Electrochemistry Communications 11 (2009) 2324–2327
Contents lists available at ScienceDirect
Electrochemistry Communications journal homepage: www.elsevier.com/locate/elecom
Anhydrous proton conducting membranes based on electron-deficient nanoparticles/PBI-OO/PFSA composites for high-temperature PEMFC Jin Hu a, Jiangshui Luo a,b, Peter Wagner a, Olaf Conrad a,*, Carsten Agert a a b
EWE-Forschungszentrum für Energietechnologie (Next-Energy), D-26129 Oldenburg, Germany Institut für Chemie, Universität Oldenburg, D-26129 Oldenburg, Germany
a r t i c l e
i n f o
Article history: Received 24 August 2009 Accepted 14 October 2009 Available online 20 October 2009 Keywords: Polymer electrolyte membrane fuel cell (PEMFC) Polybenzimidazole (PBI-OO) High temperature PFSA (Nafion) Electron-deficient compounds
a b s t r a c t PEMFC operating at high temperature (100–200 °C) are expected to have significant advantages but face big challenges in the development of suitable proton exchange membranes. This communication describes novel PBI-OO/PFSA blend membranes, which facilitate proton conduction under anhydrous conditions based on a ‘‘proton donor–proton acceptor” concept. The proton conductivity of the blends under anhydrous conditions exceeded that of PFSA by a factor of 50 at ambient temperature and of 2– 4 at elevated temperature. Intermolecular interaction between two polymer components was investigated by FT-IR spectroscopy. After incorporation of inorganic electron-deficient compounds (BN nanoparticles), the anhydrous proton conductivity of the composites was higher than that of the bare PFSA by three orders of magnitude at ambient temperature and more than one order of magnitude at 140 °C. Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction Polymer electrolyte membrane fuel cells (PEMFC) operating at high temperature (100–200 °C) will yield significant benefits such as faster electrode reaction kinetics, higher CO tolerance, easier water and heat management and higher energy efficiency [1–3]. Thus, the availability of proton conducting membranes retaining satisfactory conductivity at high temperature is considered a desirable pathway to efficient PEMFC in automotive and residential applications. The state-of-the-art PEMFC technology is based on perfluoroÒ sulfonic acid (PFSA) polymer membranes (e.g., Nafion ) [4]. However, Nafion exhibits low conductivity and therefore poor performance under low humidification and at elevated temperatures (above 100 °C), because of the water loss. Considerable efforts have been made to modify PFSA membranes for high-temperature operation, including impregnating the membranes with hygroscopic oxide particles and solid inorganic proton conductors [1,5,6]. However, in many cases, these strategies lead to insufficient mechanical behavior of the composite membranes because of the high filler content (e.g., 3–54 wt.%) required to impart acceptable performance. Since the poor proton conductivity of Nafion at high temperature is due to the loss of water, an alternative approach could blend PFSA with insoluble and nonvolatile compounds, which can act as proton solvents (proton acceptor) similar to water. * Corresponding author. Tel.: +49 441 99906 310. E-mail address:
[email protected] (O. Conrad). 1388-2481/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.elecom.2009.10.020
Kreuer et al. proposed imidazole (Im) as a very effective proton solvent, containing both proton donor (NH) and proton acceptor (N) groups [7,8]. However, besides its high vapor pressure at elevated temperature, Im is soluble in water and is an effective adsorbent on platinum surfaces [9]. Although unsuitable for practical fuel cells application, this work has broken ground for a new research direction towards anhydrous proton conduction. Recent progress was achieved for the Im-based proton solvents towards a full immobilization of the Im-cycle into polymer backbones [10,11]. However, the disadvantages of relative low thermal-oxidative stability and high water/methanol solubility still impair the usability of these Im-grafted polymers as practical proton solvents for high-temperature polymer electrolyte membrane [12]. In this communication, a new water/methanol insoluble and thermo-stable polymer, poly-[(1-(4,40 -diphenylether)-5-oxybenzimidazole)-benzimidazole] (PBI-OO), which contains benzimidazole (BzIm) moieties in the main chain, was investigated as solid proton solvent for high-temperature PEMFC application. The possible proton conduction mechanism relies on a ‘‘proton donor–proton acceptor” concept is shown in Scheme 1. The proton conduction behaviors of the Nafion/PBI-OO blends under anhydrous states were investigated. In addition, the anhydrous conductivity was greatly improved by the strategy of compositing inorganic electron-deficient compounds (e.g., B2O3, H3BO3 and BN). To the best of our knowledge, this is the first time that electron-deficient compounds (anion receptors) are reported as functional additives for proton conducting membranes.
