This article was downloaded by: ["Queen's University Libraries, Kingston"]

On: 29 September 2014, At: 15:17

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Instrumentation Science & Technology

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/list20

FABRICATION OF AUTOMATIC
ELECTROLYTE FILLING MACHINE FOR
DYE-SENSITIZED SOLAR CELLS
a a
P. V. V. Jayaweera & S. Kaneko
a
SPD Laboratory, Inc., HI-Cube , Wajiyama , Hamamatsu , Japan
Accepted author version posted online: 11 Jun 2012.Published
online: 07 Nov 2012.

To cite this article: P. V. V. Jayaweera & S. Kaneko (2012) FABRICATION OF AUTOMATIC ELECTROLYTE
FILLING MACHINE FOR DYE-SENSITIZED SOLAR CELLS, Instrumentation Science & Technology, 40:6,
490-503, DOI: 10.1080/10739149.2012.692142

To link to this article: http://dx.doi.org/10.1080/10739149.2012.692142

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 Instrumentation Science and Technology, 40:490–503, 2012
Copyright # Taylor & Francis Group, LLC
ISSN: 1073-9149 print/1525-6030 online
DOI: 10.1080/10739149.2012.692142

FABRICATION OF AUTOMATIC ELECTROLYTE FILLING MACHINE

FOR DYE-SENSITIZED SOLAR CELLS
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

P. V. V. Jayaweera and S. Kaneko

SPD Laboratory, Inc., HI-Cube, Wajiyama, Hamamatsu, Japan

& An electrolyte filling machine, capable of filling liquid electrolyte (including viscous ionic
liquid) into an enclosed cell with a single hole, has been developed. This system utilizes a
vacuum-aided filling technique and has four major steps in the filling process, namely evacuation,
discharging, filling, and returning extra electrolyte. The electrolyte is not open to ambient con-
ditions throughout the filling process, and the return step minimizes the waste of electrolyte. This
technique allows air bubble–free filling, better insertion of electrolyte into the mesoporous titanium
oxide layer of dye-sensitized solar cells (DSC), and filling time is 40 s with a nonvolatile
liquid-based electrolyte. Large area DSC modules that are 15
15 cm in size were tested using this
machine, and conversion efficiency of 7.4% was obtained under the standard air mass (AM) 1.5
simulated sunlight.

INTRODUCTION
In order to face the rapidly growing energy demand, the world is now
searching for more and more safe green energy sources rather than poten-
tially dangerous nuclear power plants. The photovoltaic cell is a clean
renewable energy source that converts sunlight directly into electricity.
The dye-sensitized solar cell (DSC), which was invented by O’Regan and
Grätzel,[1] gets more attention as a low-cost alternative technique. In the
simplest form, DSC is composed by sandwiching two plates so called photo-
and counter-electrodes. The capillary space between these electrodes is
filled with low-volatile-liquid[2] or ionic-liquid electrolyte containing
iodide=tri-iodide redox couple. In general, the photo-electrode consists
of a dye-coated mesoporous TiO2 layer deposited on a glass with a

Address correspondence to P. V. V. Jayaweera, SPD Laboratory, Inc., HI-Cube, Wajiyama,

Hamamatsu 432-8003, Japan. E-mail: viraj@spd-lab.com
 Automatic Electrolyte Filling Machine 491

transparent conducting oxide (TCO) layer of fluorine-doped tin oxide

(FTO). The counter-electrode can have lightly platinized transparent con-
ducting glass, Cr=Pt sputtered glass,[3] or lightly platinum coated titanium
metal plate.[4] Typically, photo- and counter-electrodes are bonded
together using UV curing glue[3] or a hot melt sheet[5] (Surlyn by DuPont)
making a 30–50 mm gap between the two electrodes. A silver grid-line pat-
tern is printed on the FTO layer to efficiently collect charge carriers, which
also separate the TiO2 coated area into strips. Depending on the design,
there are several TiO2 strips in the DSC (e.g., 8 strips in 10
10 cm2 mod-
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

