pubs.acs.org/NanoLett
Aligned Electrospun Nanofiber Composite Membranes for Fuel Cell Electrolytes Takuya Tamura, and Hiroyoshi Kawakami* Department of Applied Chemistry, Tokyo Metropolitan University, Hachioji, Tokyo 192-0397, Japan ABSTRACT We have synthesized the novel composite membranes composed of sulfonated polyimide nanofibers and sulfonated polyimide for proton exchange membrane fuel cell. It was clear that the polyimides within nanofiber were significantly oriented or aggregated when electrospun; as the result, the membrane stability, such as oxidative and hydrolytic stabilities, of the composite membrane was significantly improved with an increase in nanofiber, and oxygen permeability of the composite membrane also decreased when compared to that determined in the membrane without nanofibers. In addition, the proton conductivity of the membrane in the parallel direction indicated a significantly higher value when compared to that determined for the membrane in the perpendicular direction or for the membrane without nanofibers prepared with conventional solvent-casting method. Consequently, nanofibers proved to be promising materials as a proton exchange membrane and the composite membrane containing nanofibers may have potential application for use in fuel cells. KEYWORDS Aligned nanofiber, electrospining, fuel cell, proton conductivity, sulfonated copolyimide
T
received a lot of attention.11-15 Electrospinning is capable of producing fibers with diameters in the nanometer range, and the electrospun nanofibers possess many unique properties including a large specific surface area, superior mechanical properties, and use as nanoscale building blocks. However, there are only a few reports in the literature on the proton conductivity of electrospun nanofibrous mats, in addition, the fibrous structure was an isotropic nonwoven mat.16-18 If the polymer electrolyte membrane is composed of uniaxially aligned nanofibers having a high proton conductivity, the proton may be rapidly transported through the nanofiber. Here, we describe the first such composite membranes composed of uniaxially aligned sulfonated polyimide nanofibers and sulfonated polyimide for using proton exchange membrane fuel cell, which revealed a dramatically enhanced proton conductivity, lower gas permeability, and longer durability. The electrospinning process is a method of discharging a polymer solution in air from a nozzle under high voltage and producing a nanofiber by exploiting electrostatic repulsion of the polymer solution. When electrospinning is carried out for the polymer having both hydrophobic and hydrophilic domains, the hydrophobic and hydrophilic domains in the polymer may be separated to the outside as the air surface and the inside of the polymer solution, respectively. Consequently, the proton channel structure due to the network of the sulfonic acid groups formed within the nanofiber may lead to the rapid transport of the proton without forming a proton transport site by exploiting the phase separation like a sulfonated block copolymer (Figure 1b). Furthermore, since the polymers within the uniaxailly aliged nanofiber are strongly oriented in the axial direction of the nanofiber, the mechanical strength of the nanofiber may be remarkably improved and, additionally, phase
he proton exchange membrane fuel cell converts chemical energy directly into electrical energy with a high efficiency and low emission of pollutants and is the most promising power source for portable and automotive applications. The proton exchange membrane is one of the key components in fuel cell systems.1,2 The important technical focus is on developing polymer electrolyte membranes that are able to achieve a high proton conductivity, low gas permeability into the fuel and oxidant, sufficient thermal stability, and long-term durability. The perfluorosulfonated membranes such as Nafion have been widely used because of their excellent oxidative and chemical stability as well as high-proton conductivity.3,4 They however have limitations like a low thermal stability due to their low Tg and high gas permeability. Recently, much effort has gone into the development of novel polymer electrolyte membranes based on the sulfonated aromatic hydrocarbon polymers, which have been widely synthesized as alternate candidates due to their excellent chemical and thermal stabilities and good mechanical strength.5-10 However, most of them contain a large number of sulfonic acid groups to enhance the proton conductivity, resulting in an unfavorable swelling of the membranes and a dramatic loss in their mechanical properties. In addition, they have critical problems in realizing both a high proton conductivity and low gas permeability into the fuel, because the polymer electrolyte membranes with a high proton conductivity generally have a high gas permeability. Recently, nanofibers prepared through an electrically charged jet of a polymer solution/melt (electrospinning) have * To whom correspondence should be addressed. E-mail: kawakami-hiroyoshi@ c.metro-u.ac.jp. Phone: +81-426-77-1111 (Ext) 4972. Fax: +81-426-77-2821. Received for review: 12/15/2009 Published on Web: 03/26/2010 © 2010 American Chemical Society
1324
DOI: 10.1021/nl1007079 | Nano Lett. 2010, 10, 1324–1328
