Professional Documents
Culture Documents
Development of Asymmetric Polysulfone Membrane For The Application in Tropicalized Lead Acid Battery"
Development of Asymmetric Polysulfone Membrane For The Application in Tropicalized Lead Acid Battery"
Development of Asymmetric Polysulfone Membrane For The Application in Tropicalized Lead Acid Battery"
SEPTEMBER 2004
UNIVERSITI TEKNOLOGI MALAYSIA
PSZ 19:16 (PIND. 1/97)
BORANG PENGESAHAN STATUS TESIS
JUDUL : DEVELOPMENT OF ASYMMETRIC POLYSULFONE MEMBRANE
FOR THE APPLICATION IN TROPICALIZED LEAD ACID BATTERY
TIDAK TERHAD
Disahkan oleh
Sir,
(i) The materials of this research are advanced discovery that is considered
new in Malaysia.
(ii) The materials of this research are in the process of patent filing.
(iii) The materials of this research possess high commercial value and
potential.
Thank you.
Sincerely yours,
Signature :
Name of Supervisor : Prof. Dr. Ahmad Fauzi Bin Ismail
Date : September 25th, 2004
BAHAGIAN A – Pengesahan Kerjasama*
Adalah disahkan bahawa projek penyelidikan tesis ini telah dilaksanakan melalui
kerjasama antara _____________________ dengan _______________________
Disahkan oleh: -
Pemeriksa Dalam II :
Signature :
Name of Author : Wan Ahmad Hafiz bin Wan Hassan
Date : September 25th, 2004
iii
ACKNOWLEDGEMENT
Special thanks also goes to all of the Membrane Research Unit (MRU) team
members for their continuous assistance, encouragement and moral support
throughout my study especially to the lecturers, research officers as well as the
postgraduate students. They have been like none other than the second family to me.
Thank you guys.
To my beloved family, thank you for the continuous support and advice
during this study. My appreciation also goes to all the people who have contributed,
directly or indirectly towards the establishment of this study. Above all, I thank Allah
the Almighty for His grace, mercy and guidance throughout my life.
v
ABSTRACT
The objectives of this research are to develop asymmetric polysulfone membrane for
the application in lead acid battery and to investigate the effect of fabrication
conditions on the performance of the membrane-assisted lead acid battery and its
morphology. The function of the membrane is to prevent electrolyte losses by
minimizing the rate of vaporized electrolyte disposed into the atmosphere. Lead acid
batteries in tropical countries normally face the problem of water decomposition,
caused by the combine factors of the charge-discharge process and the heat
accumulation caused by hot tropical climate and heat generated from engine
compartment. This causes unpleasant setbacks such as corrosion of parts inside the
car bonnet and having to top up the electrolyte whenever it reduces. The membrane
will be functioning as a water-retaining device when applied onto the battery vent
holes by delaying the permeation of hydrogen, oxygen and traces of water vapor
during charge-discharge processes. Effects of polymer concentration and casting
shear rate were investigated in order to produce the most suitable membrane. The
prepared membranes were tested on lead acid battery during charging process to
evaluate their performance in retaining electrolyte as well as minimizing pressure
build up. Results showed that decreasing polymer concentration in casting solution
decreases compactness of the polymer matrixes in the membrane structure, therefore
decreases pressure build-up in battery casing during charging process. Increasing
shear rate resulted in a decrease of skin layer thickness and in turn, lowers the
pressure build-up during battery charging process. Membrane with 13 wt/wt %
Polysulfone (PSF) and 87 wt/wt % of N-N-dimethylformamide (DMF) and at shear
rate of about 233.33 s-1 was found to be the most suitable membrane to be applied on
the maintenance free battery. At room temperature, the electrolyte losses of a
membrane-assisted lead acid battery is about 6.67 grams per hour, while for
conventional battery is about 26.67 grams per hour. During charging process at
temperature about 80oC, membrane assisted lead acid battery can save up to 40% of
electrolyte losses compared to the conventional battery. Membrane characterization
using Scanning Electron Microscopy (SEM) and Raman Spectroscopy were also
carried out to investigate the structure of the membrane and to differentiate the
components of the permeated electrolytes from both membrane-assisted and
conventional batteries. Based on SEM images, it can be observed that the membrane
cast at the shear rate of about 233.33s-1 has the most suitable skin layer thickness.
The membrane is permeable enough to minimize the pressure build up inside the
battery and sufficiently selective enough to minimize the electrolyte losses. In the
Raman spectrometer analysis, it was proven that with the application of membrane,
the intensities of hydrogen and oxygen gases in the permeated vaporized electrolyte
had been suppressed to about 50-60% and 10-30% respectively. Therefore this
research is considered as a revolutionary step in the establishing a new age of
maintenance free battery with the application of membrane technology into lead acid
battery.
vi
ABSTRAK
TABLE OF CONTENTS
CHAPTER PAGE
TITLE PAGE i
DECLARATION ii
DEDICATION iii
ACKNOWLEDGEMENT iv
ABSTRACT v
ABSTRAK vi
TABLE OF CONTENTS vii
LIST OF FIGURES xii
LIST OF TABLES xv
NOMENCLATURE xvi
LIST OF APPENDICCES xvii
1 INTRODUCTION
2 LITERATURE REVIEW
3 RESEARCH METHODOLOGY
5.1 Conclusions 90
5.2 Recommendations for Future Work 92
REFERENCES 94
xi
PUBLICATIONS 102
APPENDICES 103
Appendix A-B
xii
LIST OF FIGURES
LIST OF TABLES
NOMENCLATURE
LIST OF APPENDICES
APPENDIX PAGE
INTRODUCTION
Practical lead acid batteries began with the research and inventions of
Raymond Gaston Plante in 1860. In Plante’s fabrication method, two long strips of
lead foil and intermediate layers of coarse cloth were spirally wound and immersed
in a solution of about 10% sulfuric acid. The early Plante cell had little capacity,
since the amount of stored energy depended on the corrosion of one lead foil to lead
dioxide to form the positive active material, and similarly the negative electrode was
formed by roughening of the foil (on cycling) to form an extended surface. But since
then, there had been so much improvement on the traditional Plante cell, on the
aspect of it’s design, capacity, and even practicality, in order to suit so many
applications related to the portable power source.
A lead acid battery is a device, which converts the chemical energy contained
in its active materials directly into electric by means of an electrochemical oxidation-
reduction (redox) reaction (Linden, 1995). This type of reaction involves the transfer
of ions from one material to another through an electric circuit. A cell, the basic
electrochemical unit, is a circuit where ions are transferred during charging and
discharging state. Each cell must consists of three major components such as:
2
Physically the anode and cathode are electronically isolated in the cell to
prevent internal short-circuiting, but are surrounded by the electrolyte. In practical
cell designs a separator material is used to separate the anode and the cathode
electrodes mechanically. The separator, however, is permeable to the electrolyte in
order to maintain the desired ionic conductivity. In some cases the electrolyte is
immobilized for a non-spill design. Electrically conducting grid structures or
materials may also be added to the electrodes to reduce internal resistance.
3
Pb → Pb2+ + 2e (1.1)
At positive electrode: cathodic reaction (reduction, gain of electrons)
Electron Flow
LOAD
_ +
C
A A
N T
O H
Flow of cations
D O
E D
E
Flow of anions
Pb2+ + 2e → Pb (1.4)
_ Electron flow
+
DC Power
_ C +
A A
T N
H Flow of anions O
O D
D E
E
Flow of cations
Nevertheless there is another side reaction that occurs during charging of lead
acid battery instead of the primary reduction-oxidation (redox) reaction as mentioned
above. During the final stage of charging, whenever the cell nearly approaches full
charge, or when most of the lead sulfate (PbSO4) has been converted into lead (Pb)
and lead oxide (PbO2 ), the cell voltage will become greater than the gassing voltage.
At this point and onwards, overcharge reactions occurs, and as a result, hydrogen and
oxygen is produced at anode and cathode. The reaction, known as the gassing
reaction is shown as below:
At negative electrode
2H+ + 2e → H2 (1.7)
At positive electrode
1
H2O → O2 + 2H+ + 2e (1.8)
2
Overall reaction
1
H2O → H2 + O2 (1.9)
2
Table 1.1: Types, characteristics and applications of lead acid battery (Linden, 1995)
advances are the use of the lighter weight plastic containers, the improvement of
shelf life, the dry charge process and the maintenance free design (Linden, 1995).
