close

Вход

Забыли?

вход по аккаунту

?

Silica Nafion Modified Composite Membranes for Direct Methanol Fuel Cells.

код для вставкиСкачать
Dev. Chem. Eng. Mineral Process. 14(1/2), pp. 119-131,2006.
Silica Nafion Modified Composite
Membranes for Direct Methanol Fuel Cells
J.I. Garnica Rodriguez, A.L. Dicks, M.C. Duke and
J.C. Diniz da Costa*
ARC Centrefor Functional Nanomaterials, School of Engineering,
The University of Queensland, Bris bane, Queensland 4072, Australia
~
~~
~
~
~~
Direct Methanol Fuel Cells (DMFCs) operate by electro-oxidation of methanol and
the transport of the protons by means of a polymer electrolyte membrane. Current
systems employ Nafion (perjluorosulphonic acid) membranes as the transport media
for the protons, which experience a high methanol crossover reducing the efjiciency
by the oxygen reduction reaction at the anode side of the fuel cell. This work
investigated the microstructural modification of Nafion membranes with silica
nanoparticles. It was shown that this mod4cation led to reduced methanol crossover,
whilst methanol permeability was slightly reduced without compromising the
conductivity at the normal DMFC operation temperature (75-80°C). Methanol uptake
was reduced by 55% with the incorporation of the silica nanoparticles into the Nafion
matrix. In addition, pure methanol uptake appeared to be less effective than methanol
mixtures. These results indicated the presence of water assisted methanol permeation
as the hydrophilic sulphonic group and the hydrophobic polymeric backbone of
Nafion were sensitive to methanol solvation.
Introduction
Fuel cells are electrochemical devices that directly convert the energy of a chemical
reaction into electricity. They operate similarly to batteries which store energy,
however, fuel cells can produce electricity as long as fuel and air are supplied. Proton
'Authorfor correspondence (joedac@cheque.uq.edu.au).
I19
J.I. Garnica Rodriguez, A.L. Dick, M.C.Duke and JC. Diniz da Costa
Exchange Membrane fuel cells (PEMFC)are one of the main types of fuel cells
which use solid polymer electrolyte membrane at low temperature (less than 100OC).
Direct methanol fuel cells (DMFC)are a specific type of PEMFC which use an
aqueous solution of methanol as a fuel, as shown in Figure 1. At the anode the
methanol oxidises releasing electrons, protons, and energy whilst at the cathode
oxygen reacts with electrons taken from the electrode and with protons from the
electrolyte to produce water according to the following reactions (Gurau, 2002;
Barragan, 2002):
CH,OH + H,O-CO,
+6Ht +6e-
Anode Reaction
(1)
1.50, + 6 H t + 6 e - +3H,O
Cathode Reaction
(2)
CH,OH + 7.50, +2H,O + CO,
Overall Reaction
(31
Figure 1. Schematic of the DMFC fuet cell process (SmartFuelCells, 2005 [ZO]).
During the last decade DMFC research and development has been focused on
portable power applications such as mobile phones, laptop computers, and portable
cameras and games devices. The long-term goal is to replace the high performance
rechargeable batteries in the US$6 billion portable electronic devices market, based
on the superior theoretical specific energy density of methanol (6000 W a g )
compared with the best rechargeable battery (lithium ion polymer: theoretical,
I20
Silica Nafion Modified Composite Membranes for Direct Methanol Fuel Cells
600 Wh/kg) (Dillon, 2004). Peduorosulfonic acid polymer-based membranes have
been extensively used in PEM fuel cells because of its good chemical stability and
proton conductivity (Jung, 2001), under trademarks such as Aciplex" (Asahi
Chemical Industry), Dow@ (Dow Chemical), Flemion* (Asahi Glass Company),
Primes@ (Gore), and Nafion@(DuPont) (Yoshida, 1998; Doyle, 2003). The latter has
become the preferred commercial electrolyte for DMFC applications due to its
chemical structure and performance.