trodes.20 We also considered that the electrostatic forces formed between the charged nanofiber and the aluminum plate determined the aligned direction of the nanofibers. The mean diameter of the nanofibers was 199 ( 37 nm (n ) 30). In this study, the proton conductivities and stabilities for the composite membranes composed of the sulfonated random copolyimide and the uniaxially aligned polyimide nanofibers were investigated. The thermal and chemical stabilities of the proton exchange membrane are key issues in fuel cell application. The coefficient of thermal expansion (CTE) and the oxidative and hydrolytic stabilities for the membrane and nanofibers prepared from NTDA-BDSA-r-APPF were measured. As apparent from Table 1, the CTE value for the nanofiber was remarkably reduced when compared to that determined in the membrane and the composite membrane containing nanofibers and indicated excellent thermal stability, suggesting that the polyimides within nanofiber were significantly oriented or aggregated when electrospun.21-23 In addition, the composite membranes exhibited more stability for Fenton’s reaction than the membrane due to the orientation of the polyimides. The hydrolytic stability of the composite membrane was also remarkably improved due to the introduction of the nanofibers. Figure 2 shows the proton conductivity of the composite membrane containing the nanofibers. In this study, the proton conductivities of the composite membranes were measured in transverse direction, although the proton transport current in a fuel cell is perpendicular to the membrane surface (Figure 2a). In addition, the proton conductivity was determined by measuring it in the perpendicular and parallel directions to the aligned nanofiber direction. The proton conductivity estimated in the parallel direction for the composite membrane significantly increased the increasing nanofibers and was significantly enhanced when compared to the dense membrane without nanofibers. In most polymer electrolyte membranes, the proton conductivities of the membranes are strongly related to the amount of sulfonic acid groups or the water content of the membrane, and polymer electrolyte membranes absorbing a large amount of water typically have a high proton conductivity.24-26 Therefore, there is a good correlation between the IEC or water uptake and the proton conductivity. However, as apparent from Table 2, IEC or water uptake values of the composite membrane are almost constant regardless of the amount of nanofibers, indicating that the polymer structure, such as phase separation formed in the nanofiber, influenced the proton conductivity. The water uptake has been expressed as the number of water molecules per sulfonic group (λ)24,27
FIGURE 1. (a) Chemical structure of NTDA-BDSA-r-APPF. (b) Schematic representation of sulfonated copolyimide nanofiber. (c) TEM image of cross-sectional aligned nanofiber in radial direction. (d) TEM image of cross-sectional aligned nanofiber in axial direction. (d) SEM image of aligned nanofiber electrospun on special designed collector.
separation in the polymer having both hydrophobic and hydrophilic domains may be also facilitated. Moreover, since the gas diffusion through a nanofiber is inhibited by the aggregated structure, the gas permeability or gas crossover in a nanofiber may also be remarkably reduced. To confirm these hypotheses, the phase separation structure in a sulfonated random copolyimide nanofiber was observed using TEM (Figure 1c,d). The chemical structure of the polyimide, NTDA-BDSA-r-APPF, was shown in Figure 1a. The nanofiber was stained with Pb+ ions. The dark areas represent hydrophilic domains and the brighter areas represent hydrophobic domains. TEM images in the radial and axial directions disclosed many amounts of sulfonic acid groups the inside of the nanofibers, indicating that a proton may be efficiently transported within the nanofiber. Figure 1e shows the SEM image of the sulfonated random copolyimide nanofiber aligned on the collector after the electrospinning. The electrospinning was performed on a specially designed collector equipped with conductive aluminum plates and glass insulator materials.19 This image clearly demonstrated that the nanofibers were uniaxially aligned and were individually deposited across the gap between the aluminum plates without any aggregation. Li et al. reported that the nanofibers electrospun on dual grounded collection plates were uniaxially aligned because of the two sets of electrostatic forces in the charged nanofibers; the first set originating from the split electric field and the second one between the charged fiber and image charges induced on the surface of the dual grounded elec© 2010 American Chemical Society
λ)
1325
n(H2O) -
n(SO3 )
)
WS 18IEC
(1)
DOI: 10.1021/nl1007079 | Nano Lett. 2010, 10, 1324-–1328
TABLE 1. Coefficient of Thermal Expansion (CTE), Oxidative, and Hydrolytic Stabilities of Composite Membranes Containing Aligned Nanofibers weight ratio membrane
membrane-nanofiber
CTEa (ppm/K)
oxidative stabilityb (h)
hydrolytic stabilityc (h)
membrane composite membrane composite membrane composite membrane nanofiber
100:0 99:1 95:5 90:10 0:100
29
10 10 11 13
750 1000 1400 1600
25 8.3
a The measurement temperature was from 80 to 150 °C. b Fenton stability was characterized by the time that the membranes was completely dissolved in 3% H2O2 containing 2 ppm FeSO4 at 80 °C. c Hydrolytic stability was characterized by the time that the membranes was completely dissolved in water at 80 °C.