ule and 12 strips in 15
15 cm2 module), and each strip should have two
holes (on the counter-electrode) in order to fill the electrolyte by using
conventional filling methods. In the literature, Sastrawan et al.[6] reported
a coloration and electrolyte-filling machine that was used for their glass frit
sealed DSC modules. It needs two holes per cell and first circulates dye
through the cell. As explained in their article, ‘‘After the coloring is com-
plete, the electrolyte is pumped through the module for about 1 min until
the cells are clean from residues of the dye solution. Then the process is
stopped and the electrolyte remains in the module. (p. 1685)’’[6] This
method has practical problems when using large scale production due to
waste of electrolyte to wash the dye inside the cell. The main disadvantages
of the two-hole filling methods are described in detail in the next para-
graph. A few other publications[2,7] have mentioned the term ‘‘vacuum
back-filling system’’ that they have used to fill the cells, but details or refer-
ences were not provided. The well-known vacuum back-filling approach is
to vacuum a DSC module dipped in an electrolyte bath. For this method,
the electrolyte bath with the DSC module should be placed in a vacuum
chamber, then the chamber should be vacuumed and the vacuum should
be released. Air inside the DSC will bubble through the electrolyte during
the vacuum period, and electrolyte is entered to the DSC when the vacuum
releases. This technique has several disadvantages: (i) When vacuuming,
the electrolyte can start to boil (especially acetonitrile-based electrolytes
can easily boil, even at very low vacuum levels) and change the composition
of the electrolyte; (ii) this method needs a large amount of electrolyte to
dip the whole DSC module and a considerable amount of the electrolyte-
coated outer surface of the DSC is wasted during the cleaning; and (iii)
the electrolyte can be easily contaminated with the silver-electric contact
pads outside the DSC, if not masked properly.
Figures 1a and 1b show two typical approaches (usually manual opera-
tions) used to inject electrolyte into the cell. Both of these methods need
two holes in the cell to perform the operation. In the first method, a needle
inserted into the filling hole injects electrolyte into the cell, while a second
hole stays open to release the air inside the cell. In the second method
(shown in Figure 1b), the needle is replaced by a cylindrical tip with a
 492 P. V. V. Jayaweera and S. Kaneko
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

FIGURE 1 Advantages of a new electrolyte-filling method over typical double-hole filling approaches.
(a) Schematic cross-section of needle-aided filling technique showing trapped air bubbles at the corners
and electrolyte leaks around the hole. (b) Schematic cross-section of an o-ring–based filling tip also gen-
erates air bubbles at the corners. (c) The proposed new single-hole, vacuum-aided filling technique pro-
vides air bubble–free filling. (d) The top view of the 15 cm
15 cm DSC has 12 parallel strips. Typical
filling methods need 24 holes (two holes per strip), but the new technique needs just one hole to fill
an entire cell. (color figure available online.)
 Automatic Electrolyte Filling Machine 493

rubber o-ring to achieve good sealing between the cell surface and filling
tip. These techniques have several drawbacks:

1. Air bubbles can get trapped in the corners of the cells as shown in
Figures 1a and 1b. Cells should be as straight as possible without corners
or bends, and filling holes should be placed very close to the edges to
minimize the air bubbles.
2. Wasting of electrolyte, leaks (overflow) around the needle and hole
3. Wasting of extra electrolyte remaining in the bottom part of the tip
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

when removing filling tip from the DSC

4. Exposure of electrolyte into ambient air, allowing water vapor absorption
5. The necessity of drilling two holes for each cell
6. Difficulties in automating the process for large scale production line. It
has been shown that if water vapor enters the cell, the lifetime of DSC is
reduced. Therefore it is very important to minimize the exposure of
electrolyte to ambient conditions throughout the filling process.[8]