Perfluorinated membranes differ from conventional ion exchange membranes in
that they are not cross-linked polymeric structures but thermoplastic polymers with
pendant groups (sulphonic or carboxylic) partially or completely neutralized to salts
(Colomban, 1992). Nafion contains a hydrophilic region (ion clusters which allow the
proton migration through the membrane) scattered into hydrophobic regions
(composed of the polymer backbone which restricts the fuel crossover across the
membrane) as shown in Figure 2.
Figure 2. Schematic representation of Nafon microstructure (Kreuer, 2001).
I21
J.I. Garnica Rodriguez, A.L. Dick, M.C. Duke and J.C. Diniz da Costa
The protonic transport properties of perfluorinated membranes are strongly
influenced by the water content of the membrane (Barragan, 2002). In the dry state
Nafion behaves like a proton (and electric) insulator, but when hydrated the
electrolyte membrane becomes proton conductive as a function of the water content.
Eikerling (2003) reported that in perfluorinated membranes at high levels of hydration
proton transport follows the Grotthuss mechanism where the protons hop from one
water molecule to the next via hydrogen bonding.
One of the technical hurdles in DMFCs is related to methanol crossover from the
anode to the cathode. This causes a reduction on fuel efficiency by wastehl oxidation
of methanol at the cathode side while seriously depolarizing the cathode (Gurau,
2002; Cruickshank, 1998; Ramya, 2003). The depolarisation is mainly due to the
parasitic superposition of methanol oxidation currents with oxygen reduction currents.
One of best strategies known to date to minimise the methanol-crossover rate using
Nafion membrane-based cells is to operate the cell with lower methanol
concentrations < 2 M and temperatures < 70°C (Doyle, 2003; Dillon, 2004).
In addition, a large number of structural modifications of PEMs for DMFC
applications have been studied in order to reduce the methanol crossover. These
include grafting of styrene monomer on Nafion by means of supercritical COz
impregnation to build a hydrophobic (methanol repellent) bamer (Sauk, 2004), porefilling type PEM using teflon as substrate and acrylic acid-vinyl sulfonic acid
copolymer as the proton conductive matrix (Yamaguchi, 2003), Nafion modification
with polyvinyl alcohol (Shao, 2002), and silicon dioxide-Nafion composites (Mauritz,
1998; Dimitrova, 2002; Jung, 2002,). All these studies have reached a reduction on
the methanol permeation, but at the same time an undesired reduction of the proton
conductivity.
In this work we investigated the effect of methanol crossover and conductivity in
Nafion modified membranes with silica nanoparticles. An in-situ sol-gel method was
used to hydrolyse and condense silica inside the Nafion matrix as a physical barrier
for methanol diffusion. Synthesised composite membranes with different silica
loadings were tested to determine the potential of this technology for DMFC
application.
I22
Silica Nafion Modfied Composite Membranesfor Direct Methanol Fuel Cells
Experimental Details
Nafion 117 membranes (DuPont) cut into squares with an approximate area of 4 cm2
were modified with silica following the technique developed by Ladewig (2004). In
this technique, silica nanoparticles were embedded into the hydrophilic clusters of the
perfluorinated membrane using a sol-gel process. The advantage of this approach
when compared to similar studies (Mauritz, 1998) is the reduced size of the clusters
that leads to smaller particles. While Mauritz’s technique uses a swelling step in
methanol before silica incorporation, leading to an expansion of the free volume of
the hydrophilic clusters where the silica nanoparticles are formed, our approach
reduces the free volume of the clusters by means of a stabilisation step in low relative
humidity atmosphere. After pre-treatment, the membranes were dned at 100°C and
atmospheric pressure. Then the membranes were stabilised in an atmosphere with a
controlled relative humidity of 40% for 30 minutes and subsequently immersed in a
solution of tetraethyl ortho silicate (TEOS - Aldrich) and ethanol in the molar ratio
1:8 for 1 minute. The water accumulated into the clusters during the stabilisation step
reacted with the TEOS to form silica nanoparticles. After this step, the membranes
were rinsed in pure ethanol to eliminate any build-up of silica on the surface of the
membranes, and returned to the oven for 30 minutes when a higher silica loading was
required. The silica content of the membranes was determined by means of weighing
before and after each cycle. Three membranes with different silica loadings of 1.1,2.6
and 4.8% (w/w) were synthetised in 1, 4, and 8 consecutive modification cycles,
respectively.
