form large, relatively insoluble crystals over
time. These large crystals are difficult to convert back into lead or lead oxide, and hence they reduce battery capacity if the battery is
left in its discharged state.

Other effects that relate to the physical changes experienced by the electrode or electrolyte are that the reactant products seldom have
the identical density as the reactants, and hence the electrode undergoes physical changes in its size. If the mechanical stresses are too
large, the electrode material may flake off, hence permanently reducing capacity. The relative physical changes in size may be
exacerbated at high or low temperatures, as density differences may increase as the temperate changes.

Finally, as the electrode material is re-formed during charging, the electrode may change its shape. In lead acid batteries, this is
circumvented by the fact that the solubility of the lead ion Pb2+ is very low, and hence Pb2+ is rapidly converted to Pb in the close
physical proximity to where it was dissolved, thus preventing significant changes of shape of the electrode. Alternately, either the
products during discharging or the original battery material during charging may form so as isolate regions from charging or
discharging, thus permanently reducing battery capacity.

Battery Characteristics
The use of batteries in photovoltaic systems differs from the use of batteries in other common battery applications. For photovoltaic
systems, the key technical considerations are that the battery experience a long lifetime under nearly full discharge conditions.
Common rechargeable battery applications do not experience both deep cycling and being left at low states of charge for extended
periods of time. For example, in batteries for starting cars or other engines, the battery experiences a large, short current drain, but is at
full charge for most of its life. Similarly, batteries in uninterruptible power supplies are kept at full charge for most of their life. For
batteries in consumer electronics, the weight or size is often the most important consideration. This section provides an overview of the
critical battery characteristics or specifications, including battery voltage, capacity, charging/discharging regimes, efficiency, etc.

Battery Capacity
"Battery capacity" is a measure (typically in Amp-hr) of the charge stored by the battery, and is determined by the mass of active
material contained in the battery. The battery capacity represents the maximum amount of energy that can be extracted from the battery
under certain specified conditions. However, the actual energy storage capabilities of the battery can vary significantly from the
"nominal" rated capacity, as the battery capacity depends strongly on the age and past history of the battery, the charging or discharging
regimes of the battery and the temperature.

Units of Battery Capacity: Ampere Hours

The energy stored in a battery, called the battery capacity, is measured in either watt-hours (Wh), kilowatt-hours (kWh), or ampere-
hours (Ahr). The most common measure of battery capacity is Ah, defined as the number of hours for which a battery can provide a
current equal to the discharge rate at the nominal voltage of the battery. The unit of Ah is commonly used when working with battery
systems as the battery voltage will vary throughout the charging or discharging cycle. The Wh capacity can be approximated from the
Ahr capacity by multiplying the AH capacity by the nominal (or, if known, time average) battery voltage. A more accurate approach
takes into account the variation of voltage by integrating the AH capacity x V(t) over the time of the charging cycle. For example, a 12
volt battery with a capacity of 500 Ah battery allows energy storage of approximately 100 Ah x 12 V = 1,200 Wh or 1.2 KWh.
However, because of the large impact from charging rates or temperatures, for practical or accurate analysis, additional information
about the variation of battery capacity is provided by battery manufacturers.

Impact of Charging and Discharging Rate on Capacity

The charging/discharging rates affect the rated battery capacity. If the battery is being discharged very quickly (i.e., the discharge
current is high), then the amount of energy that can be extracted from the battery is reduced and the battery capacity is lower. This is
due to the fact the necessary components for the reaction to occur do not necessarily have enough time to either move to their
necessary positions. The only a fraction of the total reactants are converted to other forms, and therefore the energy available is
reduced. Alternately, is the battery is discharged at a very slow rate using a low current, more energy can be extracted from the battery
and the battery capacity is higher. Therefore, the battery of capacity should include the charging/discharging rate. A common way of
specifying battery capacity is to provide the battery capacity as a function of the time in which it takes to fully discharge the battery
(note that in practice the battery often cannot be fully discharged).

Temperature

The temperature of a battery will also affect the energy that can be extracted from it. At higher temperatures, the battery capacity is
typically higher than at lower temperatures. However, intentionally elevating battery temperature is not an effective method to increase
battery capacity as this also decreases battery lifetime.

Age and history of battery

The age and history of the battery have a major impact on the capacity of a battery. Even when following manufacturers specifications
on DOD, the battery capacity will stay at or close to its rated capacity for a limited number of charge/discharge cycles. The history of
the battery has an additional impact on capacity in that if the battery has been taken below its maximum DOD, then battery capacity
may be prematurely reduced and the rated number of charge/discharge cycles may not be available.

Battery Charging and Discharging Parameters

The key function of a battery in a PV system is to provide power when other generating sourced are unavailable, and hence batteries in
PV systems will experience continual charging and discharging cycles. All battery parameters are affected by battery charging and
recharging cycle.

Battery State of Charge (BSOC)

A key parameter of a battery in use in a PV system is the battery state of charge (BSOC). The BSOC is defined as the fraction of the
total energy or battery capacity that has been used over the total available from the battery.

Battery state of charge (BSOC or SOC) gives the ratio of the amount of energy presently stored in the battery to the nominal rated
capacity. For example, for a battery at 80% SOC and with a 500 Ah capacity, the energy stored in the battery is 400 Ah. A common
way to measure the BSOC is to measure the voltage of the battery and compare this to the voltage of a fully charged battery. However,
as the battery voltage depends on temperature as well as the state of charge of the battery, this measurement provides only a rough idea
of battery state of charge.

Depth of Discharge

In many types of batteries, the full energy stored in the battery cannot be withdrawn (in other words, the battery cannot be fully
discharged) without causing serious, and often irreparable damage to the battery. The Depth of Discharge (DOD) of a battery
determines the fraction of power that can be withdrawn from the battery. For example, if the DOD of a battery is given by the
manufacturer as 25%, then only 25% of the battery capacity can be used by the load.

Nearly all batteries, particularly for renewable energy applications, are rated in terms of their capacity. However, the actual energy that
can be extracted from the battery is often (particularly for lead acid batteries) significantly less than the rated capacity. This occurs
since, particularly for lead acid batteries, extracting the full battery capacity from the battery dramatically reduced battery lifetime. The
depth of discharge (DOD) is the fraction of battery capacity that can be used from the battery and will be specified by the
manufacturer. For example, a battery 500 Ah with a DOD of 20% can only provide 500Ah x .2 = 100 Ah.

Daily Depth of Discharge

In addition to specifying the overall depth of discharge, a battery manufacturer will also typically specify a daily depth of discharge.
The daily depth of discharge determined the maximum amount of energy that can be extracted from the battery in a 24 hour period.
Typically in a larger scale PV system (such as that for a remote house), the battery bank is inherently sized such that the daily depth of
discharge is not an additional constraint. However, in smaller systems that have a relatively few days storage, the daily depth of
discharge may need to be calculated.

Charging and Discharging Rates

A common way of specifying battery capacity is to provide the battery capacity as a function of the time in which it takes to fully
discharge the battery (note that in practice the battery often cannot be fully discharged). The notation to specify battery capacity in this
way is written as Cx, where x is the time in hours that it takes to discharge the battery. C10 = Z (also written as C10 = xxx) means that
the battery capacity is Z when the battery is discharged in 10 hours. When the discharging rate is halved (and the time it takes to
discharge the battery is doubled to 20 hours), the battery capacity rises to Y. The discharge rate when discharging the battery in 10
hours is found by dividing the capacity by the time. Therefore, C/10 is the charge rate. This may also be written as 0.1C. Consequently,
a specification of C20/10 (also written as 0.1C20) is the charge rate obtained when the battery capacity (measured when the battery is
discharged in 20 hours) is discharged in 10 hours. Such relatively complicated notations may result when higher or lower charging
rates are used for short periods of time.