In this article, we are introducing a new vacuum-aided filling technique that

eliminates the limitations of the previous approaches. As shown in Figure 1c,
this method only needs one hole per cell. As the DSC schematic Figure 1d
shows, it needs only one hole to fill the entire cell using new method. The
typical approaches needs 24 holes to fill it, and takes about 12 times longer
time to complete the job. This new technique utilizes a filling tip that has two
homocentric tubes. In the first step, the outer tube is used to evacuate the
cell. Here we show unique feature in this technique: the electrolyte is not
in the vicinity during the evacuation period, unlike conventional ‘‘vacuum
back-filling,’’ which is evacuated the DSC through an electrolyte column.
In the second step, an inner tube is used to release electrolyte into the filling
tip. In this method, a large amount (several times larger than cell volume) of
electrolyte can be released to the tip without measuring the actual cell vol-
ume. When the pressure is released, electrolyte is automatically sucked into
the cell and fills the entire cell. In the final step, extra electrolyte remaining
in the filling tip is pumped back to the bottle through the middle tube. The
proposed completely automated new filling process can be directly applied
to an industry-scale production line. A detailed description of the working
principle, control software, and experimental results of DSC modules pre-
pared using this filling machine is presented in this article.

DESIGN AND OPERATING PRINCIPLES OF VACUUM-AIDED

FILLING TECHNIQUE
The filling process of this machine is governed by controlling the
vacuum and nitrogen supplies through four electric solenoid valves.
 494 P. V. V. Jayaweera and S. Kaneko

Electrolyte flow is controlled by an air-operated valve inside the filling head.

This system can be divided to three main components: the head, the elec-
trolyte container, and the vacuum and nitrogen flow control valves.

Filling Head
The heart of the machine is the filling head, which allows evacuation of
the cell, filling electrolyte, and safely returning extra electrolyte remaining
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

in the tip to the bottle. As shown in Figure 2, the filling-head has three
ports, named as A, B, and C, and a rubber tip that can be placed on the
filling hole of the DSC. The rubber tip, which is made of silicone, has a
5 mm hole in the middle and directly opens to Port A. The bottom end
of the rubber tip acts like a vacuum seal with the surface of DSC. A tube
positioned in the middle of the rubber tip is used to transport the electro-
lyte, which is connected to Port B, through a pneumatic valve F. When
compressed air is supplied through Port C, the piston is moved to left side
by overcoming spring tension and opens Port B to the middle tube of the
rubber tip. This filling head is mounted on a vertically moving linear actu-
ator (z axis) so that it can be easily moved down to the filling hole on the

FIGURE 2 Cross-section of filling head positioned on a DSC module’s filling hole. (color figure avail-
able online.)
 Automatic Electrolyte Filling Machine 495

DSC before starting the filling process and moved up from the DSC after
finishing the process.

Flow Control Design

The schematic diagram of the complete electrolyte filling system is
shown in Figure 3. The machine itself contains the filling head, two bottles,
five solenoid valves (KOGANEI 030E1 series, Japan), a digital vacuum
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

sensor=switch (SUNX DP2-20Z, Japan), a N2 pressure regulator (KOGANEI

PR-200, Japan), and USB data acquisition module. Solenoid valves A and B
are two-port positive pressure type, solenoids C and D are two-port vacuum
type, and solenoid valve E is a three-port positive pressure type. This
machine requires an external vacuum pump, air compressor, and N2 gas

FIGURE 3 Schematic diagram of the electrolyte-filling system showing the N2, vacuum, and com-
pressed air pathways and solenoid valves with electronic control system. (color figure available online.)
 496 P. V. V. Jayaweera and S. Kaneko

supply. Nitrogen pressure is regulated to þ15 kPa by a pressure regulator at

the entrance port. Port A of the filling head is connected to an empty bottle
that acts like a safety trap to prevent electrolyte entering into the solenoids
and vacuum pump area, if there is accidental overflow from the filling
head. This empty bottle also works as a washing liquid collector during
the automatic cleaning process, which will be described later under the sec-
tion Self Cleaning. As shown in Figure 3, the left side empty bottle is con-
nected to a vacuum sensor, solenoid valve A, and solenoid valve C. This
configuration allows us to connect the vacuum pump or N2 supply to Port
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

A by opening solenoid valve C or A. The vacuum sensor has an analog volt-

age output that is directly read through the USB data acquisition module,
and two digital outputs (OUT 1, OUT 2) that are turned to digital ‘‘high’’
state when vacuum level reaches two preset pressure values: P1 and P2,
respectively. Port B of the filling head is connected to the electrolyte bottle,
and a second tube coming from the bottle is connected to solenoid valve D
and solenoid valve B. This configuration allows increasing or decreasing the
pressure in the electrolyte bottle by connecting it to the N2 supply or vac-
uum pump through the corresponding solenoid valve. Port C is connected
to the compressed air supply through the solenoid valve E that is used to
control electrolyte flow valve F inside the filling head. A ommercially avail-
able USB data acquisition board (16 digital inputs=outputs and 12-bit ana-
log to digital inputs) combined with a home build optocoupler-based
driving circuit is used to control the solenoid valves through a computer
program. The vacuum pump is powered through a relay and can be turned
on=off through the computer program, which allows power savings by
turning off the pump when the machine is idle for a long time.