The methanol and water permeability of the membranes was determined using a
pervaporative system (see Figure 3). The polymer electrolyte membrane was sealed
between two metallic bulbs, and fed at one side with a liquid stream (methanol-water
solution) where the concentration was maintained constant by means of refluxing
using a peristaltic pump. At the other side, the permeate vapour was carried by a
helium stream and collected in a glass condenser immersed in liquid nitrogen. The
weight of permeate was determined using a digital balance. The concentrations of
methanol and water were determined by gas chromatography using a Shimadzu
GC17A and a column HPLOT-U. Testing was performed at various temperatures
123
J.I. Garnica Rodriguez, A.L. Dick, MC.Duke andLC. Dinu da Costa
Figure 3. Permeability measurement set-up.
(25, 50, 75°C) and methanol concentrations (0, 50, 100% (vlv)). State-of-the-art fuel
cells use 2 M (8% v/v) methanol solutions as a fuel in order to reduce the detrimental
effects of methanol crossover on the electric generation performance. However, we
performed a fimdamental study along the whole range of methanol concentration in
order to acquire valuable information not just related to the methanol transport
mechanisms but also to the microstructural changes of Nafion and silica-Nafion
composite membranes when stabilised in different media.
It is important to note that the permeability values obtained do not include the
electro osmotic drag factor (i.e. methanol and water molecules are "pulled" through
the polymer electrolyte membrane by migrating protons during fuel cell operation
(Barragan, 2004)), the data obtained provided valuable information for preliminary
membrane selection, where the membranes with the most reduced methanol
permeability profiles will be integrated with the electrodes and the catalyst (MEA
assembly) and tested for electricity generation.
Total and selective water-methanol uptake tests were performed in order to
determine the differences in the interaction between the membranes and methanol
andor water because of the silica-modification process. After stabilisation in a
specific methanol concentration solution, membranes were manually dried using filter
paper, quickly weighed on a digital scale, and returned to the methanol solution. This
procedure was repeated five times in order to improve the reliability of the technique.
I24
Silica Nafon Modified Composite Membranes for Direct Methanol Fuel Cells
The normal direction conductivity of the hlly hydrated membranes at 20, 50 and
75°C was measured by means of impedance spectroscopy on a Solartron 1260 (Stand
Alone mode, frequency sweep 10 MHz to 1 Hz).The membrane resistance for proton
conduction was obtained by the difference between the measured resistance and the
contribution of the short circuited electrodes.
Results and Analysis
Table 1 lists the thicknesses of Nafion and composite membranes when fully
hydrated. The silica loading varied from 1.1% to 4.8% by weight in the composite
membranes. The embedding of silica particles in the Nafion structure allowed an
incremental increase of the membrane thicknesses as a hnction of the silica loading.
As Nafion contains hydrophilic clusters, membranes swelled to accommodate the
extra volume of absorbed water. These results suggest silica did not hinder water
absorption which is beneficial for DMFCs as proton conduction occurs via the
Grotthuss mechanism.