The charging rate, in Amps, is given in the amount of charge added the battery per unit time (i.e., Coulombs/sec, which is the unit of
Amps). The charging/discharge rate may be specified directly by giving the current - for example, a battery may be charged/discharged
at 10 A. However, it is more common to specify the charging/discharging rate by determining the amount of time it takes to fully
discharge the battery. In this case, the discharge rate is given by the battery capacity (in Ah) divided by the number of hours it takes to
charge/discharge the battery. For example, a battery capacity of 500 Ah that is theoretically discharged to its cut-off voltage in 20 hours
will have a discharge rate of 500 Ah/20 h = 25 A. Furthermore, if the battery is a 12V battery, then the power being delivered to the
load is 25A x 12 V = 300W. Note that the battery is only "theoretically" discharged to its maximum level as most practical batteries
cannot be fully discharged without either damaging the battery or reducing its lifetime.

Charging and Discharging Regimes

Each battery type has a particular set of restraints and conditions related to its charging and discharging regime, and many types of
batteries require specific charging regimes or charge controllers. For example, nickel cadmium batteries should be nearly completely
discharged before charging, while lead acid batteries should never be fully discharged. Furthermore, the voltage and current during the
charge cycle will be different for each type of battery. Typically, a battery charger or charge controller designed for one type of battery
cannot be used with another type.

Battery Efficiency
As with any other component in a PV system, efficiency is an important issue in component selection due to the relatively high cost of
power generated by PV modules. The overall battery efficiency is specified by two efficiencies: the columbic efficiency and the
voltage efficiency.

Columbic Efficiency

The columbic efficiency of battery the ratio of the number of charges that enter the battery during charging compared to the number
that can be extracted from the battery during discharging. The losses that reduce columbic efficiency are primarily due to the loss in
charge due to secondary reaction, such as the electrolysis of water or other redox reactions in the battery. In general, the columbic
efficiency may be high, in excess of 95%.

Voltage Efficiency

The voltage efficiency is determined largely be the voltage difference between the charging voltage and voltage of the battery during
discharging. The dependence of the battery voltage on BSOC will therefore impact voltage efficiency. Other factors being equal, a
battery in which the voltage varies linearly with BSOC will have a lower efficiency than one in which the voltage is essentially
constant with BSOC.

Energy, Volumetric and Power Density

Energy density is a parameter used chiefly to compare one type of battery system to another. The energy density of a battery is the
capacity of the battery divided by either the weight of the battery, which gives the gravimetric energy density in Wh/kg, or by the
volume, which gives a volumetric energy density in Wh/dm3 (or Wr/litre3). A battery with a higher energy density will be lighter than
a similar capacity battery with a lower energy density. In portable systems, the energy density is a critical parameter but in
conventional PV systems which provide power for a stationary object, the energy density may be less important. Nevertheless, the
costs of transporting batteries to remote locations are considerably high, so a high energy density battery is typically an advantage.

The power density of a battery is related to its energy density, as well as the ability of the battery to discharge quickly. While the power
density is important in some applications, particularly transport, it is typically not critical in photovoltaic systems.

Battery Lifetime and Maintenance

Battery Lifetime

The lifetime of a battery may be specified in several different ways depending on the application and hence on which mechanism asre
most significant. For applications in which the battery is regularly charged and discharged (such as in photovoltaic systems), the most
appropriate measure of lifetime is the number of charge/discharge cycles over which the battery maintains a given fraction of its
capacity.

Since batteries inherently involve chemical reactions that are reactive, the materials used in batteries are susceptible to alternate
reactions that degrade battery performance. While certain catastrophic battery failure mechanisms are possible, battery lifetime is
typically controlled by the gradual degradation in battery capacity which accompanies charge/discharge cycles. Consequently, battery
lifetime is typically given as the number of charge/discharge cycles which it can undergo and still maintain its original capacity.
However, in systems which do not frequently experience charge/discharge cycles (such as in uninterruptable power supplies), battery
lifetime is more appropriately specified in years. Improper use of the battery can greatly accelerate battery aging and further decrease
the number of cycles over which a battery can be used.

Battery life is defined either in years (if it remains fully charged or in # of cycles under a given set of conditions (including temperature
and DOD).

Maintenance Requirements

The type of battery used will also have an important impact on the maintenance requirements of the battery. Some types of battery
reactions evolve gasses and other products which change the volume of the components in the battery. In cases in which the volume of
a battery changes, it is more difficult to seal the battery, and the battery will need to have certain chemical components (usually simply
water) added to compensate for the evolution of gasses. A hermetically sealed battery does not exchange any materials with its
surrounding environment. Such a battery will have lower maintenance requirements than a battery in which the various battery
elements interact with the surroundings. Nearly all small common primary batteries are hermetically sealed and require no
maintenance, but many secondary batteries, particularly lead acid batteries, require a strict maintenance schedule.

Failure Modes

A battery can degrade or can fail catastrophically. Modes are: shorts, degradation of electrode material, freezing, increases in
resistance.

Battery Safety and Disposal

Most battery systems, including those used in renewable energy systems, contain corrosive or dangerous chemicals and the safety
regulations for each type of battery should be carefully checked. Additional safety concerns relate to their ability to produce large
current. Finally, for lead-acid battery systems, the evolution of hydrogen is a potential issue.

Batteries should not be thrown away as most batteries contain toxic and/or corrosive material.

Battery Voltage
The voltage of a battery is a fundamental characteristic of a battery, which is determined by the chemical reactions in the battery, the
concentrations of the battery components, and the polarization of the battery. The voltage calculated from equilibrium conditions is
typically known as the nominal battery voltage. In practice, the nominal battery voltage cannot be readily measured, but for practical
battery systems (in which the overvoltages and non-ideal effects are low) the open circuit voltage is a good approximation to the
nominal battery voltage.
Since the electric potential (voltage) from most chemical reactions is on the order of 2V while the voltage required by loads is typically
larger, in most batteries, numerous individual battery cells are connected in series. For example, in lead acid batteries, each cell has a
voltage of about 2V. Six cells are connected to form a typical 12V lead acid battery.

Voltage Variation with Discharging

Due to the polarization effects, the battery voltage under current flow may differ substantially from the equilibrium or open circuit
voltage. A key characteristic of battery technology is how the battery voltage changes due under discharge conditions, both due to
equilibrium concentration effects and due polarization. Battery discharge and charging curves are shown below for several different
battery systems. The discharge and charge curves are not necessarily symmetric due to the presence of additional reactions that may be
present at the higher voltages encountered in charging.

Cut-Off Voltage

In many battery types, including

lead acid batteries, the battery
cannot be discharged below a
certain level or permanent damage
may be done to the battery. This
voltage is called the "cut-off
voltage" and depends on the type of
battery, its temperature and the
battery's rate of discharge.

Measuring State of Charge

Based on Voltage

While the reduction of battery

voltage with discharge is a negative Figure: Variation of voltage with state of charge for several different types of batteries.
aspect of batteries which reduces
their efficiency, one practical aspect
of such a reduction, if it is approximately linear, is that at a given temperature, the battery may be used to approximate the state of
charge of the battery. In systems where the battery voltage is not linear over some range of state of charge of the battery or in which
there are rapid variations in the voltage with the BSOC will be more difficult to determine the BSOC and therefore will be more
difficult to charge. However, a battery system that maintains a more constant voltage with discharge rate will have a high voltage
efficiency and will be more easily used to drive voltage sensitive loads.

Effect of Temperature on Voltage

Battery voltage will increase with the temperature of the system, and can be calculated by the Nernst Equation for the equilibrium
battery voltage.

Other Electrical Battery Parameters

4.6.1 Internal Series Resistance
The internal series resistance of a battery determines the maximum discharge current of the battery. Consequently, for applications in
which the batteries are required to provide high instantaneous power, the internal series resistance should be low. In addition, the series
resistance will affect the battery's efficiency but may change as the battery ages.

4.6.2 Self-Discharge

Self-discharge refers to the fact that even in the absence of a connected load, the discharge reaction will proceed to a limited extent and
the battery will therefore discharge itself over time. The rate of self-discharge depends primarily on the materials involved in the
chemical reaction (i.e., the type of battery system) and on the temperature of the battery.