ELECTROLYTE FILLING PROCEDURE

The electrolyte filling process can be divided into four main steps, as
illustrated in Figure 4, and the operation principle of each step is described
below. The tube connecting the electrolyte bottle and Port B of the filling
head should be completely filled with electrolyte before starting this filling
procedure. This filling tube operation will be explained in a later section
called Removing Air From the System (Nitrogen Purging).

1. Evacuation of the Cell

In the first step, the interior of the DSC is evacuated by opening solen-
oid valve C until the vacuum level reaches the predetermined value P2
(e.g., 99.8 kPa). All other solenoid valves are in the closed position during
this operation. Please see Figure 4(1). Solenoid-valve C is closed after reach-
ing the desired vacuum level and proceeds to the next step.
 Automatic Electrolyte Filling Machine 497
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

FIGURE 4 The four main steps of the electrolyte filling procedure: 1. Evacuation of the cell, 2. dis-
charge of electrolyte, 3. fill into the cell, and 4. return extra electrolyte. (color figure available online.)

2. Discharge Electrolyte
In the second step [see Figure 4(2)], electrolyte is released into the rub-
ber tip by opening solenoid valve E. Due to the pressure difference in the
cell and the electrolyte bottle, electrolyte rapidly flows through the middle
tube and is collected inside the rubber tip (only a small amount is trans-
ferred into the cell, since inside the cell and tip have the same low pressure,
and the surface tension across the 1 mm diameter filling hole). Solenoid
valve E is closed when the electrolyte level inside the rubber-tip reaches
about two-thirds of its height. This discharged amount is not a critical fac-
tor; the requirement is that the volume of electrolyte collected in the rub-
ber tip should be greater than the volume of the vacant space inside DSC
for air bubble–free filling.

3. Filling the Cell

As shown in Figure 4(3), in the third step, nitrogen is sent through Port
A by opening solenoid valve A, which increases the pressure inside the
 498 P. V. V. Jayaweera and S. Kaneko

rubber tip and pushes the electrolyte, collected in the tip, into the cell. In
the real control program, solenoid valve A is opened for short intervals
while monitoring the vacuum gauge in order to slowly increase the pressure
inside the tip. During this period, solenoid valve E also stays open to release
more electrolytes into the tip and maintain the electrolyte level inside
the tip.

4. Return Extra Electrolyte

Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

As the last step, extra electrolyte remaining in the rubber tip and the
middle tube needs to be returned to the electrolyte bottle before lifting
off the filling head from the cell. In order to perform this, solenoid valve
D is opened for few seconds to lower the pressure in the electrolyte bottle,
and then solenoid valve E is turned on to open the valve F inside the filling
head. Electrolyte flows from the rubber tip to the bottle due to the pressure
difference [see Figure 4(4)]. In order to maintain pressure in the rubber
tip, solenoid valve A is opened for short periods while continuously moni-
toring the vacuum gauge reading. Pressure inside the rubber tip is kept
slightly lower than atmospheric pressure (5 kPa) to prevent leaking elec-
trolyte through the rubber tip and DSC surface. After returning all electro-
lytes, pressure inside the rubber tip is increased to 1 atm. Finally, all the
open solenoids are closed and the filling head is raised from the surface
of DSC.

ASSOCIATE FUNCTIONS
There are several important associate functions described below that
must be performed before and after the basic filling procedure described
in the previous section. These functions minimize the exposure of electro-
lyte in air by keeping the system under N2 atmosphere, ensure 100% air
bubble–free filling, minimize the waste of electrolyte by returning the
electrolyte trapped inside the filling head and tubes, and help to clean
the system without disassembling.