Membrane
Thickness (pm)
Nafion
Nafion-silica 1.I %
Nafon-silica 2.6 %
Nafion-silica 4.8 %
187*2
197* 1
202 f 1
201 3
*
Figures 4 and 5 show that the permeabilities for water and methanol increased
with temperature for all membranes tested. This is mainly attributed to the relaxation
of the Nafion polymeric chains which results in lower resistance to the permeation of
molecules. At the same time, high temperature led to an increase in the activation
energy for diffusion of both water and methanol molecules, providing a double fold
effect in high permeation levels. The incorporation of silica particles in the polymer
matrix clearly indicated a reduction of permeation as compared to Nafion, except in
12s
J.I. Garnica Rodriguez, A.L. Dick. M.C. Duke and J.C. Diniz da Costa
25
30
35
40
45
50
55
60
Temperature ("C)
65
70
75
65
70
75
Figure 4. Distilled water permeability
25
30
35
40
45
50
55
60
Temperature ("C)
Figure 5. 50% (vh) solution and pure methanol permeability
126
Silica Nafon Modified Composite Membranesfor Direct Methanol Fuel Cells
the case of pure methanol permeation. In Figure 4, the silica-Nafion membranes with
h g h (4.8%) and low (1.1%) silica content experienced reduced permeation
(especially at high temperature), as opposed to the medium (2.6%) silica content
membrane which resulted in permeation levels much closer to pure Nafion. This
variability warrants further investigation which may be attributed to the silica particle
size and to the in-situ silica synthesis process. The permeation of 50% (vh) methanol
is generally three to five times higher than water permeation. Pure methanol
permeation at 75OC on the other hand is slightly higher than water permeation at the
same conditions. The silica-Nafion composite membranes with low (1.1%) and high
(4.8%) silica content showed similar permeability profiles for methanol 50% solutions
as the silica-Nation membrane with medium (2.6) silica content, therefore only the
last one is included on Figure 5 for comparison with Nation. For pure methanol
permeation, only the silica-Nafion 2.6%membrane was tested.
Total uptake results shown in Figure 6 indicated that the membrane weight
generally increased resulting in maximum uptake at 75% v/v methanol concentration.
Rivin (2001) suggested that the solubility of alcohols in Nafion is due to the
interaction between the h c t i o n a l groups of Nafion where its sulphonic acid residues
are involved in strong mutual pair-wise or multiple hydrogen bonding interactions
which act as effective cross links in the absence of water. Thus, the peak observed on
the uptake profile of Nafion is produced by the disruption of the sulphonic acid
hydrogen bonding by water. Silica Nafion composite membranes also peaked at
75% v/v methanol with a much lower uptake as compared to Nation. Thisobservation,
combined with the resufts, strongly suggest silica nanoparticles embedded in the
Nafion matrix promoted some level of interference on the sulphonic group and
watedmethanol interactions. In other words, water may facilitate the transport of
methanol through the membrane, and its reduction after the maxima at 75% (v/v)
methanol concentration also leads to a reduction in methanol uptake.
The difference in the interaction of water and methanol within the Nafion matrix
was evident by a five-fold larger uptake from 24% water to 125% methanol as shown
in Figure 6. Pure methanol uptake appeared to be less effective than methanol
mixtures where water is competing with the inter-sulphonic acid hydrogen bonds.
127
J.I. Garnica Rodriguez, A . L . Dick, M.C. Duke and J.C. Diniz da Costa
135
SN 2.6%
SN 4.0%
120
105
z
L
I
Q
Y
75
-
3
$ 6 0
I45
30
0
10
20
30
40
50
60
70
80
90
100
Methanol Concentration (vh)
Figure 6. Total uptake (w/w)profiles at 20°C.
Rivin (2001) reported that the appreciable solubility of the dry alcohols may be
attributed to the solvation of the fluoroether rich regions of Nafion. This result may
also suggest a semi-hydrophilic zone between the hydrophilic sulphonic group and the
hydrophobic polymeric backbone of Nafion sensitive to methanol solvation as
proposed by Berezina (2004).
Despite the 50% reduction of methanol uptake on the 2.6% silica-Nafion
composite membrane, its permeation profiles for methanol solutions are similar to
Nafion (see Figure 5 ) . This highlights the complexity of the transport mechanisms of
methanol in Nafion, and indicates that the diffusion of methanol through the polymer
electrolyte membrane occurs mainly through the semi-hydrophillic zone of Nafion
which was not affected by the silica-modification process.