4.6.3 Cold Cranking Current

The maximum amount of current a battery can provide for a short period of time is called the cranking current. This parameter is often
specified for transport applications, in which the battery must provide enough current to start a large engine. However, it is typically
not an important parameter in PV systems.

Summary and Comparison of Battery Characteristics

There are a large number of battery parameters. Depending on which application the battery is used for, some parameters are more
important than others. The following is a list of parameters that may be specified by a manufacturer for a given type of battery. For
example, in a typical battery for a general car, the energy density is not relevant - a battery is a small fraction of the total battery weight
and consequently this parameter would typically not be listed for a conventional car battery. However, in electric vehicle applications,
the battery weight is a significant fraction of the overall weight of the vehicle and so the energy densities will be given.
Parameter Applications Examples Comments
Usually standardized to
Nominal Voltage All
12V

Given as a function of
Battery Capacity All
discharge rate
Internal Series
High current applications Running motors
Resistance

Applications that are only recharged

Self Discharge Rate
periodically
Energy/Power Mobile applications using significant Electric vehicles, satellites,
Density amounts of battery energy portable electronics, etc
Mobile applications using significant Electric vehicles, satellites,
Cut-off voltage
amounts of battery energy portable electronics, etc
Application in which charging is restricted
Efficiency Stand-alone energy systems
or heat should be minimized

Applications which experience regular, Stand alone energy systems,

Depth of discharge
deep discharge traction batteries

Applications which experience large Stand alone energy systems, some

Temperature range
variations in temperature transport applications

Specified as either time

Battery lifetime All
or cycle life
Maintenance
All
requirements
Safety All

Lead Acid Batteries

5 Lead Acid Batteries

5.1 Introduction
Lead acid batteries are the most commonly used type of battery in photovoltaic systems. Although lead acid batteries have a low
energy density, only moderate efficiency and high maintenance requirements, they also have a long lifetime and low costs compared to
other battery types. One of the singular advantages of lead acid batteries is that they are the most commonly used form of battery for
most rechargeable battery applications (for example, in starting car engines), and therefore have a well-established established, mature
technology base.

5.2 Operation of Lead Acid Batteries

A lead acid battery consists of a negative electrode made of spongy or porous lead. The lead is porous to facilitate the formation and
dissolution of lead. The positive electrode consists of lead oxide. Both electrodes are immersed in a electrolytic solution of sulfuric
acid and water. In case the electrodes come into contact with each other through physical movement of the battery or through changes
in thickness of the electrodes, an electrically insulating, but chemically permeable membrane separates the two electrodes. This
membrane also prevents electrical shorting through the electrolyte. Lead acid batteries store energy by the reversible chemical reaction
shown below.

The overall chemical reaction is:

Lead Acid Overall Reaction

 Edit
Delete
Clone

Figure: Variation of voltage with state of charge for several different types of batteries.

Read more about Lead Acid Overall Reaction

At the negative terminal the charge and discharge reactions are:

Lead Acid Negative Terminal Reaction

Edit
Delete
Clone

Read more about Lead Acid Negative Terminal Reaction

At the positive terminal the charge and discharge reactions are:

Lead Acid Positive Terminal Reaction

Edit
Delete
Clone

Read more about Lead Acid Positive Terminal Reaction

As the above equations show, discharging a battery causes the formation of lead sulfate crystals at both the negative and positive
terminals, as well as the release of electrons due to the change in valence charge of the lead. The formation of this lead sulfate uses
sulfate from the sulfuric acid electrolyte surrounding the battery. As a result the electrolyte becomes less concentrated. Full discharge
would result in both electrodes being covered with lead sulfate and water rather than sulfuric acid surrounding the electrodes. At full
discharge the two electrodes are the same material, and there is no chemical potential or voltage between the two electrodes. In
practice, however, discharging stops at the cutoff voltage, long before this point. The battery should not therefore be discharged below
this voltage.

In between the fully discharged and charged states, a lead acid battery will experience a gradual reduction in the voltage. Voltage level
is commonly used to indicate a battery's state of charge. The dependence of the battery on the battery state of charge is shown in the
figure below. If the battery is left at low states of charge for extended periods of time, large lead sulfate crystals can grow, which
permanently reduces battery capacity. These larger crystals are unlike the typical porous structure of the lead electrode, and are difficult
to convert back into lead.

5.2.1 Voltage of lead acid battery upon charging.

The charging reaction converts the lead sulfate at the negative electrode to lead. At the positive terminal the reaction converts the lead
to lead oxide. As a by-product of this reaction, hydrogen is evolved. During the first part of the charging cycle, the conversion of lead
sulfate to lead and lead oxide is the dominant reaction. However, as charging proceeds and most of the lead sulfate is converted to
either lead or lead dioxide, the charging current electrolyzes the water from the electrolyte and both hydrogen and oxygen gas are
evolved, a process known as the "gassing" of the battery. If current is being provided to the battery faster than lead sulfate can be
converted, then gassing begins before all the lead sulfate is converted, that is, before the battery is fully charged. Gassing introduces
several problems into a lead acid battery. Not only does the gassing of the battery raise safety concerns, due to the explosive nature of
the hydrogen produced, but gassing also reduces the water in the battery, which must be manually replaced, introducing a maintenance
component into the system. In addition, gassing may cause the shedding of active material from the electrolyte, thereby permanently
reducing battery capacity. For these reasons, the battery should not regularly be charged above the voltage which causes gassing. The
gassing voltage changes with the charge rate.

Lead sulphate is an insulator, and therefore the way in which lead sulfate forms on the electrodes determined how easily the battery can
be discharged.

5.3 Characteristics of Lead Acid Batteries

For most renewable energy systems, the most important battery characteristics are the battery lifetime, the depth of discharge and the
maintenance requirements of the battery. This set of parameters and their inter-relationship with charging regimes, temperature and age
are described below.

5.3.1 Depth of Discharge and Battery Capacity

The depth of discharge in conjunction with the battery capacity is a fundamental parameter in the design of a battery bank for a PV
system, as the energy which can be extracted from the battery is found by multiplying the battery capacity by the depth of discharge.
Batteries are rated either as deep-cycle or shallow-cycle batteries. A deep-cycle battery will have depth of discharge greater than 50%,
and may go as high as 80%. To achieve the same useable capacity, a shallow-cycle battery bank must have a larger capacity than a
deep-cycle battery bank.

In addition to the depth of discharge and rated battery capacity, the instantaneous or available battery capacity is strongly affected by
the discharge rate of the battery and the operating temperature of the battery. Battery capacity falls by about 1% per degree below
about 20°C. However, high temperatures are not ideal for batteries either as these accelerate aging, self-discharge and electrolyte
usage. The graph below shows the impact of battery temperature and discharge rate on the capacity of the battery.
5.3.2 Battery Lifetime
Over time, battery capacity degrades due to sulfation
of the battery and shedding of active material. The
degradation of battery capacity depends most
strongly on the interrelationship between the
following parameters:

the charging/discharging regime which the

battery has experienced
the DOD of the battery over its life
its exposure to prolonged periods of low
discharge
the average temperature of the battery over its
lifetime

The following graph shows the evolution of battery Figure: Relationship between battery capacity, temperature and discharge rate.
function as number of cycles and depth of discharge
for a shallow-cycle lead acid battery. A deep-cycle
lead acid battery should be able to maintain a cycle life of more than 1,000 even at DOD over 50%.

In addition to the DOD,

the charging regime also
plays an important part
in determining battery
lifetime. Overcharging
or undercharging the
battery results in either
the shedding of active
material or the sulfation
of the battery, thus
greatly reducing battery
life.
Figure: Relationship between battery capacity, depth of discharge and cycle life for a shallow-cycle battery.
The final impact on
battery charging relates
to the temperature of the battery. Although the capacity of a lead acid battery
is reduced at low temperature operation, high temperature operation increases
the aging rate of the battery.

Figure: Impact of charging regime of battery capacity.