Removing Air From the System (Nitrogen Purging)

Air inside the tubes and the electrolyte bottle should be replaced with
N2 whenever new electrolyte bottle is connected to the machine. This is
performed by opening solenoid valve D to vacuum the system, and then
opening solenoid valve B to fill the system with N2 (see Figure 3). This pro-
cedure can be repeated several times to ensure a water vapor–free, dry N2
environment for the electrolyte. During this procedure, electrolyte is filled
 Automatic Electrolyte Filling Machine 499

in the tube that connects Port B of the filling head and the electrolyte bot-
tle. This associate function is automatically performed, just before starting
the electrolyte filling procedure.

Returning Electrolyte
This function will help to return the entire electrolyte in the filling
head and the connecting tube, just before removing the bottle from the sys-
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

tem. As a first step, the Port A side of the filling head is filled with N2 to
prevent normal air bubbles from forming through the electrolyte during
the returning process. In order to do this, the filling head is moved down
to the cell holding table, and the head through Port A is evacuated by open-
ing solenoid valve C. After that, solenoid valve A is opened to fill the head
with N2. In the second step, solenoid valve D is opened for few seconds to
reduce the inside pressure of the electrolyte bottle, then solenoid valve E is
opened to return electrolyte to the bottle. The pressure in Port A is main-
tained as 1 atm by opening solenoid valve A intermittently while monitoring
the vacuum sensor.

Pressure Balancing
This function will fill the N2 to the filling head, tubes, and bottles until
the pressure is equal to atmospheric pressure. This is important whenever
removing two bottles from the system for maintenance, or after an emerg-
ency stop in the middle of filling process where the internal low pressure
keeps the cell stuck to the filling tip. In the normal operation, at the end
of filling procedure, the inside pressure of the filling head and two bottles
will be set to 1 atm by releasing N2.

Self Cleaning
In a situation such as changing to a different type of electrolyte, or stor-
ing the machine without using it for a long time, the filling head and the
tubes need to be cleaned in order to prevent contamination and clogging
inside the filling head. In the self-cleaning process, the electrolyte bottle is
replaced by a cleaning liquid (acetonitrile or 3-Methoxypropionitrile,
depending on the solvent used in the electrolyte) and the liquid is pumped
to the second bottle (empty bottle in Figure 3) through the filling head
intermittently to make the liquid flow turbulent. First, the filling head is
moved down to the cell holding table to close the open end of the rubber
tip. Solenoid valve C is opened to reduce the pressure inside the empty bot-
tle (see Figure 3) and the Port A side of the filling head. Solenoid valve B is
 500 P. V. V. Jayaweera and S. Kaneko

opened to increase the pressure in the cleaning liquid bottle. Then valve F
is opened and closed at 1-second intervals to inject the liquid into the rub-
ber tip and flow it through the tube connected to Port A and collected in
the second bottle. This process is stopped after 10 on=off cycles of Valve F,
and then the return electrolyte function described under the Returning
Electrolyte Section is performed. This complete process is repeated until
the cleaning liquid bottle is empty. Finally, the filling head is moved up,
and the last drops of cleaning liquid trapped inside the Port B side tubes
and middle tube of the filling head are removed by opening solenoid valve
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

B to pump N2 through the middle tube.

CONTROL SOFTWARE
A state-of-the-art, user-friendly computer program has been developed
with Microsoft Visual Basic to control the filling machine through the
USB port of a PC. Electrolyte filling operation and each associate function
described in the previous section were assigned to a command buttons on
the graphical user interface (GUI). User can execute those operations by
clicking the mouse on the corresponding command button on the com-
puter screen.

TESTING OF FILLING MACHINE OPERATION

Several dummy cells made by bonding two glass plates (one glass plate
has a hole with diameter 1 mm) with a 30 mm spacer using UV curing glue
were tested with the machine. After confirming successful filling operation,
a real DSC module (15
15 cm2 in size and 12 parallel connected cells) is
filled by using this machine. For this experiment, the filling machine was
connected to single-stage rotary oil vacuum pump (ULVAC model G-50S
by Sinku Kiko Co., Ltd., Japan), an oil-free air compressor (Hitachi
Bebicon, Japan), and commercial grade compressed N2 cylinder.