The design of lead acid batteries is varied in order to maximize the desired
type of performance. Tradeoffs exist for optimization among such parameters as
power density, energy density, cycle life, “float-service life” and cost. A starting,
lighting and ignition (SLI) battery whose main function is to start an internal
combustion engine discharges briefly but at a high current. Once the engine is
running, a generator or alternator system recharges the battery and then maintains it
on float at full charge or slight overcharge. Other characteristics for starting, lighting
and ignition (SLI) are tabulated in Table 1.2.
7
Table 1.2: Typical characteristics of a starting, lighting and ignition (SLI) battery
Characteristics and unit SLI Figures
Chemistry
Anode Pb
Cathode PbO2
Electrolyte H2SO4 (aqueous solution)
The Malaysian motor trade has started during mid 1960's where all motor
vehicles found on the roads were imported. Since then, the Malaysian motor trade
has not changed a lot where most of vehicles were imported from Europe and due to
the colonial background, the majority of the motor vehicles were British origin. In
the late 1960's, Japanese cars were slowly introduced to the Malaysian roads and
began gaining popularity because of their low price and fuel consumption. At the
same year also, the government of Malaysia approved the setting up of car assembly
plants and within ten years almost every European, Japanese and Australian make
was assembled in Malaysia by eight industrial groups producing more than fifty
variants.
As more and more companies are manufacturing their products here, the
competition of becoming the best seller in the country has begun to be stiffer. Each
company is working hard in producing new items that are not only relatively low in
price, but offers better performance than other normal products. Normally the
product that they are developing is the maintenance free battery, which needs no
topping up of deionised water whenever the electrolyte level goes low.
As the cell approaches full charge and the majority of the PbSO4 has been
converted to Pb and PbO2, the cell voltage on charge becomes greater than the
gassing voltage and the overcharge reactions begin, resulting in the production of
hydrogen and oxygen (gassing) and the resultant loss of water (Linden, 1995).
It is understood that during charging, high over voltage slows down reactions
at the electrodes substantially, as some single steps of the reaction are impeded
effectively. Thus the system is in an unstable condition and any substance that
facilitate these partial
11
Table 1.3: Losses of water in the lead acid battery (Kiehne, 1989)
Factors affecting water losses in Estimated quantity of water losses per day
lead acid battery (eg. Traction battery, one cycle per day,
charging coefficient 1.10)
Electrolysis
Due to overcharging 3.4 g
Due to self-discharge 0.06 g
Diffusion of water by gassing 0.07 g
Evaporation 0.01 g
Water in acid vapor < 2 x 10-4 g
Diffusion of water by the gas by < 0.1 g
circulation pumps
Corrosion of the positive grid < 0.2 g
reaction steps can in some cases accelerate the gassing process and therefore greatly
influence the water decomposition process (Kiehne, 1989).
It is a well-known fact that the climate for Malaysia is hot and humid
throughout the year. As the sequence, battery electrolyte will evaporate easily as the
battery casing absorbs heat whenever the car is parked outdoor in the scorching sun.
In addition to that, in certain car bonnets such as Proton cars, place their battery too
close to the engine, and this will add up to the heat build up due to the hot climate.
These two factors are considered to be the main reasons why batteries in tropical
countries need to be topped
with deionised water frequently. On average, the temperature measured inside the
battery is between 60 to 80oC, and at this point, evaporation occurs.
The lead acid battery especially in tropical countries has a critical water
decomposition problem, which seems to be deteriorating its life cycle and
performance. The combination factor of overcharging and hot weather contributed to
the water losses in lead acid battery. This is causing the electrolyte level to decrease.
12
Whenever the electrolyte level goes down, the acid concentration becomes stronger
and this leads to a corrosion problem where the strong acidity will attack the
connector between grid and terminal and to a certain extent, the connector will break.
Another subsequent to the water decomposition problem is the consumer has to top
up the electrolyte level every time it goes lower than the minimum level. These are
some of the problems faced in the flooded starting, lighting and ignition (SLI)
batteries that seems need some review in its design to improve the performance.
LITERATURE REVIEW
For a solid membrane, the two bulk phases can be a combination of miscible
or immiscible liquid phases and gaseous phases. For example, gas 1/ solid
membrane/ gas 2 (gas separation), liquid 1/solid membrane/gas (vapor) 2
(pervaporation), and liquid 1/solid membrane/miscible liquid2 (dialysis,
electrodialysis, reverse osmosis, ultrafiltration and microfiltration), gas 1/solid
membrane/liquid 2 and etc (Winton Ho and Sirkar, 1992)
For a liquid membrane, the two bulk phases can be any combination of
gaseous phases, immiscible liquid phases and a solid phase. The bulk liquid phases
must also be immiscible with the liquid membrane phase. For example, gas 1/liquid
membrane/gas 2, gas 1/liquid membrane/liquid 2, liquid 1/liquid membrane/gas 2
and liquid1/liquid membrane/liquid 2.
For membrane separation process, the bulk phases are mixtures. A membrane
process permits selective separation and controlled transfer of mass or energy
between two bulk phases interposed by the membrane. The membrane is selective to
one of the species in the mixture due to the differences in mobility or solubility of
species in the membrane. Therefore, one of the species in the mixture is allowed to
be exchanged in preference to others causing one bulk phase, either the permeating
stream or the retained phase to be enriched in that species while the other is depleted
of it.
The movement of any species across the membrane is caused by one or more
driving forces which arise from a gradient of chemical potential, electrical potential,
or even the temperature gradient. A gradient in chemical potential may be due to a
concentration gradient or pressure gradient or both.
16
Sir Thomas Graham, about three and a half decades after the founding of
J.K.Mitchell’s work, was said to have fathered gas separations via membranes,
discussed the mechanism of permeation in strikingly modern terms and demonstrated
experimentally that mixtures of gases can be separated by means of rubber
membranes (Lacey, 1972). Subsequently, further evidence of the separation
possibilities is contained in a statement made by Graham in his pioneer paper “On
the Molecular Mobility of Gases” in 1863 (Hwang and Kammermeyer, 1975). In this
treatise, which resulted in the formulation of Graham’s Law, the author mentioned
“the partial separation of mixed gas”. Only a few years later, 1866, Graham
published another paper dealing with the use of “colloid septa” for the separation of
17
gases (Hwang and Kammermeyer, 1975). Through the remainder of the 19th and the
early 20th centuries, the ability of gases to permeate membranes selectively had no
industrial or commercial use and value. The concept of the perfectly selective
membrane was, however used as a theoretical tool to develop physical and chemical
theories, such as Maxwell’s kinetic theory of gases.
From 1943 to 1945, Graham’s law of diffusion was exploited for the first
time, to separate U235F6 from U238F6 as part of the Manhattan Project. Finely
microporous metal membranes were used. The separation plant, constructed in
Knoxville, Tennessee, represented the first large scale of gas separation membranes
and remained the world’s largest membrane separation plant for the next 40 years.
However, this application was unique and so secret that it had essentially no impact
on long-term development of gas separation.
In the 1940’s to 1950’s, Barrer, van Amerongen, Stern, Meares and others
laid the foundation of the modern theories of gas permeation. The solution-diffusion
model of gas permeation developed then is still the accepted model for the gas
transport through membranes. However, despite the availability of interesting
polymer materials, membrane fabrication technology was not advanced enough at
that time to make useful gas separation membrane systems from these polymers.
The development of high flux anisotropic membranes and large surface area
membrane modules for reverse osmosis applications in the late 1960’s and early
1970’s provided the basis for modern membrane gas separation technology. The first
company to establish a commercial presence was Monsanto, which launched its
hydrogen-separating Prism membrane in 1980. The company had the advantage of
being a large chemical company with ample opportunities to test pilot and
demonstration-scale systems in its own plants before launching the product. The
economics were compelling, especially for the separation of hydrogen from
ammonia-plant purge-gas streams. Within a few years Prism systems were installed
in many such plants (Baker, 2000).
18
In the 1940’s till the 1950’s, van Amerongen has been said to be the man who
initiated the present knowledge concerning the mechanism of gas permeation
through polymeric membranes. This is followed by a research by Rogers and also in
a monograph edited by Crank and Park. A typical schematic membrane process for
gas separation is shown in Figure 2.1 (Spillman and Sherwin, 1990).