Constant current discharge curves for a 550 Ah lead acid battery at different discharge rates, with a limiting voltage of 1.85V per cell
(Mack, 1979). Longer discharge times give higher battery capacities.

5.3.3 Maintenance Requirements

The production and escape of hydrogen and oxygen gas from a battery causes water loss and water must be regularly replaced in lead
acid batteries. Other components of
a battery system do not require
maintenance as regularly, so water
loss can be a significant problem. If
the system is in a remote location,
checking water loss can add to
costs. Maintenance-free batteries
limit the need for regular attention
by preventing or reducing the
amount of gas which escapes the
battery. However, due to the
corrosive nature the elecrolyte, all
batteries to some extent introduce
an additional maintenance
component into a PV system.
Figure: Relationship between battery capacity, temperature and lifetime for a deep-cycle battery.
5.3.4 Battery
Efficiency
Lead acid batteries typically have coulombic efficiencies of 85% and energy efficiencies in the order of 70%.

5.4 Lead Acid Battery Configurations

Depending on which one of the above problems is of most concern for a particular application, appropriate modifications to the basic
battery configuration improve battery performance. For renewable energy applications, the above problems will impact the depth of
discharge, the battery lifetime and the maintenance requirements. The changes to the battery typically involve modification in one of
the three basic areas:

changes to the electrode composition and geometry

changes to the electrolyte solution
modifications to the battery housing or terminals to prevent or reduce the escape of generated hydrogen gas.

5.5 Special Considerations for Lead Acid Batteries

Flooded lead acid batteries are characterised by deep cycles and long lifetimes. However, flooded batteries require periodic
maintenance. Not only must the level of water in the electrolyte be regularly monitored by measuring its specific gravity, but these
batteries also require "boost charging".

Boost Charging

Boost or equalization charging involves short periodic overcharging, which releases gas and mixes the electrolyte, thus preventing
stratification of the electrolyte in the battery. In addition, boost charging also assists in keeping all batteries at the same capacity. For
example, if one battery develops a higher internal series resistance than other batteries, then the lower SR battery will consistently be
undercharged during a normal charging regime due to the voltage drop across the series resistance. However, if the batteries are
charged at a higher voltage, then this allows all batteries to become fully charged.

Specific Gravity (SG)

A flooded battery is subject to water loss from the electrolyte due to the evolution of hydrogen and oxygen gas. The specific gravity of
the electrolyte, which can be measured with a hydrometer, will indicate the need to add water to the batteries if the batteries are fully
charged. Alternately, a hydrometer will accurately indicate the SOC of the battery if it is known that the water level is correct. SG is
periodically measured after boost charging to insure that the battery has sufficient water in the electrolyte. The SG of the battery should
be provided by the manufacturer.

Special Considerations for Gelled, Sealed Lead Acid Batteries

Gelled or AGM lead acid batteries (which are typically sealed or valve regulated) have several potential advantages:
 they can be deep cycled while retaining battery life
they do not need boost charging
they require lower maintenance.

However, these batteries typically require a more precise and lower voltage charging regime. The lower voltage charging regime is due
to the use of lead-calcium electrodes to minimise gassing, but a more precise charging regime is required to minimise gassing from the
battery. In addition, these batteries may be more sensitive to temperature variations, particularly if the charging regime does not
compensate for temperature or is not designed for these types of batteries.

5.5.1 Failure Modes for Lead Acid Batteries

The battery for a PV system will be rated as a certain number of cycles at a particular DOD, charging regime and temperature.
However, batteries may experience either a premature loss in capacity or a sudden failure for a variety of reasons. Sudden failure may
be caused by the battery internally short-circuiting due to the failure of the electrical separator within the battery. A short circuit in the
battery will reduce the voltage and capacity from the overall battery bank, particularly if sections of the battery are connected in
parallel, and will also lead to other potential problems such as overcharging of the remaining batteries. The battery may also fail as an
open circuit (that is, there may be a gradual increase in the internal series resistance), and any batteries connected in series with this
battery will also be affected. Freezing the battery, depending on the type of lead acid battery used, may also cause irreversible failure of
the battery.

The gradual decline in capacity may be worsened by inappropriate operation, particularly by degrading the DOD. However, the
operation of one part of the battery bank under different conditions to another will also lead to a reduction in overall capacity and an
increase in the likelihood of battery failure. Batteries may be unintentionally operated under different regimes due either to temperature
variations or to the failure of a battery in one battery string leading to unequal charging and discharging in the string.

Installation

Battery installation should be conducted in accordance with the relevant standard in the country in which they are being installed. At
present, there are Australian standards AS3011 & AS2676 for battery installation. There is also a draft standard for batteries for RAPS
applications which will eventually become an Australian standard.

Among other factors to be considered in the installation of a battery system are the ventilation required for a particular type of battery
bank, the grounding conditions on which the battery bank is to be placed, and provisions taken to ensure the safety of those who may
have access to the battery bank. In addition, when installing the battery bank care must be taken to ensure that the battery temperature
will fall within the allowable operating conditions of the battery and that the temperature of the batteries in a larger battery bank are at
the same temperatures. Batteries in very cold conditions are subject to freezing at low states of charge, so that the battery will be more
likely to be in a low state of charge in winter. To prevent this, the battery bank may be buried underground. Batteries regularly exposed
to high operating temperatures may also suffer a reduced lifetime.

5.5.2 Safety
Batteries are potentially dangerous and users should be aware of three main hazards: The sulfuric acid in the electrolyte is corrosive.
Protective clothing in addition to foot and eye protection are essential when working with batteries.

Batteries have a high current generating capability. If a metal object is accidentally placed across the terminals of a battery, high
currents can flow through this object. The presence of unnecessary metal objects (e.g. jewellery) should be minimised when working
with batteries and tools should have insulated handles.

Explosion hazards due to evolution of hydrogen and oxygen gas. During charging, particularly overcharging, some batteries, including
most batteries used in PV systems, may evolve a potentially explosive mixture of hydrogen and oxygen gas. To reduce the risk of
explosion, ventilation is used to prevent the buildup of these gasses and potential ignition sources (i.e. circuits which may generate
sparks or arcs) are eliminated from the battery enclosure.

5.5.3 Maintenance
Batteries introduce a periodic maintenance component into a PV system. All batteries, including "maintenance free" batteries require a
maintenance schedule which should ensure that:

the battery terminals are not corroded

 the battery connections are tight
the battery housing should be free of cracks and corrosion.

Flooded batteries require extra and more frequent maintenance. For flooded batteries, the level of electrolyte and the specific gravity of
the electrolyte for each battery needs to be checked regularly. Checking the specific gravity of a battery by using a hydrometer should
be carried out at least 15 minutes after an equalisation or boost charge. Only distilled water should be added to batteries. Tap water
contains minerals which may damage the battery electrodes.

5.5.4 Battery Disposal and Recycling

The lead in a lead acid battery presents an environmental hazard if it is not properly disposed of. Lead acid batteries should be recycled
so that the lead can be recovered without causing environmental damage.

5.6 Electrode Materials and Configuration

The materials from which the electrodes are made have a major affect on the battery chemistry, and hence affect the battery voltage and
its charging and discharging characteristics. The geometry of the electrode determines the internal series resistance and the charging
and discharging rate.

5.6.1 Plate Material

The basic anode and cathode materials in a lead acid battery are lead and lead dixodie (PbO2). The lead electrode is in the form of
sponge lead. Sponge lead is desirable as it is very porous, and therefore the surface area between the lead and the sulfic acid electrolyte
is very large. The addition of small amounts of other elements to the lead electrode to form lead alloys can reduce several of the
disadvantages associated with the lead. The main types of electrodes used are lead/antimony (using several percent antimony),
lead/calcium alloys, and lead/antimony/calcium alloys.

