Highly stable self-crosslinked anion conductive ionomers for fuel cell applications

Abstract

We report a simple method for the preparation of self-crosslinked anion conductive ionomers (SCL-ACIs) via polycondensation–Friedel–Crafts alkylation reactions without using any catalyst or crosslinking agent. The reported alkaline stable and methanol impervious SCL-ACIs with 2.45 mequiv. g⁻¹ ion-exchange capacity (IEC) and 68 mS cm⁻¹ hydroxide ion conductivity were assessed as suitable candidates for alkaline direct methanol fuel cell (ADMFC) applications.

---

Fuel cells are green and efficient power sources for stationary, mobile, and automotive applications.^1,2 Recently, ADMFCs have showed potential advantages over direct methanol fuel cells (DMFCs), because the former avoids the use of noble metal catalysts as well as catalyst poisoning, exhibits better reaction kinetics (fuel oxidation and oxygen reduction) in alkaline media and less fuel loss results in depletion of cell cost.^3–5 ACIs based on polystyrene containing quaternary ammonium hydroxide are commercially available but they are unstable in alkaline media due to membrane degradation happening by the nucleophilic (OH⁻) attack on the quaternary ammonium group.^6–8 Several other types of ACIs based on poly(phenylene oxide), poly(ether-imide), or poly(arylene ether)s have been reported in the literature.^9–15

Poly(arylene ether)s based ACIs were typically prepared by chloromethylation reaction using a carcinogenic chloromethyl methyl ether (CMME) reagent.^16–18 In this process, fine control over degree and position of functionalization is also quite difficult. Recently, Hickner et al. studied chemical stabilities of quaternary-ammonium groups containing anion exchange polymers by quantitative ¹H NMR analysis.¹⁹ Therefore, developing of highly conductive and chemical resistant ACIs in an eco-friendly manner is a challenging issue.^20,21

Research interest was rendered for developing self-crosslinked ionomers with stability and reasonable conductivity.^22–25 Zhuang et al. reported self-crosslinked tertiary quaternary ammonium poly sulfone based alkaline polymer electrolyte.²³ In spite of good dimensional stability, these membranes are sensitive for Hofmann elimination (due to presence of β-hydrogen atoms on crosslinked quaternary ammonium groups) and further deteriorates membrane conductivity. Further, Hui Na et al. also reported self-crosslinked anion-exchange membrane (AEM) by introducing bromine atoms onto benzylic positions of propenyl-containing poly(ether sulfone), which exhibits activity to undergo Friedel–Crafts electrophilic substitution C-alkylation reaction with aromatic benzene ring under heat treatment.^24,25 In this process, a fine temperature control is necessary otherwise crosslinking of all benzyl bromide groups may be possible and thus results low hydroxide conductivity (13.4–18.8 mS cm⁻¹ at 20 °C) due to less availability of benzyl bromide groups for quaternization. For improved hydroxide conductivity and stability, it is also interesting for architectural tailoring of self-crosslinked ionomers with hydrophilic and hydrophobic phase separation, in which hydrophilic block enhances conductivity and stability by hydrophobic block via in situ self-crosslinking reaction. The hydrophobic blocks were prepared by nucleophilic substitution polycondensation of bromo-substituted allyl bisphenol and 4,4′-difluorophenyl sulfone, while hydrophilic blocks by 2,2′,3,3′-tetrakis(chloromethyl)bisphenol (TCMBP) and 4,4′-difluorophenyl sulfone monomers.

We prepared highly stable SCL-ACIs using bromo substituted 2,2′-diallyl bisphenol (ABP-Br) and TCMBP monomers and in situ self-crosslinking was achieved during polymerization process. Since, polycondensation reaction has been carried out at below 200 °C, bromo-substituted benzyl groups easily undergo self-crosslinking reaction due to best leaving nature of bromide nucleophile²⁶ but not the chloromethyl groups.^27,28 The ACI membranes were obtained by dissolving alkaline polymer (quaternary ammonium hydroxide containing ionomers) in NMP under thermal treatment may be due to presence of more hydrophilic moieties in the polymer matrix generated by quaternization and alkalization of un-crosslinked chloromethyl groups.