Low Pressure
Permeate
Fast Permeating Gas
Slow Permeating Gas
Convective Flow
• Knud
en Diffusion
Molecular Sieving
(a)
Solution Diffusion
Mechanism
(b)
Figure 2.2 Mechanism for gas separation using (a) porous membrane and
diffusion). These very small pore membranes have not been used on a large scale,
but ceramic and ultramicroporous glass membranes with extraordinarily high
selectivities for similar molecules have been prepared in the laboratory.
21
where is Ji the flux of component i [g/(cm2.s)], pio and pil are the partial pressure of
component i on the either side of the membrane, l is the membrane thickness, Di is
the permeate diffusion coefficient, and KiG is the Henry’s Law sorption coefficient
[g/(cm3.pressure)]. In gas permeation it is much easier to measure the volume flux
through the membrane than the mass flux, so Equation 2.1 is usually recast as:
where is ji the volume (molar) flux expressed as [cm3 (STP) of component i/(cm2.s)],
pio and pil are the partial pressure of component i on the either side of the membrane,
l is the membrane thickness, Di is the permeate diffusion coefficient, and Ki is the
sorption coefficient with the unit [cm3 (STP) of component i /(cm3 of
polymer).(pressure)]. The product DiKi can be written as Pi which is called the
membrane permeability and is a measure of the ability of the membrane to permeate
gas. The ability of a membrane to separate two gases i and j is measured by the ratio
of their permeabilities called the membrane selectivity:
22
αij = Pi (2.3)
Pj
αij = Di Ki (2.4)
Dj Kj
The ratio Di / Dj is the ratio of the diffusion coefficients of the two gases and
can be viewed as the mobility selectivity, reflecting the size differences of the two
molecules. The ratio Ki / Kj is the ratio of the sorption coefficients of the two gases
and can be viewed as the sorption or solubility selectivity, reflecting the relative
condensabilities of the two gases.
dc
J = −D (2.5)
dx
23
c = Sp (2.6)
Both equations (equation (2.5) and equation (2.6)) are then combined and
integrated to give:
∆pΑ
Q = DS (2.7)
l
where Q is volumetric flow rate of penetrate gas permeate through membrane, A and
l is membrane effective area and thickness, while ∆p is pressure difference across
membrane (Stern, 1972).
solubility coefficients that vary with permeate and membrane material. A common
P = DS (2.8)
0.1µm~1µm
100µm~200µm
Skin Layer
Sublayer
Furthermore, the skin layer density is somewhat greater than that of the
substructure, and whose depth (everything else being equal) does not vary
appreciably with total membrane thickness. Because of its greater density, the skin
layer offers greater resistance to material transport that that provided by a
substructure layer of equivalent thickness. Indeed, where the difference in density
between the two layers is great, the gas transport resistance of the entire membrane is
approximately equal to that of the dense skin layer alone.
the cast polymer film into the non-solvent coagulation bath. The membrane formed
though polymer precipitation as a resultant of the solvent loss and an increase in
polymer concentration in the top layer of the polymer solution.
The big difference lies between the type 1(wet cast) and types 2, 3 and 4 (dry
cast), where the casting process used are not similar. The distinction refers to
whether the outlet of the casting knife that drags the casting solution from a closed
reservoir is exposed to the air or directly submerged to a non-solvent coagulation
medium. This is considered important because during the time between extrusion of
the nascent membrane and its immersion, evaporation of the solvent could occur.
Subsequently producing a more concentrated polymer region near the air-solution
interface to promote the formation of a skin layer. Nevertheless, dry cast / dry-wet
phase inversion process will be given the focus, as it is the process produces most of
the membranes today.
Most of the membranes in use today are phase inversion membranes obtained
by dry/ wet phase inversion technique. Phase inversion membranes can be prepared
from variety of polymers. The only requirement is that the polymer must be soluble
in a solvent or a solvent mixture. In the formation process of a membrane, two types
29
of phase inversion can be distinguished. The dry phase inversion takes place in the
atmosphere by evaporation of the volatile solvent. The wet phase inversion is carried
out by immersing the polymer solution membrane into a coagulation bath of a non-
solvent takes place and the membrane formed. The dry/wet phase inversion
technique is a relatively new process which arose out of the quest to fabricate ultra-
thin and defect-free on asymmetric gas separation membranes. The dry/wet phase
inversion technique requires multi-component casting solutions consisting of
polymer, at least one volatile solvent and one less volatile non-solvent then bringing
the solution to phase separation by means of solvent outflow and/or non-solvent
inflow (Wijmans et al., 1983). However, it has been found that the use of two
solvents, a primary volatile solvent and a secondary less volatile solvent, allows finer
control of solvent evaporation and polymer coagulation rates. All solvents and non-
solvents should be miscible with the coagulation bath medium and the non-solvent is
added to the casting solution until the solution nears its phase boundary (Sourirajan
et al., 1993). The evaporation process for a typical ternary casting system used for
the preparation of membranes made by dry/wet phase inversion is illustrated
schematically in Figure 2.4.
Solvent
Exchange and Vitrification Wet Phase Free Standing
Drying Separation Evaporation
Figure 2.4 Steps involved in the formation of asymmetric flat sheet membranes
using dry/wet phase separation process (Clausi and Koros, 2000).
30
The process, from the most basic standpoint, consists of the formation of a
nascent membrane, followed by interfacial phase separation induced through the
evaporation of a volatile solvent. After a short ‘healing’ period not less than 3s and
more than 10s, the nascent membrane is then immersed in a non-solvent bath where
phase separation occurs throughout the rest of the membrane (Clausi and Koros,
2000).
τ s = ωγ& θ (2.1)
where τs is shear stress and γ& is shear rate. ω and θ are constants for a particular
fluid. ω is a measure of fluid consistency; a high value of ω indicates a more
viscous fluid. θ is a measure of non-Newtonian behavior; θ < 1 indicates a shear-
thinning behavior (Ismail, 1997).
30% QR 30% QR
all the water that is recombined is also condensed and flows back into the battery, but
20% escapes as water vapor through openings that the converter has to have because
oxygen and hydrogen are not always produced at the same rate.
Figure 2.6 Valve regulated lead acid battery with immobilized electrolyte (Dietz and
Dittmar, 1992)
lead. This sequence of reactions is demonstrated in the first three equations on the
left side of figure 2.6. The addition of the three equations gives the equation below.
The kind of reaction delivering lead oxide has the same result as the direct reaction.
Functional preconditions for the oxygen cycle are:
The second means the cell has to have no liquid electrolyte, because a low
diffusion speed of oxygen in the sulfuric acid reduces the oxygen cycle on very little
current. Thus, the main characteristics of enclosed batteries is the immobilization of
the electrolyte while the sulfuric acid is soaked in a glass mat or fixed as a gel. In
both cases open pores allow easy oxygen transport to the negative electrode as
mentioned in Figure 2.6.
37
However, there is a major problem of sealed lead acid batteries and that is the
control of the gas pressure. Whereas the efficiency of the oxygen cycle in sealed lead
acid batteries is close to 100%, an analogous hydrogen cycle cannot be achieved, as
the hydrogen oxidation is kinetically hindered at the working potential of the positive
electrode. Therefore, in sealed lead acid batteries: (i) the hydrogen evolution is
usually suppressed (by charging at low constant voltage, using cell components of
high purity, especially antimony-free lead alloys); (ii) evolved hydrogen can be
oxidized, e.g., electrochemically at (catalyzed) auxiliary electrodes or by currentless
catalytic hydrogen/oxygen recombination, and (iii) excess hydrogen is removed by
means of a safety-valve together with oxygen (Dietz and Dittmar, 1992)
ii) in combination with auxiliary electrodes held at special potentials (by means
of diodes connected to the battery electrodes) to ensure oxygen reduction or
hydrogen oxidation
However, there are some obstacles that have not allowed industrial
application on a large scale yet. These are the complicated voltage control measures
and the disadvantages involved in the application of noble metal catalysis (high
costs, expensive precautions against poisoning of the catalyst and explosion hazards
at high reaction yields, risk of increasing self-discharge by the decrease of the
hydrogen over potential caused by noble metal traces in the electrolyte).