Antimony lead alloy batteries have several advantages over pure lead electrodes. These advantages include: the lower cost of
lead/antimony; the increased strength of the lead/antimony electrode; and the ability to be deeply discharged for short period of time.
However, lead/antimony alloys are prone to sulfation and should not be left at low states of charge for extended periods of time. I
addition, lead/antimony alloys increase the gassing of the battery during charging leading to high levels of water loss. Since the water
must be added to these batteries, they have higher maintenance. Furthermore, lead/antimony batteries have a high discharge rate and a
short lifetime. These problems (xx- check if both problems are caused by plating)) are caused by the dissolution of antimony from one
electrode and its deposition or plating on the other electrode. (xx the increased adhesion of PbO2 xx)

Lead calcium batteries are an intermediate cost technology. Like antimony, calcium also adds strength to the lead of the negative
electrode, but unlike antimony, the addition of calcium reduces the gassing of the battery and also produces a lower self-discharge rate.
However, lead calcium batteries should not be deeply discharged. Consequently, these types of batteries may be considered
"maintenance-free", but are only shallow cycle batteries.

Adding antimony as well as calcium to the electrodes provides some of the advantages of both antimony and lead, but at an increased
cost. Deep discharge batteries such as these can also have a high lifetime. Furthermore, trace amounts of other materials can be added
to the electrodes to increase battery performance.

5.6.2 Electrode Configuration

In addition to the material used to make the electrode plates, the physical configuration of the electrodes also has an impact on the
charging and discharging rates and on the lifetime. Thin plates will allow faster charging and discharging, but are less robust and more
prone to shedding of material from the plates. As high charging or discharging currents are not typically a required feature of batteries
for renewable energy systems, thicker plates can be used, which have lower charge and discharge times, but also have longer lifetimes.

5.6.3 Battery Housing

In an open, flooded battery, any gas which is generated can escape to the atmosphere, causing both safety and maintenance problems.
A sealed lead acid (SLA), valve-regulated lead acid (VRLA) or recombining lead acid battery prevent the loss of water from the
electrolyte by preventing or minimizing the escape of hydrogen gas from the battery. In a sealed lead acid (SLA) battery, the hydrogen
does not escape into the atmosphere but rather moves or migrates to the other electrode where it recombines (possibly assisted by a
catalytic conversion process) to form water. Rather than being completely sealed, these batteries include a pressure vent to prevent the
build-up of excess pressure in the battery. Sealed batteries require stringent charging controls to prevent the build-up of hydrogen faster
than it can recombine, but they require less maintenance than open batteries.

Valve regulated lead acid (VRLA) batteries are similar in concept to sealed lead acid (SLA) batteries except that the valves are
expected to release some hydrogen near full charge. SLA or VRLA batteries typically have additional design features such as the use
of gelled electrolytes and the use of lead calcium plates to keep the evolution of hydrogen gas to a minimum.

5.7 Types of Lead Acid Batteries

Despite the range in battery types and applications, the characteristics particularly important in PV applications are the maintenance
requirements of the battery and the ability to deep charge a battery while maintaining a long lifetime. To promote long cycle life with
deep discharge, deep cycle batteries may be either of the open-flooded type, with an excess of electrolytic solution and thick plates, or
of the immobilized electrolytic type. Sealed gelled batteries may be rated as deep cycle batteries, but they will usually withstand fewer
cycles and lower discharges than the specially designed flooded plate or AGM batteries. Shallow-cycle batteries typically use thinner
plates made from lead calcium alloys and do not typically have a depth of discharge above 25%.

Batteries for PV or remote area power supplies (RAPS)

The stringent requirements for batteries used in photovoltaic systems have prompted several manufacturers to make batteries
specifically designed for PV or other remote power systems. The batteries most commonly used in stand-alone photovoltaic systems
are either deep-cycle lead acid types, or shallower cycle maintenance-free batteries. Deep-cycle batteries may be open flooded batteries
(which are not maintenance-free) or captive electrolyte AGM batteries which are maintenance-free (but which do require care in
regulator selection). Special shallow-cycle maintenance-free batteries that withstand infrequent discharging may also be used in PV
applications, and provided that the battery bank is appropriately designed, never require a DOD of more than 25%. A long-life battery
in an appropriately designed PV system with correct maintenance can last up to 15 years, but the use of batteries which are not
designed for long service life, or conditions in a PV system, or are part of a poor system design can lead to a battery bank which fails
after only a few years.

Several other types of specific purpose batteries are available and these are described below.

Starting, lighting ignition batteries (SLI). These batteries are used in automotive applications and have high discharge and charge rates.
Most often they use electrode plates strengthened with either lead antimony in a flooded configuration, or lead calcium in a sealed
configuration. These batteries have a good life under shallow-cycle conditions, but have very poor lifetime under deep cycling. SLI
batteries should not be used in a PV system since their characteristics are not optimized for use in a renewable energy system because
lifetime in a PV system is so low.

Traction or motive power batteries. Traction or motive batteries are used to provide electric power for small transport vehicles such as
golf carts. Compared to SLI batteries, they are designed to have a greater ability to be deep-cycled while still maintaining a long
lifetime. Although this feature makes them more suited to a PV system than one which uses SLI batteries, motive power batteries
should not be used in any PV systems since their self discharge rate is very high due to the use of lead antimony electrodes. A high self
discharge rate will effectively cause high power losses from the battery and make the overall PV system inefficient unless the batteries
experience large DOD on a daily basis. The ability of these batteries to withstand deep cycling is also far below that of a true deep-
cycle battery. Therefore, these batteries are not suited to PV systems.

RV or marine batteries. These batteries are typically a compromise between SLI batteries, traction batteries and true deep-cycle
batteries. Although they are not recommended, both motive and marine batteries are used in some small PV systems. The lifetime of
such batteries will be restricted to a few years at best, so that the economics of battery replacement mean that such batteries are
typically not a long-term cost effective option.

Stationary batteries. Stationary batteries are often used for emergency power or uninterruptable power supply applications. They are
shallow-cycle batteries intended to remain close to fully charged for the majority of their lifetime with only occasional deep discharges.
They may be used in PV systems if the battery bank is sized so that it never falls below a DOD of between 10% and 25%.

Deep-cycle Batteries. Deep-cycle batteries should be able to maintain a cycle life of several thousand cycles under high DOD (80% or
more). Wide differences in cycle performance may be experienced with two types of deep cycle batteries and therefore the cycle life
and DOD of various deep-cycle batteries should be compared.

5.8 Potential Problems with Lead Acid Batteries

A lead acid battery consists of electrodes of lead oxide and lead are immersed in a solution of weak sulfuric acid. Potential problems
encountered in lead acid batteries include:

Gassing: Evolution of hydrogen and oxygen gas. Gassing of the battery leads to safety problems and to water loss from the electrolyte.
The water loss increases the maintenance requirements of the battery since the water must periodically be checked and replaced.

Damage to the electrodes. The lead at the negative electrode is soft and easily damaged, particularly in applications in which the
battery may experience continuous or vigorous movement.

Stratification of the electrolyte. Sulfuric acid is a heavy, viscous liquid. As the battery discharges, the concentration of the sulfuric acid
in the elecotrolyte is reduced, while during charging the sulfiric acid concentratin increases. This cyclicing of sulfuric acid
concentration may lead to stratification of the electrolyte, where the heavier sulfuric acid remains at the bottom of the battery, while the
less concentrated solution, water, remains near the top. The close proximity of the electrode plates within the battery means that
physical shaking does not mix the sulfuric acid and water. However, controlled gassing of the electrolyte encourages water and sulfuric
acid to mix, but must be carefully controlled to avoid problems of safety and water loss. Periodic but infrequent gassing of the battery
to prevent or reverse electrolyte stratification is required in most lead acid batteries in a process referred to as "boost" charging.

Sulfation of the battery. At low states of charge, large lead sulfate crystals may grow on the lead electrode as opposed to the finely
grained material which is normally produced on the electrodes. Lead sulphate is an insulating material.

Spillage of the sulfuric acid. If sulfuric acid leaks from the battery housing it poses a serious safety risk. Gelling or immobilizing the
liquid sulfuric acid reduces the possibility of sulfuric acid spills.

