Sulfonated multiblock copolynaphthalimides for polymer electrolyte fuel cell application
Graphical abstract
Introduction
In recent years, polymer electrolyte fuel cells (PEFCs) have been drawing great attention as energy sources for transportation, stationary and portable devices because of the environmental friendship and the high fuel efficiency [1]. Polymer electrolyte membrane (PEM) in a PEFC system plays a key role to transport protons from anode to cathode and to separate the fuel and oxidant. Sulfonated aromatic polymers have been extensively studied as alternative materials to the state-of-the-art perfluorosulfonic acid polymers [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44]. However, most of these aromatic polymers showed low proton conductivity at low relative humidities. High performance PEMs are required to have high through-plane proton conductivity under the conditions of low relative humidity (<50 %RH) and high temperature (>80 °C). One of the effective strategies to overcome this drawback is to develop PEMs based on hydrophilic–hydrophobic multiblock copolymers and having proper microphase-separated morphology for PEFC applications [12], [13], [14], [15], [16], [17], [20], [21], [22], [36].
Recently, McGrath et al. developed multiblock copolymers consisting of fully disulfonated poly(arylene ether sulfone) and partially fluorinated poly(arylene ether ketone) as hydrophilic and hydrophobic segments, respectively [14]. The membranes of the copolymer with both block length of 15 showed a clear lamellar morphology with well-connected hydrophilic channels, resulting in high proton conductivity comparable to that of Nafion (NRE211). Miyatake and Watanabe et al. developed sulfonated multiblock copoly(arylene ether sulfone ketone)s with highly sulfonated hydrophilic blocks, of which the membranes had well microphase-separated morphology, resulting in improved proton conductivity at low relative humidities and high temperatures and also high PEFC performance at 30% and 53% RH and 100 °C [20]. It is noted that these block copolymer membranes showed the anisotropic membrane swelling with larger through-plane swelling, whereas the random copolymer membranes showed the isotropic membrane swelling.
Sulfonated random copolynaphtalimides (random co-SPIs) derived from 1,4,5,8-naphthalenetetracarboxylic dianhydride (NTDA) and sulfonated and nonsulfonated diamines have been reported to have the high fuel cell performance and durability for PEFC applications [32], [34], [38], [39], [40], [42], [43]. To further improve the proton conductivity under the low relative humidities, the multiblock co-SPIs were also investigated [45], [46], [47], [48]. In general, the random co-SPIs show the anisotropy in membrane swelling and proton conductivity, namely, in-plane membrane swelling is smaller than through-plane swelling and through-plane proton conductivity is smaller than in-plane conductivity [7], [31], [32], [33], [34], [35], [37], [48], [49]. This is attributed to that the polymer chains tend to align to the surface of a glass plate. Therefore, it is essentially important for evaluation of multiblock co-SPIs to investigate both and together with in-plane and through-plane dimensional change ( and , respectively).
In a previous paper, we reported on the physical properties, morphology and PEFC performance of multiblock co-SPIs derived from NTDA, 2,2′-bis(4-sulfophenoxy)benzidine (BSPOB) and 4,4′-diaminophenyl hexafluoropropane (BAHF) or 4,4′-bis(3-diamino-phenoxy) diphenylsulfone in comparison with those of the corresponding random co-SPIs [48]. The chemical structures of typical random and multiblock co-SPIs are shown in Fig. 1. The membranes of BSPOB-based multiblock co-SPIs with both block length of 20 had well-defined microphase-separated morphology with hydrophilic and hydrophobic layer-like domains oriented in the plane direction. Compared to the random co-SPI membranes having the homogeneous morphology, the multiblock co-SPI membranes exhibited a higher water uptake and the enhanced anisotropy in membrane swelling and proton conductivity, that is, the through-plane swelling was enhanced but the in-plane swelling was rather reduced, whereas the in-plane conductivity was largely enhanced but the through-plane conductivity was largely reduced. As a result, they exhibited a much lower PEFC performance.
Random co-SPIs derived from BSPOB, 2,2′-benzidine disulfonic acid, 2,2′- or 3,3′-bis(3-sulfopropoxy)benzidine and 2,2′-bis(4-sulfophenyl)benzidine as sulfonated diamine are classified into the group having the larger membrane swelling anisotropy [7], [31], [32], [33], [34]. They have the rigid imide backbone from NTDA and benzidine with sulfonated pendants, which cause the better alignment of polymer chain in the plane direction. On the other hand, random co-SPIs derived from other sulfonated diamines such as 4,4′-bis(4-aminophenoxy)biphenyl (BAPBDS), 4,4′-diamino-diphenyl-ether-2,2′-disulfonic acid, bis[4-(4-aminophenoxy)phenyl]-sulfone-3,3′-disulfonic acid and 4,4′-bis(4-aminophenoxy)-3,3′-bis(4-sulfophenyl)biphenyl (BAPSPB) are classified into another group having the smaller or mild swelling anisotropy [7], [31], [35], [37], [49], [50]. In the latter groups, flexible ether bonds make the imide backbone flexible to some extent, resulting in the less alignment of polymer chain.
Random co-SPIs based on BAPSPB bearing sulfophenyl pendant and two ether bonds belong to the latter groups. In a previous paper, we reported that they exhibited the high water stability and PEFC performance comparable to those of BSPOB-based ones [35]. Therefore, it is interesting to study BAPSPB-based multiblock co-SPIs as PEM for PEFC applications.
In this paper, multiblock co-SPIs have been prepared from NTDA, BAPSPB and BAHF and their physical properties, morphology and PEFC performance have been investigated in comparison with those of the random co-SPIs.
Section snippets
Materials
BAPSPB was prepared according to the literature [35]. BAHF, benzoic acid, m-cresol, isoquinoline, methanol, 2-propanol and other reagents were purchased from Wako and used as received. NTDA (Aldrich) was purified by vacuum sublimation before use. Triethylamine (TEA) was purified by distillation under reduced pressure and dehydrated with 4 Å molecular sieves. Ultra-pure water was obtained from a Millipore Milli-Q purification system.
Synthesis of multiblock co-SPIs
Multiblock co-SPIs were synthesized by a two-pot method as
Synthesis and characterization
Multiblock co-SPIs were synthesized by a two-pot method as described in Scheme 1. The anhydride-end-capped hydrophilic oligomer and an amine-end-capped hydrophobic oligomer were synthesized separately in two polymerization systems. These two oligomer solutions were mixed very carefully and kept at 80 °C for 24–48 h and then at 180 °C for 24 h to react adequately and to obtain high molecular weight polymers. The block length (l) of the hydrophilic and hydrophobic segments was set for 10 or 20 by
Conclusions
The BAPSPB-based multiblock co-SPI (BA1) with block length of 20/10 and IEC of 1.67 meq g−1 exhibited larger WU, larger in whole humidity range and larger in water than the randomco-SPI (RA4) with the similar IEC. The multiblock co-SPI (BA2) with the longer block length of 20/20 exhibited the large and comparable to those of BA1, in spite of the smaller IEC of 1.35 meq g−1. Both BA1 and BA2 showed the moderate anisotropy in membrane swelling and proton conductivity
References (50)
- et al.
Prog Polym Sci
(2000) J Membr Sci
(2001)J Membr Sci
(2001)- et al.
Solid State Ionics
(2002) - et al.
Polymer
(2004) - et al.
Polymer
(2006) - et al.
Polymer
(2008) - et al.
Polymer
(2006) - et al.
Polymer
(2007) - et al.
Polymer
(2009)