Many workers have been studying on how to suppress the hydrogen evolution
and oxygen reduction in lead acid batteries. This is because the main factor that
causes gassing reaction to be occurring in lead acid batteries. During the production
of lead acid battery, additives were introduced into lead paste to make it more
durable and strong, as lead is a very soft substance. Normally, substances like
antimony and calcium were used added in lead paste to the earlier mentioned
objective, however these substances cause marked influence on gassing reaction to
occur.
using the oxygen cycle, could improve their performance in the presence of bismuth
(Bi), stanum (Sn) and selenium (Se) (Maja and Penazzi, 1988).
iii) Indirectly increasing PbSO4 reduction current, I (PbSO4 red), i.e., by lowering
the state-of-charge of the negative electrode.
However, method (i) has practical relevance since neither I (O2 red) nor the
Ampere-hour capacity are decreased. Efficient removal of H2 is also possible by
catalyzing the hydrogen/ oxygen recombination (Dietz, 1990).
The problem can be solved by the addition of electrolyte to the cell. The new
design allows the more concentrated electrolyte from the reservoir to fall down by
gravity and mix with the depleted electrolyte in the cell as discharge is proceeding,
thus maintaining the electrolyte concentration. The system works properly during
discharge, but needs an extra overcharging process to homogenize the electrolyte
reservoir. This reduces the cycle efficiency and accelerates the electrode corrosion
and plate ageing (Armenta-Deu, 1998).
In lead acid batteries, the cell performance has noticeably been improved
with the use of a circulation system that homogenizes the electrolyte concentration,
achieves longer discharge time, and increases the output voltage. Further
improvements have been obtained when the battery has been attached to an
electrolyte reservoir supplying sulfuric acid to the main cell. In this case the acid
depletion at the electrode interface is compensated by the forced flow created by the
electrolyte circulation, and hence the higher concentration gradient between the bulk
solution and the interface. The result is the enhancement of the discharge process
(Armenta-Deu, 1998).
CHAPTER 3
RESEARCH METHODOLOGY
Material Selection
losses from the lead acid battery as well as minimizing the pressure build up inside
the battery casing. The optimized membrane will be tested in the membrane-assisted
lead acid battery charging test to observe the effect of high temperature condition on
the electrolyte losses in lead acid battery during charging process. Membrane
characterization using Scanning Electron Microscopy (SEM) and Raman Spectroscopy
were also carried out to investigate the structure of the resultant membrane and to
differentiate the components of the permeated electrolytes from both membrane-assisted
and conventional batteries.
3.2.1.1 Materials
Before casting of a flat sheet membrane can be made, a casting solution has
to be prepared first. Generally, a casting solution will consist of a polymer, a solvent
and a nonsolvent. For this research, polysulfone has been selected as the base
polymer for all the casting solution made. As for the solvent, N,N-dimethylacetamide
(DMAc), dimethylformamide (DMF) and tetrahydrofuran (THF) have been selected
due to the strong interaction with polymer and miscibility with water. Meanwhile the
nonsolvent used in this experiment is an aliphatic alcohol, ethanol. It is used as a
nonsolvent additive for polysulfone.
CH 3
SO 2 O C O
CH 3 n
Apparatus for forming polymer solution had been set up as shown in Figure
3.3. Polysulfone pellets were dried in a vacuum oven at 70oC overnight before used
in making dope solution. This is to remove any absorbed water or moisture contents
on the polymer surface. The polymer and less volatile solvent, N,N-
dimethylacetamide (DMAc) were weighted according to the percentage on an
electronic scale and kept inside an aluminium foil sealed beaker. Once the dope
solution apparatus is set and the stirrer is controlled at medium speed, the less
volatile solvent is poured into it.
When the temperature of the solvent has reached 50 to 55 oC, polymer pellets
were added gradually into the flask. Afterwards, the polymer pellets were ensured to
be totally dissolved in the solvent before inserting new pellets into the vessel. This is
done to avoid agglomeration of polymer. Right after the addition of the polymer
pellets into the flask, temperature is slowly increased to 70 to 90 oC, while the stirrer
is set at a much higher speed to ensure the polymer pellets were fully dissolved.
After all of the pellets were added into the vessel and fully dissolved, the more
volatile solvent, tetrahydrofuran (THF) is added into the vessel. After almost an hour,
non-solvent, ethanol is poured slowly into the vessel.
The solution is then stirred continuously until the all bubbles inside the
solution disappear. Once the solution is ready, the stirrer, the heater, the blower and
45
the cooling water is switched off. The dope solution is then kept tightly inside a
bottle. In order to remove residual bubbles trapped inside the dope solution, the
bottle is immersed inside an ultrasonic bath for about 3-4 hours before casting.
Stirrer
Condenser Thermometer
Feed Funnel
Flange Reaction
Flask
Heating Unit
ambient atmosphere (30oC and 84% relative humidity). The flat sheet casting shear
rate can be calculated using the Equation 3.1 as follows:
v
γ& =
g
(3.1)
where v is velocity of casting knife and g is gap setting of casting knife (Ismail,
1997).
After that, the nascent membrane was quenched immediately into an aqueous
bath at room temperature (28°C) and remained there for 2 days. Finally, the washed
membrane was solvent-exchanged with methanol for 1 day and then air-dried for 1
day.
Qi = ( V / t ) (3.2)
where Qi is the flow rate of the permeating gas (cm/s), V is the volume of the gas
collected ( 10 cm3 ) and t is the time needed for gas to permeate (seconds).
Therefore, permeability can be determined by using the following equation:
48
Pi = Qi / ( A.∆p ) (3.3)
After various types of membranes were fabricated, they were then tested on
lead acid battery to find out whether they are capable of maintaining electrolyte level
inside the battery. In order to put the membrane up to the real test, the membrane is
applied on the battery cap and then the battery is charged by an alternator. This
analysis is done in three stages whereas to optimize the fabricating conditions so that
electrolyte losses can be minimized. The tests were done accordingly as below:
i) First Stage
- Preliminary membrane-assisted lead acid battery charging test
ii) Intermediate Stage
- Membrane-assisted lead acid battery charging test
iii) Final Stage
- Membrane-assisted lead acid battery charging test with heat supply
Before putting the membrane up to the real test, a preliminary test is done to
see which membrane formulation that suits to be applied on the lead acid battery.
This is done to screen down the suitable membrane formulation to be used in the
membrane assisted lead acid battery-charging test. Through this preliminary test, all
the membranes fabricated in the earlier stage were cut accordingly to the battery vent
hole size and then applied on each of the six lead acid battery vent holes with epoxy
adhesive to prevent any leakages. Then the battery is continuously charged using an
alternator and charged on average 12 to 15 Amps per hour. The objective of this test
is to select a membrane formulation with the most suitable polymer concentration.
This is done by observing the electrolyte level inside the battery. The membrane that
has the most retained electrolyte level throughout the experiment will be the chosen
one. A control test, run without any application of membrane on the battery holder,
was done to differentiate how much electrolyte losses can be minimized. Prior to this
screening process, the selected membrane formulation will proceed to the second
stage experiment, which is the membrane applied lead acid battery-charging test.
50
After the first stage of experiment has been established, the most suitable
membrane formulation will be chosen and reproduced with different casting shear
rates. The objective of this experiment is to determine the best shear rate so that the
electrolyte losses can be further minimized. Therefore during the membrane-assisted
lead acid battery-charging test with heat supply, the performance is already
optimized.
As the membranes were produced with different shear rates, they were cut
into circular sheets with the diameter of 3.5 cm. The membrane will be applied into a
battery cap as shown in Figure 3.7. Subsequently, this spiral end stainless steel
battery cap is applied on battery vent hole, and then sealed with epoxy sealant to
prevent any leakages. Charged using an alternator at the average rate of 15 Amps per
hour, the battery is charged for approximately three straight hours. Gas permeation
from the battery cap is measured as such shown in Figure 3. 8, where it uses the
principle of soap bubble flow meter. During this experiment, three parameters were
subsequently observed, which are the flow rate of the permeating vaporized
electrolyte; the electrolyte losses and the pressure build up measured inside battery.
At the end of the experiment, the membrane with the best casting shear rate will be
selected and will proceed to the final stage of the experiment. The best casting shear
will be expected to exhibit the best performance in minimizing electrolyte losses as
well as minimizing pressure build up.
51
3.4.3 Membrane-assisted Lead Acid Battery Charging Test with Heat Supply
This experiment will be the third and the final stage experiment in evaluating
the effectiveness of membrane application on lead acid battery. It is quite similar to
the membrane-assisted lead acid battery charging test, where all of the equipment
configurations are all the same except that there is a water bath inclusion for this
experiment. The objective of this test is to investigate the effect of heat supply to the
battery on the electrolyte losses. Membrane with the most suitable polymer
concentration and casting shear rate which is selected from two previous
experiments, will be tested in this experiment to observe the effectiveness of the
membrane in retaining electrolyte within the existence of high temperature condition.