Freezing of the battery at low discharge levels. If the battery is at a low discharge level following the conversion of the whole
electrolyte to water, then the freezing point of the electrolyte also drops.

Loss of active material from the electrodes. The loss of active material from the electrodes can occur via several processes. One
process that can cause a permanent loss of capacity is the flaking off of the active material due to volumetric changes between xxx and
lead sulphate. In addition, xxx. Improper charging conditions and gassing can cause shedding of active material from the electrodes,
leading to a permanent loss in capacity.

Depending on which one of the above problems is of most concern for a particular application, appropriate modifications to the basic
battery configuration improve battery performance. For renewable energy applications, the above problems will impact the depth of
discharge, the battery lifetime and the maintenance requirements. The changes to the battery typically involve modification in one of
the three basic areas:

changes to the electrode composition and geometry

changes to the electrolyte solution
modifications to the battery housing or terminals to prevent or reduce the escape of generated hydrogen gas.

5.8.1 Corrosion of terminals

Corrosion consists of a set or reduction/oxidation regions in which both the reactions take place at the same electrode. For a battery
system, corrosion leads to several detrimental effects. One effect is that it converts a metallic electrode to a metal oxide.

5.8.2 Self-Discharge
All chemical reactions proceed in both the forward and reverse direction. In order for the reverse reaction to proceed, the reactants
must gain enough energy to overcome the electrochemical difference between the reactants and the products and also the overvoltage.
Usually in battery systems, the probability of the reverse reaction occurring is small, since there are few molecules with a large enough
energy. Although small, however, there are some particles that do have sufficient energy. In a charged battery, a process exists by
which the battery can be discharged even in the absence of a load connected to the battery. The amount a battery discharges upon
standing is known as self-discharge. Self-discharge increases as temperature increases because a greater fraction of products will have
enough energy for the reaction to proceed in the reverse direction.

An ideal set of chemical reactions for a battery would be one in which there is a large chemical potential which releases a large number
of electrons, has a low overvoltage, spontaneously proceeds in only one direction and is the only chemical reaction which can occur.
However, in practice, there are several effects that degrade battery performance, due to unwanted chemical reactions, to effects such as
the change in phase of volume of the reactants or products and also to the physical movement of reactants and products within the
battery.

5.8.3 Change in form of materials

While undergoing chemical reactions, many materials undergo a change either in phase, or if they stay in the same phase, the volume,
density of the material may be altered by the chemical reaction. Finally, the materials used in the battery, primarily the anode and
cathode, may change their crystallinity or surface structure, which will in turn affect the reactions in the battery. Many components in
redox reactions undergo a change in phase during either oxidation or reduction. For example, in the lead acid battery, sulfate ions
changes from being in solid form (as lead sulfate) to being in solutions (as sulfuric acid). If the lead sulfate recrystallizes anywhere but
the anode or cathode, then this material is lost to the battery system. During charging, only materials connected to the anode and
cathode can participate in electron exchange, and therefore if the material is not touching the anode or cathode, then it can no longer be
recharged. The formation of a gaseous phase in a battery also presents special problems. First of all, the gaseous phase will usually
have a larger volume that the initial reactants, thus giving rise to a change in pressure in the battery. Secondly, if the intended products
are in the gaseous change, they must be confined to the anode and cathode, or they will not be able to be charged.

A change in volume will also usually be detrimental in battery operation.

5.8.4 Modifications to the electrolyte

A standard "flooded" lead acid battery has the electrodes immersed in liquid sulfuric acid. Several modifications to the electrolyte are
used to improve battery performance in one of several areas. The key parameters of the electrolyte which control the performance of
the battery are the volume and concentration of the electrolyte and forming a 'captive' electrolyte.

5.8.5 Electrolyte Volume and Concentration

Changes in the volume of the electrolyte can be used to improve the robustness of a battery. Increasing the volume of an electrolyte
makes the battery less sensitive to water losses, and hence makes regular maintenance less critical. Adding to the volume of the battery
will also increase its weigth and reduce the energy density of the battery.

5.8.6 Captive Electrolyte Lead Acid Batteries

In 'captive' electrolyte batteries, the sulfuric acid is immobilised by either 'gelling' the sulfuric acid or by using an 'absorptive glass
mat'. Both have lower gassing compared to a flooded lead acid battery and are consequently often found in "maintenance-free" sealed
lead acid batteries.

Gelling. In a "gelled" lead acid battery, the electrolyte may be immobilized by gelling the sulfuric acid using silica gel. The gelled
electrolyte has an advantage in that gassing is reduced, and consequently, the batteries are low-maintenance. In addition, stratification
of the electrolyte does not occur with gelled batteries and therefore boost charging is not required, and because the electrolyte is gelled,
the chances of spilling sulfuric acid are also reduced. However, in order to further reduce gassing, these "gel-cell" batteries also
typically use lead calcium plates, making them unsuited to deep discharge applications. A further drawback is that the charging
conditions of a gelled lead acid battery must be more carefully controlled to prevent overcharging and damage to the battery.

Absorbative Glass Matting. A second technology which can be used to immobilize the sulfuric acid is "absorptive glass mat" or AGM
batteries. In an AGM battery, the sulfuric acid is absorbed in a fiberglass mat which is placed between the electrodes plates. AGM
batteries have numerous advantages including the ability to be deeply discharged without affecting lifetime, allowing high rates of
charge/discharges and an extended temperature range for operation. The key disadvantage with these batteries is their need for more
carefully controlled charging regimes and their higher initial cost.

Characteristics of Lead Acid Batteries

For most renewable energy systems, the most important battery characteristics are the battery lifetime, the depth of discharge and the
maintenance requirements of the battery. This set of parameters and their inter-relationship with charging regimes, temperature and age
are described below.

Depth of Discharge and Battery Capacity

The depth of discharge in conjunction with the battery capacity is a fundamental parameter in the design of a battery bank for a PV
system, as the energy which can be extracted from the battery is found by multiplying the battery capacity by the depth of discharge.
Batteries are rated either as deep-cycle or shallow-cycle batteries. A deep-cycle battery will have depth of discharge greater than 50%,
and may go as high as 80%. To achieve the same useable capacity, a shallow-cycle battery bank must have a larger capacity than a
deep-cycle battery bank.

In addition to the depth of discharge and rated battery capacity, the instantaneous or available battery capacity is strongly affected by
the discharge rate of the battery and the operating temperature of the battery. Battery capacity falls by about 1% per degree below
about 20°C. However, high temperatures are not ideal for batteries either as these accelerate aging, self-discharge and electrolyte
usage. The graph below shows the impact of battery temperature and discharge rate on the capacity of the battery.

Battery Lifetime

Over time, battery capacity degrades due to

sulfation of the battery and shedding of
active material. The degradation of battery
capacity depends most strongly on the
interrelationship between the following
parameters:

the charging/discharging regime

which the battery has experienced
the DOD of the battery over its life
its exposure to prolonged periods of
low discharge
the average temperature of the battery
over its lifetime

The following graph shows the evolution of Figure: Relationship between battery capacity, temperature and discharge rate.
battery function as a number of cycles and
depth of discharge for a shallow-cycle lead
acid battery. A deep-cycle lead acid battery should be able to maintain a cycle life of more than 1,000 even at DOD over 50%.

In addition to
the DOD, the
charging
regime also
plays an
important part
in determining
battery
lifetime.
Overcharging
or
undercharging
the battery Figure: Relationship between battery capacity, depth of discharge and cycle life for a shallow-cycle battery.
results in either
the shedding of
active material
or the sulfation of the battery, thus greatly reducing battery life.

The final impact on battery charging relates to the temperature of the

battery. Although the capacity of a lead acid battery is reduced at
low temperature operation, high temperature operation increases the
aging rate of the battery.

Figure: Impact of charging regime of battery capacity.

 Figure: Relationship between battery capacity, temperature and lifetime for a deep-cycle battery.

Constant current discharge curves for a 550 Ah lead acid battery at different discharge rates, with a limiting voltage of 1.85V per cell
(Mack, 1979). Longer discharge times give higher battery capacities.