J. Hu et al. / Electrochemistry Communications 11 (2009) 2324–2327
CF2 CF2
x
CF
CF2
2325
y
CF2
Nafion
CFCF3 O CF2
H+
PBI-OO
CF2SO3H
N
N
O
O
N
N
H+
H+
n
Scheme 1. The proposed proton conducting mechanism of PBI-OO and Nafion based on a ‘‘proton donor–proton acceptor” concept.
2. Experimental 2.1. Membrane preparation Nafion solution was purchased from Sigma–Aldrich, PBI-OO from FuMA-Tech GmbH, and nano-sized BN (average particle size 137 nm) from IoLiTec GmbH. Nafion/PBI blend membranes were prepared by solvent-casting method with the desired weight ratio of Nafion and PBI in DMF solution. For the composite with inorganic fillers, the BN/DMF suspension was added dropwise into the Nafion/PBI-OO/DMF solution under vigorous stirring. The resulting solutions or suspensions were poured into a Teflon dish with a glass bottom and placed in a vacuum oven at 80 °C until the solvents were evaporated completely. The final treatments followed the procedure described in [13]. A molar ratio of the Im-cycle: SO3H = 0.125 corresponds to 1.7 wt.% of PBI-OO in the blend membrane, 0.25 corresponds to 3.3 wt.% of PBI-OO in the blend membrane and 0.375 corresponds to 4.8 wt.% of PBI-OO, respectively. 2.2. Characterizations Proton conductivity data of the membranes was obtained from impedance spectra, which were collected on a Solartron 1255B/ 1287A FRA/Electrochemistry workstation in the frequency range of 1–106 Hz with a perturbation of 10 mV. The PTFE test cell has a ‘‘sandwich” structure with the membrane clamped between two stainless steel electrodes. The impedance measurements were performed under fully anhydrous conditions from ambient temperature to 140 °C in an oil bath. FT-IR spectra were collected on a Bruker Vertex-70 FT-IR Spectrometer using the attenuated total reflection (ATR) technique. Before proton conductivity measurement and FT-IR test, samples were rested at 90 °C under high vacuum for one night. 3. Results and discussion The temperature dependence of the proton conductivity for recast Nafion, PBI-OO and the Nafion/PBI-OO blends in the temperature range from 25 to 140 °C are shown in Fig. 1. The negligible proton conductivity of recast Nafion at anhydrous states was comparable to the values reported in literature [14]. After blending with PBI-OO, the anhydrous proton conductivity of the Nafion – 1.7 wt.% PBI-OO blend at ambient temperature was 2.5 106 S cm1, exceeding that of Nafion by a factor of about 50. At elevated temperature (100–140 °C), the conductivities ranged from 7.3 106 S cm1 to 3.2 105 S cm1, which is about 2–4 times
Fig. 1. Arrhenius plots of proton conductivity as a function of temperature for recast Nafion, PBI-OO and Nafion/PBI-OO blends under anhydrous states.