Similar to the membrane applied lead acid battery charging test, membranes
are cut into circular sheets with the diameter of 3.5 cm and applied into battery caps
as shown in Figure 3.7. Subsequently, this spiral end stainless steel battery cap is
applied on battery vent hole, and then sealed with epoxy sealant to prevent any
leakages. Charged using an alternator at the average rate of 15 Amps per hour, the
battery is charged for approximately three straight hours. This time, instead of
placing the lead acid battery on a weight scale, it is put inside a temperature
controlled water bath, as shown in Figure 3.9. To measure the water losses from the
lead acid battery, the initial mass of the battery is measured before starting the
experiment. After the test is done the mass of the battery is measured again to see the
electrolyte losses.
53
Figure 3.8 Membrane-assisted lead acid battery charging test with heat supply
Raman
Spectroscopy
Computer
Battery
Alternator
In the gas permeation test, the two main gases, oxygen and hydrogen, that
were formed in the battery during charging were tested using gas permeation test rig.
Every membrane sample from different types of formulations was tested to observe
the effect of different casting formulation on the performance of lead acid battery.
There are eight different types of membrane with different formulations that were
successfully cast. The solution formulations are tabulated in Table 4.1.
Each and every type of the cast membranes was tested for its permeation of
pure hydrogen and oxygen gases. Data for low polymer concentration membranes
such as GS 6, GS 7 and GS 8 cannot be obtained due to the rapidness of gas
permeating from these membranes. Therefore, permeation rates of these membranes
cannot be calculated.
58
Polymer Solution GS 1 GS 2 GS 3 GS 4 GS 5 GS 6 GS 7 GS 8
Composition
Figure 4.1 depicted the membrane produced from ternary system such as in
GS 1, GS 2 and GS 3. It was observed that the permeation rate for GS 3 is higher
than that of GS 1. This phenomenon occurs because of higher polymer concentration
in the casting solution produces thicker skin layer, thus resulted in lower permeation
rate. The scenario also happens in binary system casting solutions. It seems that the
polymer concentration for GS 5 is lower than of GS 4, therefore permeation rate for
GS 5 is higher than GS 4. The only difference between binary and ternary system is
the structure of the resultant membrane. Binary system tends to form membranes
having more uniform structure rather than that of ternary system as proven in the
SEM images.
59
H2 Test O2 Test
18
16
14
Gas Permeation Rate,
cm3/ cm2.s. cmHg
12
10
8
6
4
2
0
GS 1 GS 2 GS 3 GS 4 GS 5
The solutions were cast at the shear rate of 210 s-1 prior to coagulation bath.
Shear rate was calculated from equation (3.3) with the gap of casting knife of 150 µm
and casting velocity at 3.15 cm/s.
In this study, the performance of membranes cast from ternary and binary
systems were tested in retaining electrolyte from being disposed to the atmosphere.
Since no previous work has been done on this area, the experiment had to be done
based on direct observation on the lead acid battery during charging process.
minimal compared to GS1, GS2 and GS3. This is due to the thickness of the skin
layer possessed by the ternary system (one polymer, one solvent and one non-
solvent) solution cast membranes. This result agreed with the conclusion made by
previous works (Chung et al., 1997; Hachisuka et al., 1996).
vaporized electrolyte and at the same time shows minimum pressure build-up inside
the battery casing. It can be justified that GS 6 membrane, has passed the preliminary
test and practically can be used as an electrolyte-retaining device.
300 7s
262.5 8s
233.33 9s
210 10s
190.91 11s
175 12s
As shown in Figure 4.2, the chart shows the profile of electrolyte losses of
three selected membrane shear rates and one control test. The other three profiles can
be seen in Appendix A2. The electrolyte loss for the control test was continuously
increasing until the end of the test. At the end of the three-hour experiment, the
control test battery lost about 80 grams of electrolyte, and has an average electrolyte
losses rate of 26.67 grams per hour. Meanwhile the overall electrolyte losses were
tabulated in Table 4.4. As it can be seen here, for batteries that were applied with
membranes, the electrolyte losses were found to be the coherent in shear rates of 7s,
11s and 12s, with the rate of losses of 20 grams per hour. Whereas for the shear rates
of 8s, 9s and 10s, the average rate of electrolyte losses was minimally around 6.67
grams per hour.
Figure 4.3 shows that 7s, 11s and 12s membranes were able to decrease the
electrolyte losses by 25%. However, 8s, 9s and 10s membranes showed 75%
reduction in electrolyte losses. This indicates that the critical shear lies within the
range of 210s-1 to 262.5s-1.
In order obtain out the critical shear rate, a comparison analysis must be made
on other parameters, which are the pressure build-up inside the battery casing and the
flow rate of the vaporized electrolyte profile. The critical shear rate will feature a
minimal pressure build-up inside the battery casing and smoothest flow rate profile
of vaporized electrolyte.
Table 4.4: Total and average rate of electrolyte losses during membrane-assisted
lead acid battery charging test
13.85
13.84
13.83
13.82
13.81
Electrolyte Losses (kg)
13.8
Control Test
13.79
7s
9s
13.78
12s
13.77
13.76
13.75
13.74
13.73
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180
13.72
Figure 4.2 Effect of membrane casting shear rate on the performance of membrane in minimizing electrolyte losses
67
Figure 4.3 Minimization of electrolyte losses with comparison to the conventional lead
acid battery
Membranes cast at six different shear rates show different values of maximum
pressure build-up with similar trend of profile. Each of the membranes had a sudden
increase of pressure build-up during the first twenty minutes of charging. The next
phenomenon seen after the sudden increase is the pressure build-up starts to show
tendency in becoming constant as the flow rate of vaporized electrolyte becomes faster.
68
During the first twenty minutes of charging, the vaporized electrolyte could not
be permeated smoothly since the pressure build-up is still low. Since membrane flux is
pressure driven, the vaporized electrolyte could not permeate through the selective skin
and therefore is trapped inside the battery casing. The pressure build-up has increased
until it reaches the point where the pressure becomes constant and the vaporized
electrolyte permeation rate becomes constant too. This is the point when the condition
reaches its “working pressure”, where the pressure build-up is stabilized with the smooth
and constant permeation of vaporized electrolyte.
In selecting the critical shear rate within the six different shear rates applied, the
pressure profile must feature a minimum value of “working pressure” as well as
minimizing electrolyte losses during the charging process. In this case, the 9s membrane
(cast at 233 s-1) had the lowest pressure build-up profile with the final build-up pressure
of only 0.73 bar after three hours of charging. Compared to the other shear rates, 7s
membrane recorded 0.93 bar; 8s membrane - 1.02 bar; 10s membrane - 0.88 bar; 11s
membrane – 0.96 bar and 12s membrane – 0.93 bar, 9s membrane relatively has the
lowest final pressure.
69
1.1
0.9
0.8
0.7
Build-up Pressure (bar)
0.6 7s
8s
0.5 9s
10s
0.4 11s
12s
0.3
0.2
0.1
0
0 10 20 30 40 60 80 100 120 140 160 180
Charging Time (minute)
Figure 4.4 Effect of shear ate on the pressure build-up inside battery casing
70
As shown in Figure 4.5, the flow rate of gas permeation from the battery during
charging in the control test is continuously increasing. This shows that, electrolyte is
continuously decomposed without any application of membrane. Most of the membranes
have shown a slow increase of vaporized electrolyte flow rate except for 9s membrane.
The rate of increment for vaporized electrolyte flow rate in 7s, 8s, 10s 11s and 12s
membranes were not that distinct from what had been shown in the 9s membrane flow
rate profile. The flow rate of gas permeation from the battery during charging with the
application of 9s membrane is quite slow at the early stages of charging and the rate is
increasing as time elapse. During the first sixty minutes of charging, 9s membrane have
shown slower flow rate of vaporized electrolyte due to its electrolyte retaining
characteristics. However after the first sixty minutes, 9s membrane’s profile almost
emulates the flow rate of the control test, and this proves that the vaporized electrolyte is
being well circulated inside the battery without tolerating the minimization of electrolyte
losses during charging. This indicated that as charging proceed, the permeation of gas
will continuously increase and pressure build-up inside the battery casing could be
minimized.