Maintenance Requirements

The production and escape of hydrogen and oxygen gas from a battery cause water loss and water must be regularly replaced in lead
acid batteries. Other components of a battery system do not require maintenance as regularly, so water loss can be a significant
problem. If the system is in a remote location, checking water loss can add to costs. Maintenance-free batteries limit the need for
regular attention by preventing or reducing the amount of gas which escapes the battery. However, due to the corrosive nature the
elecrolyte, all batteries to some extent introduce an additional maintenance component into a PV system.

Battery Efficiency

Lead acid batteries typically have coloumbic efficiencies of 85% and energy efficiencies in the order of 70%.

Lead Acid Battery Configurations

Depending on which one of the above problems is of most concern for a particular application, appropriate modifications to the basic
battery configuration improve battery performance. For renewable energy applications, the above problems will impact the depth of
discharge, the battery lifetime and the maintenance requirements. The changes to the battery typically involve modification in one of
the three basic areas:

changes to the electrode composition and geometry

changes to the electrolyte solution
modifications to the battery housing or terminals to prevent or reduce the escape of generated hydrogen gas.

Operation of Lead Acid Batteries

A lead acid battery consists of a negative electrode made of spongy or porous lead. The lead is porous to facilitate the formation and
dissolution of lead. The positive electrode consists of lead oxide. Both electrodes are immersed in a electrolytic solution of sulfuric
acid and water. In case the electrodes come into contact with each other through physical movement of the battery or through changes
in thickness of the electrodes, an electrically insulating, but chemically permeable membrane separates the two electrodes. This
membrane also prevents electrical shorting through the electrolyte. Lead acid batteries store energy by the reversible chemical reaction
shown below.

The overall chemical reaction is:

Lead Acid Overall Reaction

Edit
Delete
Clone

Read more about Lead Acid Overall Reaction

At the negative terminal the charge and discharge reactions are:

Lead Acid Negative Terminal Reaction

Edit
Delete
Clone

Read more about Lead Acid Negative Terminal Reaction

At the positive terminal the charge and discharge reactions are:

Lead Acid Positive Terminal Reaction

Edit
Delete
Clone

Read more about Lead Acid Positive Terminal Reaction

As the above equations show, discharging a battery causes the formation of lead sulfate crystals at both the negative and positive
terminals, as well as the release of electrons due to the change in valence charge of the lead. The formation of this lead sulfate uses
sulfate from the sulfuric acid electrolyte surrounding the battery. As a result, the electrolyte becomes less concentrated. Full discharge
would result in both electrodes being covered with lead sulfate and water rather than sulfuric acid surrounding the electrodes. At full
discharge, the two electrodes are the same material, and there is no chemical potential or voltage between the two electrodes. In
practice, however, discharging stops at the cutoff voltage, long before this point. The battery should not, therefore, be discharged below
this voltage.

In between the fully discharged and charged states, a lead acid battery will experience a gradual reduction in the voltage. Voltage level
is commonly used to indicate a battery's state of charge. The dependence of the battery on the battery state of charge is shown in the
figure below. If the battery is left at low states of charge for extended periods of time, large lead sulfate crystals can grow, which
permanently reduces battery capacity. These larger crystals are unlike the typical porous structure of the lead electrode, and are difficult
to convert back into lead.

Voltage of lead acid battery upon charging.

The charging reaction converts the lead sulfate at the negative electrode to lead. At the positive terminal the reaction converts the lead
to lead oxide. As a by-product of this reaction, hydrogen is evolved. During the first part of the charging cycle, the conversion of lead
sulfate to lead and lead oxide is the dominant reaction. However, as charging proceeds and most of the lead sulfate is converted to
either lead or lead dioxide, the charging current electrolyzes the water from the electrolyte and both hydrogen and oxygen gas are
evolved, a process known as the "gassing" of the battery. If current is being provided to the battery faster than lead sulfate can be
converted, then gassing begins before all the lead sulfate is converted, that is, before the battery is fully charged. Gassing introduces
several problems into a lead acid battery. Not only does the gassing of the battery raise safety concerns, due to the explosive nature of
the hydrogen produced, but gassing also reduces the water in the battery, which must be manually replaced, introducing a maintenance
component into the system. In addition, gassing may cause the shedding of active material from the electrolyte, thereby permanently
reducing battery capacity. For these reasons, the battery should not regularly be charged above the voltage which causes gassing. The
gassing voltage changes with the charge rate.

Lead sulphate is an insulator, and therefore the way in which lead sulfate forms on the electrodes determined how easily the battery can
be discharged.

Other Battery Types

6.1 Alkaline Battery

The alkaline battery consists of a redox reaction in which the anode is Zn (which also usually the casing) with an electrolyte consisting
of a paste of NaOH or KOH. The half reactions are:

Alkaline Battery Reaction

Edit
Delete
Clone

Read more about Alkaline Battery Reaction

In this battery, if one mole of zinc is present, and two moles of MnO2, then the Zn is the limiting component for the reaction.

6.2 Nickel-Cadmium Batteries

While lead-acid batteries are undoubtedly the most commonly used batteries in photovoltaic systems, in some photovoltaic
applications, nickel-cadmium may be cost effective on a life-cycle/cost basis. Nickel-cadmium batteries consist of a positive electrode
of nickel (or hydroxide) and a negative electrode of cadmium hydroxide. They are commonly used in a sealed configuration in small
household appliances, but larger vented or sealed batteries are also available for PV applications. Nickel-cadmium batteries have
several advantages as listed below.

Long lifetime and long storage life. In nickel-cadmium batteries, the positive and negative electrodes undergo oxidation and reduction
reactions. Material does not enter the electrolyte and then re-plate to the electrodes as it would in lead-acid batteries. This means that
the active material does not shed from the plates, and that a process analagous to sulfation of a lead-acid battery does not occur. As
these processes reduce the lifetime of lead-acid batteries, nickel-cadmium batteries have a higher lifetime. Furthermore, the electrolyte
in nickel-cadmium is less corrosive to battery parts than in a lead-acid battery which also increases lifetime.

Can be fully discharged. Nickel-cadmium batteries can be fully discharged without damage to the battery.

Can be overcharged. Nickel-cadmium batteries are less sensitive to overcharging, thereby reducing the requirements during the
charging regime. Due to the ability to completely discharge, the tolerance to overcharging and the charging regimes for these batteries,
in some cases the battery regulator may be eliminated.

Reduced sensitivity to temperature. Since the electrolyte composition does not change during charging or discharging, nickel-cadmium
batteries are not more susceptible to freezing at low levels of charge, in the same way that lead-acid batteries are. Consequently, nickel-
cadmium batteries are less sensitive to colder temperature, tolerating temperatures of -50 C. In addition, the lifetime of

nickel-cadmium batteries are not as strongly affected by high temperature operations as lead-acid.

Low maintenance requirements. As nickel-cadmium batteries emit fewer corrosive elements and have lower gassing, they require less
frequent maintenance.

However, they also have a number of disadvantages. Some of the disadvantages include;

Expense. Nickel-cadmium batteries are typically at least twice as expensive than lead-acid batteries. However, some of this cost may
be offset by the ability to fully discharge, eliminating the need for oversizing the battery, and by the possible elimination of the
regulator. Consequently, in applications which are not critical, nickel-cadmium batteries can be used, assuming that they will be nearly
fully discharged each night. If, however during a charging cycle there is a cloudy day, then no power would be available. Nickel-
cadmium batteries, therefore can only be used in non-critical loads.

Lower efficiencies. Nickel-cadmium batteries have both lower coloumbic efficiencies, between 75% to 85%, and lower overall
efficiencies, between 60% to 75%.

Memory effect. Some nickel-cadmium batteries can require full discharge to prevent "memory" development, and subsequent inability
(in a normal discharge cycle) to discharge below the level it has been subjected to in the past. Elimination of this effect requires a slow,
full discharge/charge cycle.