per sulfonic group calculated from eq 2. Although the IEC values of the membranes were very similar, their λ values slightly increased with the increasing amount of the nanofibers. In addition, their λb values were enhanced with nanofiber contents, while their λf values indicated no significant difference among the membranes. The λb of the composite membrane (3) significantly increased, as apparent from Table 2. The network of the sulfonic acid groups formed within the nanofiber may lead to the enhanced λb. We believe that the high proton conductivities of the composite membranes are responsible for the amount of the bound water. On the other hand, the proton conductivity in the perpendicular direction of the composite membrane also increased when compared to the dense membrane not containing nanofibers as shown in Figure 2a, although the increased rate was lower than that measured in the parallel direction. This may be because the uniaxially alignment of nanofibers on the top surface was disordered due to a decrease of the electrostatic force formed between the charged nanofiber and the aluminum plate collector at the large amount of electrospun nanofiber and that the proton conductivity in the perpendicular direction increased. It was found that the proton conductivity of the composite membrane strongly depends on the direction of an aligned nanofiber and that the proton conductivity in the parallel direction for the composite membrane containing 10 wt % nanofibers was similar to that for Nafion. In addition, the proton conductivity measured in the parallel direction at low relative humidity of 30% for the composite membrane indicated 3 and 10 times larger value when compared to that in the perpendicular direction for the membrane and in the membrane without nanofibers (2.7 × 10-5 s/cm), respectively. This may be because the composite membrane has a large amount of the bound water as shown in Table 2. However, we are very interested in a proton conductivity of one nanofiber. Therefore, the apparent proton conductivity of one nanofiber was calculated with the following equation
FIGURE 2. (a) Proton conductivity of composite membrane containing aligned nanofibers for parallel (σ//) and perpendicular (σ⊥) directions at 80 °C and 98% RH. (b) Calculation of apparent proton conductivity of one nanofiber. (c) Apparent proton conductivity of nanofiber estimated by simulation.
where n(H2O) is the H2O mole number, n(SO3) is the SO3 group mole number, WS is the water uptake value by weight, IEC is the ion exchange capacity, and 18 corresponds to water’s molecular weight. The hydrated state of the membrane was estimated from the endothermic peaks corresponding to the water melting measured by DSC.28,29
λ ) λf + λb
(2)
where λ is the total hydration number calculated from eq 1, λf is the number of free water molecules per sulfonic group measured by DSC, and λb is the number of bound waters © 2010 American Chemical Society
δ ) δ1φ1 + δ2φ2
1326
(3)
DOI: 10.1021/nl1007079 | Nano Lett. 2010, 10, 1324-–1328
TABLE 2. Ion Exchange Capacity (IEC), Water Uptake, and Number of Water Molecules Per Sulfonic Group of Composite Membranes Containing Aligned Nanofibers weight ratio
IEC valuea
membrane
membrane-nano fiber
(experiment/theory)
water uptakeb (%)
λc
λfd
λbe
membrane composite membrane composite membrane composite membrane
100:0 99:1 95:5 90:10
1.41/1.50 1.42/1.50 1.42/1.50 1.41/1.50
22 22 27 29
8 8 10 12
2 1 1 2
6 7 9 10
a IEC value determined by titration method of membrane that immersed in 1 N NaOH. b Membrane was immersed in water at room temperature. c λ ) Water sorption/(18 × IEC); λ ) λf + λb. λ: the number of water molecules per sulfonic group. d λf: the number of free water molecules per sulfonic group. e λb: the number of bound water molecules per sulfonic group.