In the region prior to the critical shear rate (233.33 s-1), formation of dense skin
occur due to demixing and precipitation mechanism, which was considered to be
independent of shear rate. Increasing shear rate enhance molecular orientation in skin
layer and in turn, improving selectivity in asymmetric membrane (Ismail et al., 1997;
Shilton, 1997).
71
2.25
1.75
1.5
Ctrl Test
Vaporized Electrolyte
Flowrate (cm3/s)
7s
1.25 8s
9s
1 10s
11s
12s
0.75
0.5
0.25
0
10 20 30 40 60 80 100 120 140 160 180
Charging Time(minutes)
Figure 4.5 Effect of shear rates on the flow rate profile of vaporized electrolyte
72
This has been proven where electrolyte losses were seemed to be further
minimized. Meanwhile the permeation rate of asymmetric membrane was also found to
increase when the shear rate is approaching to the critical shear, which was consistent
with the reduction in skin thickness, where this is evindenced by pressure build-up
which seemed to be reduced (Shilton et al., 1994).
Since the solution used in this study is a shear thinning, when shear rate reached
beyond the critical point (233.33 s-1), a severe decrease in solution viscosity occurred
presumably due to chain entanglement losses in solution. In this case, membrane might
undergo an instant demixing and precipitation to result in a highly oriented skin layer.
(Sharpe et al., 1999; Chung et al., 2000). Furthermore, casting over the critical shear rate
(233.33 s-1) which is considered high shear rate, could pull molecular chains or phase
separated domains apart and began to create slight defects or imperfections in skin layer
(Sharpe et al., 1999). As shown in the Figure 4.4, casting polysulfone membrane at shear
rates of 262.15 s-1 to 300 s-1 causes a decrease in selectivity where electrolyte losses
seemed to be higher than those feature near the critical shear rate region.
73
4.1.4 Membrane-assisted Lead Acid Battery Charging Test with Heat Supply
Analysis
As depicted in Figure 4.6, the chart shows the total electrolyte losses for both
batteries, membrane applied and conventional ones, during the membrane-assisted lead
acid battery charging test with the temperature set at 60 oC and 80oC respectively. This
test, eventually simulates the condition within the engine compartment, where the
battery is charge at the average rate of 15 Amps per hour and with the effect of high
temperature of 60 oC and 80oC. The objective is to observe how the membrane performs
in minimizing electrolyte losses, even with the existence of extreme temperature effect.
For the control test results, the electrolyte losses for the battery that was charged
for three straight hours in a water bath set at 60 oC was 60 grams . As calculated, the
average electrolyte losses rate for this conventional battery is 20 grams per hour.
Meanwhile the electrolyte losses for the GS6-9s membrane applied battery in the same
condition was only 40 grams , where the average electrolyte losses rate is 13.33 grams
per hour. As resulted in the control test, the electrolyte losses for the battery charged for
three straight hours in a water bath set at 80 oC was 100 grams, while the losses for the
GS6-9s membrane applied lead acid battery in the same condition was only 60 grams.
On average, the rate of electrolyte losses for a conventional battery when the
temperature is 80 oC is 33.33 grams per hour while for the membrane applied battery is
only 20 grams per hour.
As shown in Figure 4.7 and by calculating using equation (4.1) during lead acid
battery charging process in the engine compartment, when the temperature reaches 60
o
C, the membrane-assisted lead acid battery can save up to 33.33% of electrolyte
compared to the conventional battery. Meanwhile, when the temperature reaches 80 oC,
74
0.12
0.1
Electrolyte Losses (kg)
0.08
0.06
0.04
0.02
0
Ctrl Test; With GS6-9s; Ctrl Test; With GS6-9s;
T=60 oC T=60 oC T=80 oC T=80 oC
Figure 4.6 Total electrolyte losses during membrane-assisted lead acid battery charging
test with heat supply
the membrane-assisted lead acid battery can save up to 40% of electrolyte during the
charging process.
Comparing to the results from the previous charging test, the average electrolyte
losses rate for the membrane applied lead acid battery at room temperature is 6.67 grams
per hour, which is about 50% less than when the temperature is set at 60 oC, and 66.67%
less than when the temperature is set at 80 oC. This means that high temperature
conditions has an enormous effect on the electrolyte losses for a membrane-assisted lead
acid battery during charging. Although with the membrane application on lead acid
battery can minimize up to 40% of electrolyte losses in conventional battery during the
temperature is 80oC, the losses can still be further minimized with preventive measures.
This can be done with better heat insulation of the battery casing or altering the battery
placement in engine compartment; which prevents heat being transferred into the
battery.
75
0.45
0.4
0.35
0.3
Percentage (%)
0.25
0.2
0.15
0.1
0.05
0
Water bath Water bath
set at 60oC set at 80oC
Figure 4.7 Minimization of electrolyte losses with compared to the conventional lead
acid battery
The pressure profile of the membrane applied lead acid battery charged in a
water bath set at 60 oC and 80 oC showed almost similar increment of pressure for a
battery that is charged without any heat supply. The only difference existing is the final
pressure is slightly higher than the previous test. When the temperature is set at 60 oC the
final pressure is about 0.77 bar, while the final pressure when the temperature set at 80
o
C is about 0.78 bar. Compared to the battery charged without any heat supply, the final
pressure was about 0.73 bar. This means that within the effect of the high
76
0.8
0.6
Build-up Pressure (bar)
With GS6-
0.4
9s; T=60oC
With GS6-
9s; T=80oC
0.2
0
0 10 20 30 40 60 80 100 120 140 160 180
Charging Time (minute)
Figure 4.8 Effect of water bath temperature on pressure build-up in a lead acid battery during charging
77
2.5
2.3
2.1
Vaporized Electrolyte Flowrate (m3/s)
1.9
1.7
Control Test;
1.5
T=60oC
With GS6-9s;
1.3 T=60oC
Control Test;
1.1 T=80oC
With GS6-9s;
0.9 T=80oC
0.7
0.5
10 20 30 40 60 80 100 120 140 160 180
Charging Time (minute)
Figure 4.9 Effect of water bath temperature on the flow rate profile of vaporized electrolyte
78
temperature during charging, the pressure build-up is increased by only minimal margin.
Therefore, in high temperature conditions, the pressure build-up will not differ largely
from the battery that is being charged at room temperature. This is due to the fact that
the flow rate of vaporized electrolyte have been consistently fast even on the early stages
of charging process. With the smoothly permeating vaporized electrolyte, this condition
has assisted the system to reduce the pressure build-up inside the battery.
For the vaporized electrolyte flow rate profile, the vaporized electrolyte flow rate
for both control tests run at 60 oC and 80 oC were permeating very fast. The fact that the
effect of high temperature on the formation of more vaporized electrolyte had increased
the flow rate of the permeating vaporized electrolyte. But compared to the membrane
applied lead acid battery, the flow rate of vaporized electrolyte during the early stages of
charging was quite low due to electrolyte retaining characteristics possessed in the
membrane. As the charging elapses, the pressure had increased and the vaporized
electrolyte flow rate became faster as to minimize the pressure build-up inside the
battery.
concentration of molecules (Henn, 1996). Therefore, in this analysis, the results can
describe the concentration of each hydrogen and oxygen gases through the Raman
integrated intensity data. This indeed to show the effectiveness of membrane application
in retaining hydrogen and oxygen gases from being disposed into the atmosphere.
As shown in the Figure 4.10(a), the chart shows the example Raman spectrum
for a vaporized electrolyte gas sample collected from a conventional battery, while
Figure 4.10(b) features the Raman spectrum for the membrane-assisted lead acid battery
vaporized electrolyte gas sample. It can be seen that a strong and potent peak discovered
at the range of 3800-3900 cm-1 , where this indicated the existence of hydrogen element
in the sample with quite high concentration. The chart also depicts the indication of
oxygen element in the range of 1500-1650 cm-1 but compared to hydrogen, the quantity
is smaller.