Reported SCL-ACIs contain two vicinal quaternary ammonium groups and avoid degradations such as Hofmann elimination and Sommelet–Hauser rearrangement. Vicinal ammonium groups also resist nucleophilic (OH⁻) attack; further suppress the Stevens rearrangement as well as S_N2 substitution reaction due to steric hindrance. ABP-Br monomer was prepared in two steps; synthesis of 2,2′-diallyl bisphenol (ABP) monomer and followed by treatment with N-bromosuccinamide (NBS). The TCMBP was synthesized by in situ pre-chloromethylation of bisphenol A (BPA). BPA monomer was treated with potassium carbonate to achieve electrophilic-substitution reaction via formation of phenoxides, which acts as an electron donating groups. Further, reaction of bisphenol phenoxides with formaldehyde and thionyl chloride (SOCl₂) gives TCMBP monomer. The detailed procedure for synthesis of ABP, ABP-Br and TCMBP has been discussed in ESI.† Prepared monomers were characterized by ¹H NMR spectrum (Fig. S1(a)–(c) in ESI†).

Fig. S1(a)† ABP: δ 1.77 (s, 6H, –C(CH₃)₂), 3.62 (d, 4H, –CH₂), 4.94–5.14 (dd, 4H, =CH₂), 5.45–5.90 (m, 2H, =CH), 6.71–6.85 (m, 2ArH ortho to –C(CH₃)₂), 7.73 (s, 2ArH ortho to –C(CH₃)₂), 7.12–7.28 (m, 2ArH ortho to –OH). Fig. S1(b)† ABP-Br: δ 1.89 (s, 6H, –C(CH₃)₂), 5.42–5.33 (d, 2H, –CHBr), 5.1–5.27 (dd, 4H, =CH₂), 5.9–6.1 (m, 2H, =CH), 6.7–6.8 (m, 2ArH ortho to –C(CH₃)₂), 8.10 (s, 2ArH ortho to –C(CH₃)₂), 7.04–7.15 (m, 2ArH ortho to –OH). Fig. S1(c)† TCMBP: δ 4.61 (s, 4H, ortho to –C(CH₃)₂), 4.74 (s, 4H, ortho to –OH), 1.53 (s, 6H, –C(CH₃)₂), 7.00 (d, 2ArH, ortho to –C(CH₃)₂), 7.06 (d, 2ArH, ortho to –OH). Because of strong activating –OH group by resonance effect, electrophilic substitution occurred at both ortho positions, and a single –CH₂Cl peak along with aromatic singlet peak were expected. But, in ¹H NMR spectrum of TCMBP, two –CH₂Cl peaks were observed along with a double doublet of two dissimilar aromatic protons (appeared to be merged). These observations confirmed electrophilic substitution at both ortho-positions of –OH and –C(CH₃)₂ groups (caused by the activating nature of –C(CH₃)₂ group by inductive effect).

The degree of chloromethylation (DCM) for TCMBP monomer was estimated by ¹H NMR spectra (Fig. S1(c) in ESI†) by the integral peaks ratio of ‘4’, and ‘5’ (assigned for –CH₂Cl group) to those of peak ‘3’ (assigned for –C(CH₃)₂ group) by following equation.

where A(H₄ + H₅) is the sum of integral area for both H₄ and H₅ peaks (assigned for –CH₂Cl group), and A(H₃) is the area of the H₃ peak (assigned for –C(CH₃)₂ group) (Fig. S1(c)†). TCMBPs with different DCMs (90%, 55% and 35%) were prepared under varied conditions (Table S1, ESI†).

SCL-ACIs was synthesized by a nucleophilic substitution polycondensation of ABP-Br, 4,4′-difluorophenyl sulfone, TCMBP monomers in the presence of potassium carbonate in dry DMAc followed by amination with trimethylamine solution (35%) for 12 h at room temperature. The detailed membrane preparation procedure (Scheme 1) has been discussed in ESI.† Since the reaction temperature was maintained around 140–170 °C, conversion of terminal double bond into internal double bond (isomerization) and crosslinking of vinyl bromides was achieved simultaneously. In this case, the self-crosslinking (C-alkylation) was not achieved by chloromethyl groups due to the presence of two vicinal –CH₂Cl groups on aromatic rings. This would not allow for the formation of stable carbocation which can easily undergoes C-alkylation with aromatic ring of another molecule in the copolymer. On the other hand, alkenyl bromide groups presented in copolymer easily undergoes Friedel–Crafts C-alkylation due to the best leaving nature of bromide nucleophile.^24–26 The prepared self-crosslinked ionomers were designated as SCL ACI-X where X is the DCM (90%, 55%, and 35%) corresponding to the 2.45, 1.75 and 1.46 meq. g⁻¹ ion exchange capacity (IEC) as estimated from ¹H NMR spectra (Fig. S2, ESI†).

Scheme 1 Reaction scheme for SCL-ACIs preparation.