where δ is the proton conductivity measured using electrochemical impedance spectroscopy, δ1 is the proton conductivity of the membrane without the nanofiber, and δ2 is the apparent proton conductivity of one nanofiber. φ1 is the volume fraction of the membrane and φ2 is the volume fraction of one nanofiber. The apparent proton conductivity of the nanofiber markedly increased when compared to the composite membrane containing the nanofibers (Figure 2b). Although the proton conductivity should have been constant without depending on the amount of nanofibers, the conductivity decreased, however, with an increase in the nanofibers. This means that eq 3 employed for calculation of the apparent proton conductivity of one nanofiber did not necessarily work well. Although the equation indicates that the protons transport through only the polymer and nanofiber, the nanofiber surface slightly dissolved by the polymer solution during the membrane preparation should have also been considered as a novel proton transporting site. However, it is very difficult to precisely measure the dissolved surface area, therefore, the proton conductivity of one nanofiber was simulated by values of the apparent proton conductivity of the nanofiber calculated using eq 3 and the experimental proton conductivity determined from the composite membrane containing aligned nanofibers (Figure 2c). Apparent proton conductivity of the nanofiber obtained by simulation in the parallel direction showed a 15 times higher value than that of the dense membrane without nanofibers and, in addition, that was superior to that in perpendicular direction. These findings may indicate that the proton channel structure formed in the nanofibers rapidly and efficiently transports the proton. The developing polymer electrolyte membranes are also required to be able to achieve a low gas permeability into the fuel and oxidant. Table 3 shows gas permeability of the composite membrane containing aligned nanofibers. The gas permeability measurements in the composite membrane were carried out with thickness direction. The oxygen permeability of the composite membrane containing nanofibers decreased with an increase in the amount of nanofibers. This may be because the gas permeability decreased due to the suppressed gas diffusion induced by the aggregated nanofiber structure. It was found that the introduction of nanofibers within the membrane resulted in the control of the gas permeability or the gas crossover due to the sup© 2010 American Chemical Society
TABLE 3. Oxygen Gas Permeability Coefficient (PO2), Diffusion Coefficient (DO2), and Solubility Coefficient (SO2) of Composite Membranes Containing Aligned Nanofibers weight ratio membrane
membrane-nanofiber
PO2a
DO2b
SO2c
membrane composite membrane composite membrane composite membrane nafion
100:0 99:1 95:5 90:10
2.2 2.1 1.9 1.7 1.1
0.96 0.94 0.93 0.91 4.6
2.2 2.3 2.1 1.8 0.23
c
a PO2: 10-10 (cm3(STP) cm/(cm2 sec cmHg)). SO2: 10-2 (cm3(STP)/(cm3 sec cmHg)).
b
DO2: 10-8 (cm2/sec).
pressed gas diffusion. Although in this study measurements of the oxygen permeability for the composite membranes were carried out in the dry state, we consider that the hydrogen permeability will also decrease with the amount of nanofibers in the same manner. Finally, we show the preliminary results of typical polarization curves and power density curves of fuel cells with the composite membrane containing 10 wt % aligned nanofibers and membrane without nanofiber under fully humidified conditions at 80 °C (Figure 3). These measurements of fuel cells were carried out with thickness direction. It was found that the fuel cell performance of the composite membrane was superior to that in the membrane without nanofiber. Many factors, such as the proton conductivity, gas
FIGURE 3. Cell voltage (close plots) and power density (open plots) of composite membrane containing 10 wt % aligned nanofibers and membrane without nanofiber under fullyhumidified conditions at 80 °C. (b, O), composite membrane containing 10 wt % alignednanofibers; (2,4), membrane without nanofiber. 1327
DOI: 10.1021/nl1007079 | Nano Lett. 2010, 10, 1324-–1328
permeability, and catalyst, have influence on the fuel cell performances, therefore, future research will elucidate the above factors. In summary, the composite membrane containing uniaxially aligned sulfonated polyimide nanofibers strongly suggest that they are able to achieve a high proton conductivity, low gas permeability into the fuel and good chemical and thermal stabilities at the same time. Such membrane may prove to be promising as a polymer electrolyte membrane and may be potentially useful for application in fuel cells. However, the measurement of proton conductivity for the composite membrane was conducted only in the direction of membrane surface that could be precisely analyzed, the proton conductivity measured in the direction of membrane thickness is needed for actual electrolyte membranes. Therefore, at first we would like to establish the measurement method that we can precisely estimate the proton conductivity in the direction of membrane thickness in a future study. In addition, we will fabricate the composite membrane containing uniaxially nanofibers aligned in the direction of membrane thickness.
(3)
Acknowledgment. We would like to thank Mr. S. Matsuno and Mr. H. Kuromatsu of Kaneka Corp. for help with TEM and fuel cell measurements. This work was supported by a grant from New Energy and Industrial Technology Development Organization.
(18)
(4) (5) (6) (7) (8) (9)
(10) (11) (12) (13) (14) (15) (16) (17)
(19) (20) (21) (22)
Supporting Information Available. Materials, methods, preparation of composite membrane with uniaxially aligned nanofiber, ionic-exchange capacity (IEC) and water uptake, chemical stability and proton conductivity, gas permeability, fuel cell performance, additional references, and figures of the preparation of aligned nanofiber with proton conductivity, solubility of membrane without nanofiber and nanofibers. This material is available free of charge via the Internet at http://pubs.acs.org.