higher than that of Nafion. The improved proton conductivity may indicate that PBI-OO functions as a solid proton solvent in the blends involving a Grotthuss-type mechanism to facilitate proton conduction. [15] The basic nitrogen in the BzIm-moiety acts as a proton acceptor (proton solvent), while the sulfonic acid group in Nafion is a proton donor (proton source). The PBI-OO thus enables partial deprotonation of Nafion and provides carrier mobility based on a ‘‘proton donor–proton acceptor” concept (as shown in Scheme 1). For effective transport of charge through this mechanism the formation of a polymer alloy is desirable. Alternatively, maximizing the interfacial area of segregated polymer phases might be required. We believe that the breakdown in proton conductivity back to the level of pure Nafion at a PBI-OO content of higher than 4 wt.% can be rationalized with the incompatibility of the aromatic and perfluorinated backbones of PBI-OO and Nafion, respectively, leading to the formation of too large phase domains and a collapse of proton conduction pathways. To verify a chemical interaction between Nafion and PBI-OO, FT-IR absorption spectra of recast Nafion, pure PBI-OO, and Nafion/PBI-OO blend membranes at fully anhydrous states were recorded (Fig. 2a). Of particular interest are the vibration modes of the sulfonate moiety and the BzIm ring [16,17]. In pure Nafion, vibration bands associated with the –SO3H groups were observed with maxima at 1056 cm1 (symmetric stretching) and 1305 cm1 (asymmetric stretching). In the blend membranes, the symmetric stretching is red-shifted to 1054 cm1 and the asymmetric stretching blue-shifted to 1320 cm1, which is in good agreement with observations reported for the Nafion/poly(vinyl imidazole) blend system. [18] Compared to absorption bands related to the BzIm-cycle in pure PBI-OO, higher absorption frequencies were observed for the corresponding bands in the blend membranes. For example, the maximum at 1626 cm1 corresponding to C@N stretching vibration shifted to 1634 cm1 and the peak at 833 cm1 corresponding to BzIm ring stretching shifted to 839 cm1. These shifts can be attributed to a weakened conjugation effect as a result of protonation of the imino nitrogen [19,20]. These results indicate that an interaction between the sulfonic acid groups and the imino nitrogen atoms of the BzIm-moiety occurs through proton exchange reaction. If PBI-OO acts as a proton solvent in a blend membrane under anhydrous conditions, carrier mobility might be further enhanced by providing for an effective separation of charges, in analogy to
2326
J. Hu et al. / Electrochemistry Communications 11 (2009) 2324–2327
Fig. 3. Arrhenius plots of proton conductivity as a function of temperature for recast Nafion and Nafion/PBI-OO/nano-BN composites under anhydrous states.
Fig. 2. (a) FT-IR spectra of Nafion, PBI-OO and Nafion/PBI-OO blends at anhydrous states. (b) FT-IR spectra of Nafion, PBI-OO and Nafion/PBI-OO blends after compositing with nano-BN at anhydrous states.
the hydration sphere in hydrated Nafion. Then, as a consequence, more ‘‘free” protons would be conducted through the BzIm ring. Since the –SO 3 group is an electron-rich group and can act as a Lewis base, we attempted to increase the overall proton conductivity of our blend membranes through compositing with inorganic electron-deficient compounds (e.g., B2O3, H3BO3 and BN) that can act as Lewis acids. Fig. 3 shows the anhydrous proton conductivity of pure Nafion and Nafion/PBI-OO blends after compositing with BN nanoparticles. The conductivity of both Nafion and Nafion/PBI-OO blends was greatly improved after compositing with nano-sized BN. The proton conductivity of Nafion – 1.7 wt.% PBI-OO – 0.5 wt.% nano-BN composite membranes was 4.6 105 S cm1 at ambient temperature and 2.2 104 S cm1 at 140 °C. This represents an increase over bare Nafion of more than three orders of magnitude at 25 °C and one order of magnitude at 140 °C, respectively. The interaction of BN with the sulfonate group in Nafion can be proved by the red-shift of the symmetric stretching of –SO 3 (from 1056 cm1 to 1053 cm1) and blue-shift of the asymmetric stretch1 to 1310 cm1) in Nafion/BN composing of –SO 3 (from 1305 cm ites in Fig. 2b. Interestingly, compared with Fig. 2a, the relative intensity of those peaks characteristic of PBI-OO in the composites became very weak after the addition of nano-BN. This may indicate