The important membrane properties to assess their suitability for fuel cell applications such as water uptake (WU) and ion conductivity (σ) were measured by previously reported method and are summarized in Fig. 1(a).²⁹ The SCL-ACI-90 (2.45 meq. g⁻¹ IEC) exhibited 46% water uptake 10.2% swelling ratio and 68 mS cm⁻¹ hydroxide conductivity. This clearly demonstrated the dependency of water uptake and hydroxide conductivities of developed ACIs on IEC (Fig. 1(a)). Relatively low WU values were attributed to the crosslinked hydrophobic segments, which inhibited excessive membrane swelling. While presence of more number of hydrophilic segments were contributed for improved IEC/conductivity. The hydroxide conductivity of prepared SCL-ACI-90% membrane exhibited comparatively high conductivity than other crosslinked membranes reported in the literature (Table S2, ESI†).

Fig. 1 (a) Variation of hydroxide ion conductivity (σ), and water uptake (%) with IEC (meq. g⁻¹) at 30 °C: (circle) water uptake; (square) conductivity; (b) temperature dependence of hydroxide ion conductivity for SCL-ACIs with different IECs.

Hydroxide ion conductivity for prepared SCL-ACIs was varied between 68–22 mS cm⁻¹ corresponding to 2.45–1.46 meq. g⁻¹ IEC values. High hydroxide conductivity for self-crosslinked membranes was observed may be due to the formation of ion conducting channels developed by hydrophilic–hydrophobic phase separation. Further, hydroxide ion conductivity for SCL-ACIs followed Arrhenius-type temperature dependent behaviour (activation energy: 6.93–16.97 kJ mol⁻¹) with respect to IEC due to formation of inter-linked hydrophilic channels leads to fast ion conduction (Fig. 1(b)). Mechanical properties of developed SCL-ACIs were also measured (Table S4, ESI†).

The alkaline stability of SCL-ACIs was investigated by alkali treatment (6 M NaOH) at 60 °C for 300 h and percentage loss in IEC, hydroxide conductivity was measured (Table S3 in ESI†). The developed SCL-ACIs did not exhibit any degradations such as Sommelet–Hauser rearrangement (due to absent of ortho hydrogen adjacent to quaternary ammonium groups), Hofmann elimination (owing to absent of β-hydrogen). Further, the presence of two bulky vicinal quaternary ammonium groups suppresses the effective attack of OH⁻ nucleophile due to steric hindrance in spite of having more α-H atoms and attached hydroxide ions. So, the less deterioration in hydroxide conductivity was happened by both of Stevens rearrangement and nucleophilic substitution (OH⁻). Only less than 10% deterioration in conductivity and IEC was showed by SCL-ACIs in spite of having more quaternary ammonium groups after 300 hours under strong alkaline treatment confirmed their durability for ADMFC application.

Selectivity parameter (ratio of hydroxide conductivity and methanol permeability) plays a key role during ADMFCs performance assessment.³⁰ In spite of more number of quaternary ammonium groups, reported SCL-ACIs showed quite low methanol permeability (2.31–4.57 × 10⁻⁸ cm² s⁻¹) at 60 °C in 2 M MeOH and proportional with respect to DCM (Fig. 2). Self-crosslinked hydrophobic segments in the SCL-ACIs may be responsible for methanol impervious nature. Further, about (16.88–18.59) × 10⁵ S cm⁻³ s selectivity values at 60 °C for these SCL-ACIs were obtained due to their high conductivity showed by four quaternary ammonium groups (Fig. 2), and confirmed their potential application for ADMFC.

Fig. 2 Selectivity of the membranes at 60 °C.

In summary, SCL-ACIs were successfully prepared via polycondensation of TCMBP, ABP-Br and FPS followed by amination, without use of any crosslinking agent. Due to the presence of crosslinked hydrophobic and highly functionalized hydrophilic phase separation, these SCL-ACIs showed high hydroxide conductivity along with stability. The self-crosslinked membranes exhibited high alkaline stability and excellent methanol tolerance. These properties of the prepared SCL-ACIs, may improve fuel cells efficiency and their lifetimes. Additionally, the self-crosslinking of multiblock copolymer may be equally useful as ion conducting polymer membranes for diversified applications.

Acknowledgements

CSIR-CSMCRI Communication no. 065/2014. Financial assistance received from the Ministry of New and Renewable Energy (MNRE), Government of India (project no. 102/79/2010-NT) is gratefully acknowledged. Authors and thankful for Analytical Science Division, CSIR-CSMCRI for instrumental support.