An additional feature of nickel-cadmium batteries is the relatively constant voltage curve on charging and discharging. While this is an
advantage in discharging in that the voltage stays relatively constant between 10% and 80% discharge, it is a disadvantage in charging
in that the voltage is a poor indicator of battery state of charge and therefore determining SOC is more difficult.

6.3 Vanadium Redox Flow Battery

Redox flow batteries use a reductio-oxidation between two valence states in solution rather than changing the composition, and hence
the valence states of solid material on an electrode. A flow battery consists of two volumes of solution separated by a selective
membrane which allows some ions to pass but not others. The two solutions are pumped to the permeable membrane, which allows
xxxx.

Flow batteries have several potential advantages over solid batteries. A key advantage, which is particularly important in transport
applications, is that the battery may be re-charged simply by pumping out the uncharged solution and replacing the solution with
charged solution. This eliminates potentially long recharging times, such as are encountered in electric vehicles. Replacement of the
solution allows the electric car to be recharged in the same fashion in which a car is filled with fuel. Another advantage is that the
capacity of the battery is determined by the volume of solution, while the power of the battery is determined by the membrane contact
area between the two solutions.

The vanadium-Vanadium redox flow battery, developed at the University of New South Wales, is a particularly promising flow battery.
It consists of two states of Vanadium. It has high efficiencies, with coulombic efficiencies of 97% and energy efficiencies of 87%. In
addition, since both solutions (anode and cathode) in the battery use vanadium, cross contamination between the two solutions may
discharge the battery, but will not cause damage to the battery.
Function and Use of Storage
Storage is used in PV systems to increase the amount of time that the PV system can be used to power a load. Batteries are the most
common type of storage in a PV systems. However, in specific types of systems or applications, other storage components can also be
used. For example, in water pumping systems, the amount of battery storage can be greatly reduced or eliminated if extra water is
pumped and stored in a water tank for use in cloudy periods.

In stand alone electricity generating systems, some form of storage is needed unless the load is exactly matched to the time during
which the sun is shining. (Such an exact match is rare and limited to a few types of systems - for example powering a fan for cooling or
in some cases water pumping for irrigation). In stand alone systems, storage is needed not only to power loads at night, but also allow a
load to operate during cloudy weather. The number of days of storage needed depends on the weather pattern at a particular location,
with cloudier locations needing more storage. In systems with a large amount of storage, and additional utility of the storage system is
that is can buffer the system against periods of low insolation, such as in winter. For example, in telecommunications systems that
require high reliability, a large battery bank can allow high reliability without requiring the PV array to be sized to meet the worst
possible insolation conditions. In general, the larger the amount of storage included, the less sensitive the system will be to periods of
low insolation, and the more reliable the power availability will be. The figure below shows how the power availability increase with
increasing storage.

Battery state of charge over a year showing the battery discharge overnight, during cloudy weather and seasonal variations

In systems connected to the utility grid electricity supply, storage is typically not needed. PV power is used when the sun is shining,
and at night or during periods of cloudy weather, the grid provides the electricity. However, even in grid-connected systems, storage
can be included, not to increase the reliability of having power as in a stand-alone system but rather to increase the value of the PV-
generated electricity. In the load seen by many utility companies, an air conditioning load occurring on summer afternoons increases
the overall load that the utility must supply. These peaks in the load are significantly more expensive to supply power for. Since the
power output from PV is typically largest during summer months, the output from the PV system can well-matched to the peak load of
the generated electricity is stored for a few hours. The use of storage for this application is called peak shifting and is shown in the
figure below.

Functions of Batteries
Batteries are a common feature in most types of PV systems that are not connected to the utility grid. In addition to providing storage,
batteries can also be used for several other functions:

Storage. Batteries store energy being produced by a given generating source, and when this source is unavailable this energy can be
used by the load. The inclusion of storage in any energy generating system will increase the availability of the energy.

Start-up current. Batteries can provide higher currents to the load than the array alone can provide. This is especially useful if a
particular load has a high current draw on start-up. Many motors initially have a high current requirement.

Power conditioning.Batteries can function as power conditioning. Two cases where this feature is used is in directly coupled systems,
such as water pumping, and in uninterruptable power supplies.

In addition to the different mode of operation, batteries in photovoltaic systems also must meet several other criteria. As reliability and
low maintenance are desirable in photovoltaic systems, the batteries must also have a long lifetime. Further, since batteries will often
be a substantial fraction of the total cost of a PV system, cost is a significant factor in batteries for PV systems. In general, batteries
manufactured for other applications are not well suited to photovoltaic energy applications. The key characteristics of a battery in a
renewable energy system are:

efficiency of the battery

how battery capacity and lifetime is affected by deep cycling and extended states of low charge
the initial and ongoing battery costs
the maintenance requirements of the battery.

Electrolysis of Water
In battery solutions in which a component of the electrolyte is water (such as in lead acid batteries), the possibility of electrolysis water
must be taken into account when charging a battery. The electrolysis of water, which is breaking water into oxygen and hydrogen.

According to the standard potentials, the voltage of this reaction is 1.23V. However, the activation overpotential of this reaction is
large, and hence it does not proceed at a significant rate (and can therefore be neglected in battery charging or discharging) until
voltages on the order of 2.2V are reached in the battery. During high charging rates, the charging voltage may exceed this voltage, and
hence two reactions will proceed in such a battery: one the charging of the battery and the second the electrolysis of water. As the
electrolysis of water gives of hydrogen and oxygen, both of which are gases, the battery is said to be gassing. The electrolysis of water
has several impacts on the battery. Firstly, it leads to water loss in the battery, which must be replaced. Further, the evolution of
hydrogen gas forms a potential safety hazard if released in an improperly ventilated area, or can overpressure the battery case. Both of
these issues may be minimized or circumvented by preventing the gases, the battery is said to be gassing. The electrolysis of water has
several impacts on the battery. Firstly, it leads to water loss in the battery, which must be replaced. Further, the evolution of hydrogen
gas forms a potential safety hazard if released in an improperly ventilated area, or can overpressure the battery case. Both of these
issues may be minimized or circumvented by preventing the gases, particularly the hydrogen from escaping from the battery. Batteries
using this approach are called sealed or recombinant batteries. Despite the potential maintenance and safety problems associated with
gassing, it may also have beneficial impacts. For example, in lead-acid batteries gassing can be used to mix the electrolyte, thus
preventing regions of higher sulfuric acid concentration (which is denser) from sinking to the bottom (an effect called stratification).

The electrolysis of water is affected by the presence of small amounts of impurities in the lead acid batteries, and hence batteries with
additives to the lead (for mechanical strength or other practical purposes) can experience significantly different gassing voltages.
Further, since the activation energy is temperature dependent, the voltage at which gassing of a battery changes with the battery
temperature and on the details of the battery components.

Uses of batteries in PV systems

While the primary function of a storage system is to provide power when sunlight is not available, hence increasing the fraction of time
the photovoltaic system provides electricity, the addition of batteries has numerous other advantages which mean that the batteries can
be used for multiple purposes. For small systems consisting of one or two photovoltaic modules, batteries can act as a load-matching
system. Alternately, in photovoltaic systems which contain a load with a large initial current draw (such as experienced by an inductive
load, typically represented by a motor), the batteries can be used to provide initial start-up current. In grid-connected systems, battery
storage can be used for peak shifting, in which the power generated by the sun is stored for several hours in order to better match when
the peak load occurs.

Chapter 10 Quiz
Questions 1
Attempts allowed Unlimited
Available Always
Pass rate 75 %
Backwards navigation Allowed
Start Quiz

Korean Version PDF

Add new comment
1 comment(s)
The PVCDROM is now in Korean as 태양광 개론. The translaton is available as a single 12 Mb PDF at:

http://www.kier.re.kr/upload/notic/PVCDROM-KOR.pdf

The translation was done by 윤경훈 Yoon, Kyung-Hoon at the Solar Cells Research Center, Korea Institute of Energy Research. Their
website is at: http://solarpv.kier.re.kr