that the compatibility between Nafion and PBI-OO was improved, resulting in reduced phase segregation in the blend membrane. Based on these results we attribute the enhanced anhydrous proton conductivity of the Nafion/PBI-OO/nano-BN composites to improved charge separation through coordination of the sulfonate group with the Lewis-acid BN and an improved compatibility between Nafion and PBI-OO. In addition, the lower loading of 0.1 wt.% inorganic filler content was effective and increasing this level to 0.5 wt.% did not result in an enhancement of anhydrous proton conductivity. Compared to the high compositing level of the inorganic proton conductors required for a similar enhancement in proton conductivity, such low levels of nano-BN are unlikely to impact the mechanical property of Nafion. Further study of this promising new membrane concept including a full characterization and the evaluation in a PEMFC single cell under anhydrous conditions as well as low relative humidity is currently underway. 4. Conclusion PBI-OO/PFSA blend membranes were prepared by solutioncasting from clear solutions of Nafion and PBI-OO. In the Nafion/ PBI-OO blends, Nafion acted as proton donor while PBI-OO worked as proton acceptor and facilitated proton conduction under anhydrous conditions through the nitrogen atoms of BzIm ring based on a ‘‘proton donor–proton acceptor” concept. The proton conductivity of the blends under anhydrous conditions exceeded that of Nafion by a factor of 50 at ambient temperature and of 2–4 at elevated temperature. FT-IR absorption spectroscopy proved the existence of intermolecular interaction between Nafion and PBI-OO that were interpreted to be proton exchange reactions. The anhydrous proton conductivity of the blends was greatly improved by compositing with inorganic electron-deficient compounds. Incorporation with low levels of BN nanoparticles improved the anhydrous proton conductivity of the composites compared to bare Nafion by between one and three orders of magnitude at 140 °C and ambient temperature, respectively. The enhanced anhydrous proton conductivity of the Nafion/PBI-OO/nano-BN composites is attributed to the anion-receptor effect of BN and resultant better compatibility that might result in an effective charge separation
J. Hu et al. / Electrochemistry Communications 11 (2009) 2324–2327
and conduction akin to the hydration sphere in hydrated Nafion. This work demonstrates a new concept to facilitate proton conduction in composite membranes under conditions typical of PEMFC operating under anhydrous conditions and at intermediate temperature. References [1] [2] [3] [4] [5]
Q.F. Li, R.H. He, J.O. Jensen, N.J. Bjerrum, Chem. Mater. 15 (2003) 4896. G. Alberti, M. Casciola, R. Palombari, J. Membr. Sci. 172 (2000) 233. M.-S. Kang, M.-J. Lee, Electrochem. Commun. 11 (2009) 457. J.A. Kerres, J. Membr. Sci. 185 (2001) 3. P.L. Antonucci, A.S. Aricò, P. Cretì, E. Ramunni, V. Antonucci, Solid State Ionics 125 (1999) 431. [6] F. Bauer, M. Willert-Porada, J. Membr. Sci. 233 (2004) 141. [7] K.D. Kreuer, A. Fuchs, M. Ise, M. Spaeth, J. Maier, Electrochim. Acta 43 (1998) 1281. [8] K.D. Kreuer, Solid State Ionics: Science and Technology, World Science Press, Singapore, 1998.
2327
[9] C. Yang, P. Costamagna, S. Sirinivasan, J. Benziger, A.B. Bocarsly, J. Power Sources 103 (2001) 1. [10] M. Schuster, W.H. Meyer, G. Wegner, H.G. Herz, M. Ise, M. Schuster, K.D. Kreuer, J. Maier, Solid State Ionics 145 (2001) 85. [11] M.F.H. Schuster, W.H. Meyer, M. Schuster, K.D. Kreuer, Chem. Mater. 16 (2004) 329. [12] H. Steininger, M. Schuster, K.D. Kreuer, A. Kaltbeitzel, B. Bingol, W.H. Meyer, S. Schauff, G. Brunklaus, J. Maier, H.W. Spiess, Phys. Chem. Chem. Phys. 9 (2007) 1764. [13] J. Hu, V. Baglio, V. Tricoli, A.S. Aricò, V. Antounci, J. Appl. Electrochem. 38 (2008) 543. [14] J.D. Kim, T. Mori, S. Hayashi, I. Honma, J. Electrochem. Soc. 54 (2007) A290. [15] K.D. Kreuer, Solid State Ionics 94 (1997) 55. [16] K. Je-Deok, M. Toshiyuki, H. Shigenobu, H. Itsru, J. Electrochem. Soc. 154 (2007) A290. [17] P. Musto, F.E. Karasz, W.J. Mackight, Polymer 34 (1993) 2934. [18] B. Byungchan, H.H. Yong, K. Dukjoon, J. Electrochem. Soc. 152 (2005) A1366. [19] R. Bouchet, E. Siebert, Solid State Ionics 118 (1999) 287. [20] Z. Haiqiu, L. Xiangfeng, Z. Chengji, F. Tiezhu, S. Yuhua, N. Hui, J. Membr. Sci. 233 (2007) 39.