DSC MODULE PREPARATION

Two DSC modules were prepared to compare the new filling method
and the typical needle assisted method. FTO thin films of about 1 mm thick
were deposited on Corning #1737 glass sheets using a spray pyrolysis depo-
sition (SPD) apparatus (KM-150), which was reported previously.[9] Silver
grid lines (with 12 mm intervals) were deposited on the above FTO by dis-
pensing Ag paste (Noritake Co., Ltd., Japan) using a desktop robot with a
0.4 mm nozzle and sintered in air at 150 C, 350 C, and 500 C for 10 min in
 Automatic Electrolyte Filling Machine 501

each step. Mesoporous TiO2 stripes were deposited between silver grids
using the SPD apparatus. Here, a metal mask is used to cover the silver grid
and a 0.5 mm extra wide area on both sides of each silver grid. After sinter-
ing at 500 C for 30 min, samples were dipped into a 3 mM solution of N719
dye for 12 h. Counterelectrodes were prepared by lightly sputtering Pt
(50 nm) on 1.1 mm thick titanium metal plates. A single hole was drilled
on one counterelectrode (the position of the hole is shown in Figure 1d),
and 24 holes were drilled on the second counterelectrode (two holes per
strip). UV curing sealant (ThreeBond Chem. Co., Japan) is dispensed on
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

the top of each silver grid and the edges of the dye-coated photo-electro-
des. For each module, photo- and counterelectrodes were sandwiched,
and UV was irradiated to cure the sealant while using a mask to cover
the dye-coated area. A detailed description can be found in the litera-
ture.[3,4] The module with a single hole was filled using new apparatus with
an internal nitrogen environment. Vacuum time was set to 90 s for this sam-
ple, and the vacuum gauge reading was 100 kPa at the end of the vacuum
period. It was observed that this type of 15
15 cm DSC module containing
12 interconnected strips (see Figure 1d) with a single hole needed a rela-
tively long period of vacuuming (90 s) for proper filling without any air
bubbles. Usually, for a single strip cell (1
14 cm), only 20 s vacuum time
is needed to perfect filling. The second DSC module with 24 holes was
filled using an injection syringe and needle for the comparison. Finally, fill-
ing the halls of both DSC modules were closed by using a piece of glass and
UV sealant.

MEASUREMENTS AND RESULTS

Completed DSC modules were carefully inspected with the naked eye
for any visible air bubbles. The module filled using the new filling machine
showed complete filling without air bubbles. The module filled using syr-
inge and needle showed some visible air bubbles close to the corners of
each strip. In the second stage, an optical microscope was used to further
inspect the inside of the DSC for tiny bubbles. It was confirmed that
machine filled DSC module is bubble-free. Several groups of tiny bubbles
were observed in the cell filled with the needle and syringe. After the visual
inspections, the current voltage (I-V) characteristics of the DSC modules
were measured using an AM 1.5, 1000 Wm2 solar simulator (EKO Instru-
ments EXES-1001S, Japan) and I-V curve tracer (EKO Instruments MP-160,
Japan). Active areas of the DSC modules were measured as 151 cm2. The
measured short circuit photo-current, open circuit voltage, fill factor, and
conversion efficiency for machine filled and manually filed DSC modules
are shown in Table 1. The I-V characteristic curves of the DSC modules
 502 P. V. V. Jayaweera and S. Kaneko

TABLE 1 Measured Parameters for DSC Modules Completed by New Automatic Filling Machine and
Typical Needle and Syringe Method

Short Circuit Open Circuit Fill Conversion

Electrolyte Filling method Current (A) Voltage (V) Factor Efficiency (%)

Automatic filling machine 2.3 0.720 0.67 7.4

Manual (needle and syringe) 2.1 0.719 0.66 6.7
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

FIGURE 5 I-V characteristic curves of the DSC completed by new automatic filling machine and typical
needle and syringe method. (color figure available online.)