The proton conductivity values for the silica-Nafion membranes are depicted in
Figure 7. Nafion outperformed the silica composite membranes at 20"C, whilst all
values were very similar at 50°C. The latter results agree to the findings of Jung
(2002) that silica-Nafion composite membranes synthesized using a similar sol-gel
I28
Silica Nafion Modified Composite Membranesfor Direct Methanol Fuel Cells
technique showed no significant effect on the proton conduction of Nafion at loadings
below 5% (w/w). This fact suggests that there is no interference provided by the silica
nanoparticles embedded on the hydrophilic clusters of Nafion on its proton transport
mechanisms, due to the low proton conductivity of silica (i.e. 0.04 mS/cm as reported
by Daiko, 2004). At 75"C, the 2.6% silica Nafion membrane showed a slight higher
conductivity than Nafion or the other composite membranes. As DMFCs are likely to
operate at around 75-80°C, the incorporation of silica particles into the Nafion
polymeric matrix are justified by a reduction of methanol permeation and uptake
whilst conductivities are not compromised. Although conductivities as h g h as
100 mS.cm" were reported by Affoune (2005) for hydrated Nafion 117 at ambient
temperature, our results are in better agreement with the values of 15 mS.cm-' and
18 mS.cm-' reported by Cho (2004) and Xu (2005), respectively. These variances
hlghlight the fact that the final values of conductivity reported in the literature are not
dependant on the analytical technique used, but on the design of the cell used for the
testing of the membranes converting the conductivity on a relative value.
33
4-Naflon
30
-?
E
27
U)
r!
E
v
24
.-2
21
.->
+
SN1.1%
& SN2.6%
+SN4.8%
c
9
10
24
30
36
42
40
54
60
66
72
78
Temperature ("C)
Figure 7. Proton conductivity of fully hydrated membranes.
129
f.I. Garnica Rodriguez, A .L. Dicks, M.C. Duke and J.C. Diniz da Costa
Conclusions
Nafion for proton conducting membranes in DMFC was successfully modified with
various silica loadings to reduce methanol crossover. Permeation of pure water and
50% (vh) methanol solutions revealed the effects of silica modification of Nafion
compared to pure Nafion. Transport through Nafion is activated with temperature due
to changes to the polymeric structure and increased activation energy for diffusion. In
general, silica modification reduced methanol and water permeability, but great
reductions in methanol uptake were observed for solutions above 75% (v/v). The
silica nanoparticles reduced the assistance to methanol permeation provided by the
hydrophilic sulphonic groups and the polymeric network of the Nafion structure.
Silica nanoparticles are therefore effective for the reduction of methanol crossover
and uptake into proton conducting membranes for DMFC. Although the proton
conductivity of silica is low, the proton transport properties of Nafion were not
affected after the modification with silica due to the low content of silica
nanoparticles.
Acknowledgements
The authors would like to acknowledge the ARC Centre for Functional Nanomaterials
for financial and technical assistance.
References
1.
2.
Affoune, A.M., A. Yamada, and M. Umeda, Conductivity and surface morphology of Nafion
membrane in water and alcohol environments. J. Power Sources, In Press, Corrected Proof.
Barragan, V.M. and A. Heinzel, 2002. Estimation of the membrane methanol diffusion coefficient
from open circuit voltage measurements in a direct methanol fuel cell. J. Power Sources, 104(1), 6672.
2.
3.
4.
5.
6.
7.
130
Barragan, V.M., et al., 2004. On the methanol-water electroosmotic transport in a Nafion membrane.
J. Membr. Sci., 236( 1-2), 109-120.
Berezina, N.P., 2002. Transport-structural parameters of perfluorinated membranes Nation-I 17 and
MF-45K. Russian Journal of Electrochemistry, 38(8), 903-908.
Cho, K.-Y., et al., 2004. Characteristics of PVdF copolymerhlafion blend membrane for direct
methanol fuel cell (DMFC). Efectrochim. Acfn,. SO(2-3), 580-585.
Colomban, P. (ed), 1992. Proton conductors; solids. membranes, and gels: Mnterials and devices.
Cambridge University Press. Cambridge, UK.