(23) (24) (25) (26) (27) (28)
REFERENCES AND NOTES (1) (2)
(29)
Ateele, B. C. H.; Heinzel, A. Nature 2001, 41, 345. Service, R. F. Science 2004, 303, 29.
© 2010 American Chemical Society
1328
Verbrugge, M. W.; Hill, R. F. J. Electrochem. Soc. 1990, 137, 3770– 3777. Rohr, K. S.; Chen, Q. Nat. Mater. 2008, 7, 75–83. Fang, J.; Guo, X.; Harada, S.; Watari, T.; Tanaka, K.; Kita, K.; Okamoto, K. Macromolecules 2002, 35, 9022–9028. Chikashige, Y.; Miyatake, K.; Watanabe, M. Macromolecules 2003, 36, 9691–9693. Nakano, T.; Nagaoka, S.; Kawakami, H. Polym. Adv. Technol. 2005, 16, 753–757. Jouanneau, J.; Mercier, R.; Gonon, L.; Gebel, G. Macromolecules 2007, 40, 983–990. Park, M. J.; Downing, K. H.; Jackson, A.; Gomez, E. D.; Minor, A. M.; Cookson, D.; Weber, A. Z.; Balsara, N. P. Nano Lett. 2007, 7, 3547–3552. Goto, K.; Rozhanskii, I.; Yamakawa, Y.; Otsuki, T.; Naito, Y. Polym. J. 2008, 41, 95–104. Drew, C.; Liu, X.; Ziegler, D.; Wang, X.; Bruno, F. F.; Whitten, J.; Samuelson, L. A.; Kumar, J. Nano Lett. 2003, 3, 143–147. Jiaxing, H.; Kaner, R. B. Nat. Mater. 2004, 3, 783–786. Dzenis, Y. Science 2004, 305, 1917–1919. Patel, A. C.; Li, S.; Yuan, J. M.; Wei, Y. Nano Lett. 2006, 6, 1042– 1046. Formo, E.; Lee, E.; Campbell, D.; Xia, Y. Nano Lett. 2008, 8, 668– 672. Na, H.; Li, X.; Hao, X.; Xu, D.; Zhang.; Zhong, S.; Wang, D. J. Membr. Sci. 2006, 281, 1–6. Elabd, Y. A.; Snyder, J. D.; Chen, H. Macromolecules 2008, 41, 128–135. Wycisk, R.; Pintauro, P. N.; Mather, P. T.; Choi, J.; Lee, K. M. Macromolecules 2008, 41, 4569–4572. Karube, Y.; Kawakami, H. Polym. Adv. Tech., in press. Li, D.; Wang, Y.; Xia, Y. Nano Lett. 2003, 3, 1167–1171. Salalha, W.; Dror, Y.; Khalfin, R. L.; Cohen, Y.; Yarin, A. L.; Zussman, E. Langmuir 2004, 20, 9852–9855. Demir, M. M.; Ozen, B.; Ozc¸elik, S. J. Phys. Chem. B 2009, 113, 11568–11573. Rabolt, J. F.; Kakade, M. V.; Givens, S.; Gardner, K.; Lee, K. H.; Chase, D. B. J. Am. Chem. Soc. 2007, 129, 2777–2782. Mercier, R.; Genies, C.; Sillion, B.; Cornet, N.; Gebel, G.; Pineri, M. Polymer 2001, 42, 359–373. McGrath, J. E.; Wng, F.; Hickner, M.; Kim, Y. S.; Zawodzinski, T. A. J. Membr. Sci. 2002, 197, 231–242. Savard, O.; Peckham, T. J.; Yang, Y.; Holdcroft, S. Polymer 2008, 49, 4949–4959. Wu, H. L.; Ma, C. C. M.; Li, C. H.; Lee, T. M.; Chen, C. Y.; Chiang, C. L.; Wu, C. J. Membr. Sci. 2006, 280, 501–508. Kusumocahyo, S. P.; Sano, K.; Sudoh, M.; Mizoguchi, K. Sep. Purif. Technol. 2000, 18, 141–150. Shin, K. H.; Kim, D. S.; Park, H. B.; Lee, Y. M. Macromol. Res 2004, 12, 413–421.
DOI: 10.1021/nl1007079 | Nano Lett. 2010, 10, 1324-–1328