are shown in Figure 5. Machine-filled DSC shows a relatively high photo-

current and conversion efficiency compared to the manually filled DSC.
The higher photo-current observed in the machine-filled sample can be
due to three main reasons. (i) Better dispersion of electrolyte in to a meso-
porous titanium oxide layer due to vacuum step before introduce electro-
lyte; (ii) the air bubble–free, machine-filled sample uses 100% of the
TiO2 area to generate the photo-current compared to the manually filled
sample, which has some air spaces at the corners, or (iii) the manually filled
sample (with 24 filling holes) is exposed to air for a relatively long period of
time compared to the machine-filled sample (with single filling hole). It is
observed that the photo-current is reduced when the dye-coated TiO2 layer
is exposed to ambient conditions for a long period.

CONCLUSION
An automatic electrolyte filling machine was introduced for the DSC
production process; this design is applicable from the research level to
 Automatic Electrolyte Filling Machine 503

the industrial mass-production line. A vacuum-assisted single hole filling

technique has the following advantages: it can workwork under a N2
environment without exposing electrolyte to ambient conditions, the filling
procedure saves electrolyte, it is easy to clean, and the filling process keeps
the electrolyte air bubble–free. The feasibility of the new machine is veri-
fied by comparing 15 cm
15 cm DSC modules filled by the new machine
and the typical manual method. This machine will open a new path to auto-
mated DSC production processes. This filling technique is beneficial not
only for the DSC, but also for a large number of other applications where
Downloaded by ["Queen's University Libraries, Kingston"] at 15:17 29 September 2014

liquids need to be filled inside an enclosed cell under inert gas conditions.

REFERENCES
1. O’Regan, B.; Gratzel, M. A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO2
films. Nature 1991, 353(6346), 737–740.
2. Wang, P.; Klein, C.; Humphry-Baker, R.; Zakeeruddin, S. M.; Gratzel, M. Stable> ¼ 8% efficient
nanocrystalline dye-sensitized solar cell based on an electrolyte of low volatility. Appl. Phys. Lett.
2005, 86(12), 3, 123508–123510.
3. Kumara, G. R. A.; Kaneko, S.; Konno, A.; Okuya, M.; Murakami, K. Large area dye-sensitized solar
cells: Material aspects of fabrication. Prog. Photovoltaics Res. Appl. 2006, 14(7), 643–651.
4. Kumara, G. R. A.; Kawasaki, S.; Jayaweera, P. V. V.; Premalal, E. V. A.; Kaneko, S. Large area
dye-sensitized solar cells with titanium based counter electrode. Thin Solid Films 2011, 520(12),
4119–4121.
5. Hinsch, A.; Kroon, J. M.; Kern, R.; Uhlendorf, I.; Holzbock, J.; Meyer, A.; Ferber, J. Long-term stability
of dye-sensitised solar cells. Prog. Photovoltaics Res. Appl. 2001, 9(6), 425–438.
6. Sastrawan, R.; Beier, J.; Belledin, U.; Hemming, S.; Hinsch, A.; Kern, R.; Vetter, C.; Petrat, F. M.;
Prodi-Schwab, A.; Lechner, P.; Hoffmann, W. A glass frit-sealed dye solar cell module with integrated
series connections. Sol. Energy Mater. Sol. Cells 2006, 90(11), 1680–1691.
7. Wang, M. K.; Xu, M. F.; Shi, D.; Li, R. Z.; Gao, F. F.; Zhang, G. L.; Yi, Z. H.; Humphry-Baker, R.; Wang,
P.; Zakeeruddin, S. M.; Gratzel, M. High-performance liquid and solid dye-sensitized solar cells based
on a novel metal-free organic sensitizer. Adv. Mater. 2008, 20(23), 4460–4463.
8. Lu, H.-L.; Lee, Y.-H.; Huang, S.-T.; Su, C.; Yang, T. C. K. Influences of water in bis-benzimidazole-
derivative electrolyte additives to the degradation of the dye-sensitized solar cells. Sol. Energy Mater.
Sol. Cells 2011, 95(1), 158–162.
9. Yagi, I.; Kaneko, S. Growth of highly oriented SnO2 thin films from di-n-butyltin diacetate by spray
pyrolysis technique. Chem. Lett. 1992, 21(12), 2345–2348.