80
0.011
0.01
0.009
0.008
0.007
Intensity
0.006
0.005
0.004
0.003
0.002
4100 3900 3700 3500 3300 3100 2900 2700 2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
Raman Shift (cm-1)
Figure 4.10(a) Raman Spectrum for vaporized electrolyte (conventional battery) sample
taken at 100 minutes of charging process
0.0065
0.006
0.0055
0.005
0.0045
Intensity
0.004
0.0035
0.003
0.0025
0.002
4100 3900 3700 3500 3300 3100 2900 2700 2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
Raman Shift (cm-1)
The rest of the data of this analysis have been compiled in Appendix B and were
plotted in Figure 4.11 and Figure 4.12 to show the overall intensities for hydrogen and
oxygen gases during the three hour charging process. As depicted in the Raman
spectrums, the Raman shift (wave number) for the detected hydrogen and oxygen
intensity peaks would not only be fixed at 4155 cm-1 and 1555 cm-1 respectively, but
also can be found at the regions near to these two plots. This phenomenon occurred due
to the fact that gas samples were having high kinetic energy. Therefore, the mobility of
the molecules forces of the scattering process happened at different Raman shifts.
However, it was still in the near regions of the normallized differential Raman scattering
cross section.
0.014
0.012
0.01
Conventional
0.008
Battery
Intensity
Membrane
0.006 Applied Battery
0.004
0.002
0
20 40 60 80 100 120 140 160 180
Charging Time (minute)
Figure 4.11 Overall hydrogen intensities for vaporized electrolyte gas samples for conventional and membrane-assisted lead acid
batteries.
83
0.012
0.01
0.008
Intensity
Conventional
0.006
Battery
Membrane
0.004
Applied Battery
0.002
0
20 40 60 80 100 120 140 160 180
Charging Time (minutes)
Figure 4.12 Overall oxygen intensities for vaporized electrolyte gas samples for conventional and membrane-assisted lead acid
batteries.
84
Skin Layer
Transition layer
Substructure
(a)
Skin Layer
Substructure
(b)
Skin Layer
Substructure
Macrovoids
(c)
Figure 4.13 Scanning electron micrographs of membrane cross section with different
polymer concentration ( a: 15.0 wt/wt%; b: 13.0 wt/wt%; c: 12.5 wt/wt%)
87
This has been proven that casting asymmetric membranes from a dilute
polymer solution produced a thin and porous skin layer, leading to a high
permeability of vaporized electrolyte and low pressure build up. While increasing
polymer concentration of casting solution forms a denser and thicker skin layer,
resulting in a more pressure build up but less vaporized electrolyte permeation in
asymmetric membranes for the application on lead acid battery during charging.
However, as it can be seen here, the skin thickness of membrane was found
to be decreasing with the increase of shear rate during casting process. In addition to
this, as seen in Figure 4.14(c), a relatively thicker transition layer was found existing
in the polysulfone membrane with the casting shear rate of 300 s-1 . There was not
much of a difference in the outloook of substructures for the membranes cast at
different shear rates. However, casting over the critical shear rate (233.33 s-1) which
88
is considered high shear rate, could pull molecular chains or phase separated domains
apart and began to create slight defects or imperfections in skin layer and the
substructure itself. It can be seen in Figure 4.14 (c), that the voids in the substructure
of the membrane cast over the critical shear rate were arranged uniformly compared
to that shown in Figure 4.14(b) where most of the voids were arranged distributively.
89
Skin layer
Substructure
(a)
Skin layer
Substructure
(b)
Skin layer
Transition layer
Substructure
(c)
Figure 4.14 Scanning electron micrographs of membrane cross section at different
shear rates ( a: 210-1; b: 233.33 s-1 ; c: 300 s-1)
CHAPTER 5
5.1 Conclusions
Based from all the experimental work experienced and results obtained in this
study, some recommendations for a better and more comprehensive execution of the
future work in this particular area have been drawn:
2. Lead acid battery charging test parameters such as rate of charging and
charging period should also be carried out to analyze the performance of
the membrane under extreme conditions.
3. Lead acid battery charging tests should be done in a well designed battery
casing for the placement of the membrane in order to prevent leakages
and stabilize the pressure build up inside.
Billard, G. (1992), “ A vision for the Asian battery industry”, Journal of Power
Sources, 38, 3-11.
Bottino, A., Camera-Roda, G., Capannelli, G. and Munari, S., (1991), “The
formation of microporous polyvinylidene difluoride membranes by phase
separation”, J. Membrane Sci., 57, l-20.
Chung, T. S., Lin, W. H., Vora, R. H. (2000a), “The Effect of Shear Rates on Gas
Separation Performance of 6FDA-Durene Polyimide Hollow Fibers.” Journal of
Membrane Science, 167, 55-66
Chung, T. S., Qin, J. J. and Gu, J. (2000b). “Effects of Shear Rate within the
Spinneret on Morphology, Separation Performance and Mechanical Properties of
Ultrafiltration Polyethersulfone Hollow Fiber Membranes.” Chem. Eng. Sci. 55,
1077-1091.
da Silva, P. S., (1997), “ The European lead/acid industry and its future in the world
market”, Journal of Power Sources, 67, 3-6.
Dietz, H., Voss, S., Doring, H., Garche, J., and Wiesener, K., (1990), “Measures for
Minimizing Hydrogen Pressure in Sealed Lead Acid Batteries”, Journal of Power
Sources, 31, 107-113
Dietz, H., Dittmar, L., Ohms, D., Radwan, M., and Wiesener, K., (1992), “Noble
metal-free catalysts for the hydrogen/oxygen recombination in sealed lead-acid
batteries using immobilized electrolytes”, J. Power Sources, 40, 175-186
Hachisuka, H., Ohara, T. and Ikeda, K. (1996). “Gas Permeation Properties of New
Type Asymmetric Membranes.” J. Appl. Polym. Sci. 61, 1615-1619.
Hawkes, H., (1997), “Influences and trends in lead/acid battery demand, lead supply
and prices”, Journal of Power Sources, 67, 213-218.
Henn, F. E., Rouvet, C., de Guibert, A., Martue, P., (1996), “Hydrogen and Oxygen
Evolution in Sealed Laed/acid 2V Cells. In situ Gas Measurement By Raman
Spectroscopy.”, Journal of Power Sources, 63, 235-246
Hwang, J. R., Koo, S. H., Kim, J. H.,Higuchi, A. and Tak, T. M. (1996) “Effects of
Casting Solution Composition on Performance of Polyethersulfone Membrane.”
Journal of Applied Polymer Science, 60, 1343-1348.
Kakuta , A., Ozaki, O., and Ohno, M. (1978). “Gas Permeation in Freeze Dried
Cellulose Acetate Membrane”, J. Polym. Sci.Polym. Chem Ed., 16, 3249-3257
Koenhen, D. M., Mulder, M.H.V. and Smolders, C. A., (1977), “Phase separation
phenomena during the formation of asymmetric membranes”, J. Applied. Polymer.
Science., 21, 199-215.
Lai, Jui-Yih, Lin, F. C., Wang, D.M., (1996), “Asymmetric TPX membranes with
high gas flux”, J. Membrane Science. 110, 81-181
Lai, J. Y., Lin, F. C., Wu, T. T. and Wang D. M. (1999). “On the Formation of
Macrovoid in PMMA Membranes.” Journal of Membrane Science, 155, 31-43.
Linden, D., (1995). “ Lead Acid Battery”, Handbook of Batteries, Chapter 24,
McGraw Hill, New York.
Maja, M., Penazzi, N., (1988), “Effect of some elements on oxygen reduction and
hydrogen evolution at lead-acid battery negative plates”, Journal of Power Sources,
22, 1-9
Matz, R., (1972), “The structure of cellulose acetate membranes I. The development
of porous structures in anisotropic membranes”, Desalination, 10, 1-15
Pinnau, I., Koros, W. J., (1992), “Influence of Quench Medium on The Structures
and Gas Permeation Properties of Polysulfone Membranes Made by Dry/Wet Phase
Inversion Process”, Journal of Membrane Science, 71, 81-96.
Pinnau, I., Koros, W. J., (1993). “A Qualitative Skin Layer Formation Mechanism
for Membranes Made by Dry/Wet Phase Inversion.” J. Polym Sci: Part B: Polym.
Phys. 31, 419-427
Ruetschi, P.,(1978), “Review of the Lead Acid Battery Science and Technology”,
Journal of Power Sources, 2.
Sharpe, I. D., Ismail, A. F., Shilton, S. J. (1999), “A Study of Extrusion Shear and
Forced Convection Residence Time in Spinning of Polysulfone Hollow Fiber
Membranes for Gas Separation.” Separation and Purification Technology, 17, 101-
109
Shilton, S. J., Bell, G., Ferguson, J. (1994), “The Rheology of Fiber Spinning and the
Properties of Hollow Fiber Membranes for Gas Separation.” Polymer, 35, 5327-5335
Shilton, S. J., Bell, G., Ferguson, J. (1996), “The Deduction of Fine Structural
Details of Gas Separation Hollow Fiber Membrane Using Resistance Modelling of
Gas Permeation.” Polymer, 37, 485-492.