Cruickshank, J. and K. Scott, 1998. The degree and effect of methanol crossover in the direct
methanol fuel cell. J. Power Sources, 70(1), 40-47.
Dillon, R., et al., 2004. International activities in DMFC R&D: Status of technologies and potential
applications. J. Power Sources, 127(1-Z), 112-126.
Silica Nafion Modlfied Composite Membranes for Direct Methanol Fuel Cells
8.
9.
10.
II.
12.
13.
14.
15.
l5.
16.
17.
18.
18.
19.
20.
21.
Dimitrova, P., et al., 2002. Transport properties of ionomer composite membranes for direct methanol
fuel cells. J. Elecrronnnlyticnl Chemistry, 532( 1-2), 75-83.
Doyle, M. and G. Rajendran, 2003. In: Vielstich, W., A. Lamm and H.A. Gasteiger (ed.), Hnndbook
of Fuel Cells: Fundnmenmfs, technology. and npplicnlions, Vol. 3. pp.35 1-395. Wiley, Chichester.
Eikerling, M., et al., 2003. Defect structure for proton transport in a triflic acid monohydrate solid.
Chem. Phys. Len., 368(1-2), 108-1 14.
Gurau, B. and E.S. Smotkin, 2002. Methanol crossover in direct methanol fuel cells: a link between
power and energy density. J. Power Sources, 112(2), 339-352.
Jung, D.H. et al., 2001. A performance evaluation of direct methanol fuel cell using impregnated
tetraethyl-orthosilicate in cross-linked polymer membrane. fnr. J. Hydrogen Energy, 26( 12), 1263-69.
Jung, D.H., et al., 2002. Performance evaluation of a Nafion/silicon oxide hybrid membrane for direct
methanol fuel cell. J. Power Sources, 106(1-2). 173-177.
Kreuer, K.D., 2001, On the development of proton conducting polymer membranes for hydrogen and
methanol fuel cells. J. Membr. Sci., 185(1).29-39.
Ladewig, B., 2005. Personnl Communicntion
Mauritz, K.A., 1998. Organic-inorganic hybrid materials: Perfluorinated ionomers as sol-gel
polymenzation templates for inorganic alkoxides. Mnter. Sci. Eng. C, 6(2-3). I21 -1 33.
Ramya, K. and K.S. Dhathathreyan, 2003. Direct methanol fuel cells: determination of fuel crossover
in a polymer electrolyte membrane. J. Eledroonnlylicnl Chemistry, 542, 109-1 15.
Rivin, D. et al., 2001. Solubility and transport behavior of water and alcohols in
Polymer,
42(2), 623-635.
Sauk, J. et al., 2004, Grafting of styrene on to Nafion membranes using supercritical COz
impregnation for direct methanol fuel cells. J. Power Sources, 132(1-2), 59-63
Shao, Z.-G., X. Wang, and I.-M. Hsing, 2002. Composite Nafion/polyvinyl alcohol membranes for
the direct methanol fuel cell. J. Membr. Sci., 210(1), 147-153.
Skou, E., P. Kauranen, and J. Hentschel, 1997. Water and methanol uptake in proton conducting
NationtR’membranes. Solid Stole lonics, 97(1-4), 333-337.
SmnrtFuefCeits. Direct methanol fuel cell technology - Mode of operation. Available online at:
h~~p://www.smnrIfuelceN.~e/index.php?id=
I41 &L = I
Xu, W. et al., 2005. Low methanol permeable composite Nafion/silica/PWA membranes for low
temperature direct methanol fuel cells. Eleclrochim. Acln, SO( 16-17), 3280-3285.
22. Yamaguchi, T., F. Miyata, and
Nakao, 2003. Pore-filling type polymer electrolyte
membranes for a direct methanol fuel cell. J. Membr. Sci., 214(2), 283-292.
131
Документ
Категория
Без категории
Просмотров
5
Размер файла
663 Кб
Теги
fuel, direct, modified, compositum, membranes, silica, nafion, cells, methanol
1/--страниц
Пожаловаться на содержимое документа