Notes and references

1. W. Vielstich, Handbook of Fuel Cells, Wiley, Chichester, England, 2009.
2. T. J. Peckham and S. Holdcroft, Adv. Mater., 2010, 22, 4667–4690.
3. G. Couture, A. Alaaeddine, F. Boschet and B. Ameduri, Prog. Polym. Sci., 2011, 36, 1521–1557.
4. J. Sanabria-Chinchilla, K. Asazawa, T. Sakamoto, K. Yamada, T. Tanaka and P. Strasser, J. Am. Chem. Soc., 2011, 133, 5425–5431.
5. G. Merle, M. Wessling and K. Nijmeijer, J. Membr. Sci., 2011, 377, 1–35.
6. K. Kneifel and K. Hattenbach, Desalination, 1980, 34, 77–95.
7. A. A. Zagorodni, D. L. Kotova and V. F. Selemenev, React. Funct. Polym., 2002, 53, 157–171.
8. V. Neagu, I. Bunia and I. Plesca, Polym. Degrad. Stab., 2000, 70, 463–468.
9. L. Wu, T. Xu, D. Wu and X. Zheng, J. Membr. Sci., 2008, 310, 577–585.
10. G. Wang, Y. Weng, D. Chu, D. Xie and R. Chen, J. Membr. Sci., 2009, 326, 4–8.
11. M. Tanaka, M. Koike, K. Miyatake and M. Watanabe, Macromolecules, 2010, 43, 2657–2659.
12. M. Tanaka, K. Fukasawa, E. Nishino, S. Yamaguchi, K. Yamada, H. Tanaka, B. Bae, K. Miyatake and M. Watanabe, J. Am. Chem. Soc., 2011, 133, 10646–10654.
13. M. Tanaka, M. Koike, K. Miyatake and M. Watanabe, Polym. Chem., 2011, 2, 99–106.
14. X. Li, Q. Liu, Y. Yu and Y. Meng, J. Mater. Chem. A, 2013, 1, 4324–4335.
15. J. Zhou, M. Unlu, J. A. Vega and P. A. Kohl, J. Power Sources, 2009, 190, 285–292.
16. US Department of Health Human Services, Public Health Service, National Toxicology Program. Report on Carcinogens, Twelfth Edition, 2011, http://ntp.niehs.nih.gov/ntp/roc/twelfth/roc12.pdf.
17. K. Verschueren, Handbook of Environmental Data on Organic Chemicals, 2nd edn, 1983, p. 1310.
18. S. Laskin, M. Kusschner, R. T. Drew, V. P. Capiello and N. Nelson, Arch. Environ. Health, 1971, 23, 135–136.
19. S. A. Nunez and M. A. Hickner, ACS Macro Lett., 2013, 2, 49–52.
20. D. Chen and M. A. Hickner, ACS Appl. Mater. Interfaces, 2012, 4, 5775–5781.
21. C. Wang, N. Li, D. W. Shin, S. Y. Lee, N. R. Kang, Y. M. Lee and M. D. Guiver, Macromolecules, 2011, 44, 7296–7306.
22. S. Gu, R. Cai and Y. Yan, Chem. Commun., 2011, 47, 2856–2858.
23. J. Pan, Y. Li, L. Zhuang and J. Lu, Chem. Commun., 2010, 46, 8597–8599.
24. J. Ni, C. Zhao, G. Zhang, Y. Zhang, J. Wang, W. Ma, Z. Liu and H. Na, Chem. Commun., 2011, 47, 8943–8945.
25. H. Sun, G. Zhang, Z. Liu, N. Zhang, L. Zhang, W. Ma, C. Zhao, D. Qi, G. Li and H. Na, Int. J. Hydrogen Energy, 2012, 37, 9873–9881.
26. R. M. Roberts and M. B. Abdel-Baset, J. Org. Chem., 1976, 41, 1698–1701.
27. E. Avram, M. A. Brebu, A. Warshawsky and C. Vasile, Polym. Degrad. Stab., 2000, 69, 175–181.
28. X. Yan, G. He, S. Gu, X. Wu, L. Du and H. Zhang, J. Membr. Sci., 2011, 375, 204–211.
29. A. Jasti and V. K. Shahi, J. Mater. Chem. A, 2013, 1, 6134–6137.
30. A. Jasti, S. Prakash and V. K. Shahi, J. Membr. Sci., 2013, 428, 470–479.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed experimental procedures are included. See DOI: 10.1039/c4ra02507e

---