100
Shilton, S. J., Ismail, A. F., Dunkin, I. R., Gallivan, S. L., Gough, P. J. (1997),
“Molecular Orientation and the Performance of Synthetic Polymeric Membranes for
Gas Separation.”, Polymer, 38, 2215-2220
Spillman, R. W., Sherwin, M. B., (1990). “Gas Separation Membranes: The First
Decade”, Chemtech, 378-384
Stern, S. A., (1994). “Review-Polymer for gas separation: the next decade.” Journal
of Membrane Science, 1-65.
Strathmann, H., Kock, K., Amar, P. and Baker, R. W., ( 1975), “The formation
mechanism of asymmetric membranes”, Desalination, 16, 179-203.
Wang, D. L., Li, K. and Teo, W. K. (1995). “Relationship between Mass Ratio of
Nonsolvent-Additive to Solvent in Membrane Casting Solution and Its Coagulation
Value.” Journal of Membrane Science, 98, 233-240.
101
Weber, A. (1979). “Raman Spectroscopy of Gases and Liquids” 1st edition New
York: Springer-Verlag, Inc.
Wong, J., (1992), “ Status of the lead acid battery industry in Malaysia”, Journal of
Power Sources, 40, 105-111
Young, T. H., Chen, L. W., (1995), “Pore formation mechanism of membranes from
phase inversion process”, Desalination, 103, 233-247.
Ismail, A. F., Hafiz, W. A. and T. C. Khoo., (2004), “Effect of Shear Rate on the
Performance of Membrane-assisted Lead Acid Battery.”, 2nd Regional
Symposium on Membrane Science and Technology, Johor Bahru, 21-25
April 2004
Battery Electrolyte
Time Mass Losses
(sec) (kg) (kg)
10 12.72 0.02
20 12.72 0.02
30 12.72 0.02
40 12.72 0.02
60 12.7 0.04
80 12.7 0.04
100 12.7 0.04
120 12.7 0.04
140 12.68 0.06
160 12.68 0.06
180 12.66 0.08
Battery Elecrolyte
Time Mass Losses
(sec) (kg) (kg)
10 13.48 0
20 13.46 0.02
30 13.46 0.02
40 13.46 0.02
60 13.46 0.02
80 13.44 0.04
100 13.44 0.04
120 13.44 0.04
140 13.42 0.06
160 13.42 0.06
180 13.42 0.06
Battery Electrolyte
Time Mass Losses
(sec) (kg) (kg)
10 13.94 0
20 13.94 0
30 13.94 0
40 13.94 0
60 13.94 0
80 13.94 0
100 13.94 0
120 13.94 0
140 13.92 0.02
160 13.92 0.02
180 13.92 0.02
Battery Electrolyte
Time Mass Losses
(sec) (kg) (kg)
10 13.92 0
20 13.92 0
30 13.92 0
40 13.92 0
60 13.92 0
80 13.92 0
100 13.92 0
120 13.92 0
140 13.9 0.02
160 13.9 0.02
180 13.9 0.02
Battery Electrolyte
Time Mass Losses
(sec) (kg) (kg)
10 13.66 0
20 13.66 0
30 13.66 0
40 13.66 0
60 13.66 0
80 13.66 0
100 13.64 0.02
120 13.64 0.02
140 13.64 0.02
160 13.64 0.02
180 13.64 0.02
Battery Electrolyte
Time Mass Losses
(sec) (kg) (kg)
10 13.78 0
20 13.78 0
30 13.78 0
40 13.78 0
60 13.76 0.02
80 13.76 0.02
100 13.74 0.04
120 13.74 0.04
140 13.74 0.04
160 13.72 0.06
180 13.72 0.06
Battery Electrolyte
Time Mass Losses
(sec) (kg) (kg)
10 13.78 0
20 13.78 0
30 13.78 0
40 13.76 0.02
60 13.76 0.02
80 13.76 0.02
100 13.76 0.02
120 13.74 0.04
140 13.74 0.04
160 13.72 0.06
180 13.72 0.06
Appendix A2
Electrolyte Losses Profile During Membrane-assisted Lead Acid Battery Charging
Test
13.805
13.8
13.795
Electrolyte Losses (kg)
13.79
13.785
13.78
13.775
13.77
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180
Charging Time (minute)
13.825
13.82
13.815
Electrolyte Losses (kg)
13.81
13.805
13.8
13.795
13.79
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180
Charging Time(minute)
13.81
13.8
13.79
13.78
Electrolyte Losses (kg)
13.77
13.76
13.75
13.74
13.73
13.72
13.71
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180
Charging Time (minute)
Appendix A3 (a)
Membrane-assisted Lead Acid Battery Charging Test Results with Heat Supply
Control Test (Conventional Battery)
Water Bath Temperature = 60oC
Appendix A3 (b)
Membrane-assisted Lead Acid Battery Charging Test Results With Heat Supply
Membrane Used = 9s Membrane (233.33 s-1)
Water Bath Temperature = 60oC
Appendix A3 (c)
Membrane-assisted Lead Acid Battery Charging Test Results with Heat Supply
Control Test (Conventional Battery)
Water Bath Temperature = 80oC
Appendix A3 (d)
Membrane-assisted Lead Acid Battery Charging Test Results With Heat Supply
Membrane Used = 9s Membrane (233.33 s-1)
Water Bath Temperature = 80oC
0.007
0.006
Intensity
0.005
0.004
0.003
0.002
4200 4000 3800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0
Raman Shift (cm-1)
0.012
0.01
0.008
0.006 Intensity
0.004
0.002
0
4200 4000 3800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0
Raman Shift (cm-1)
0.0163
0.0143
0.0123
0.0103
Intensity
0.0083
0.0063
0.0043
0.0023
4200 4000 3800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0
Raman Shift (cm-1)
0.011
0.01
0.009
0.008
0.007
Intensity
0.006
0.005
0.004
0.003
0.002
4100 3900 3700 3500 3300 3100 2900 2700 2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
Raman Shift (cm-1)
0.014
0.012
0.01
0.008
Intensity
0.006
0.004
0.002
0
4000 3500 3000 2500 2000 1500 1000 500 0
Raman Shift (cm-1)
0.015
0.014
0.013
0.012
0.011
Intensity
0.01
0.009
0.008
0.007
0.006
4200 4000 3800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600
Raman Shift (cm-1)
0.006
0.0055
0.005
0.0045
Intensity
0.004
0.0035
0.003
0.0025
0.002
4000 3500 3000 2500 2000 1500 1000 500 0
Raman Shift (cm-1)
Raman Spectrum for Vaporized Electrolyte (Membrane Assisted Battery) Sample Taken
After 40 minutes of Charging Process
0.0045
0.004
0.0035
0.003
Intensity
0.0025
0.002
0.0015
0.001
5000 4000 3000 2000 1000 0
Raman Shift (cm-1)
Raman Spectrum for Vaporized Electrolyte (Membrane Assisted Battery) Sample Taken
After 60 minutes of Charging Process
120
0.0063
0.0058
0.0053
0.0048
0.0043
Intensity
0.0038
0.0033
0.0028
0.0023
0.0018
0.0013
4000 3500 3000 2500 2000 1500 1000 500 0
Raman Shift (cm-1)
0.0065
0.006
0.0055
0.005
0.0045
Intensity
0.004
0.0035
0.003
0.0025
0.002
4100 3900 3700 3500 3300 3100 2900 2700 2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
Raman Shift (cm-1)
Raman Spectrum for Vaporized Electrolyte (Membrane Assisted Battery) Sample Taken
After 120 minutes of Charging Process
121
0.0065
0.006
0.0055
0.005
0.0045
Iintensity
0.004
0.0035
0.003
0.0025
0.002
4000 3500 3000 2500 2000 1500 1000 500 0
Raman Shift (cm-1)
0.0065
0.006
0.0055
0.005
0.0045
Intensity
0.004
0.0035
0.003
0.0025
0.002
4000 3500 3000 2500 2000 1500 1000 500 0
Raman Shift (cm-1)
Raman Spectrum for Vaporized Electrolyte (Membrane Assisted Battery) Sample Taken
After 160 minutes of Charging Process