Design of polybenzimidazolium membranes for use in vanadium redox flow batteries

In recent years, polybenzimidazole (PBI) membranes have been proposed for vanadium redox flow batteries (VRFBs) as an alternative to perfluoroalkylsulfonic acid membranes such as Nafion™. Despite their excellent capacity retention, PBI membranes tend to suffer from a low ionic conductivity. The formation of a polybenzimidazolium through an N-alkylation of the benzimidazole core is shown to improve the ionic conductivity of the membrane, with this class of materials having found uses in alkaline fuel cell and water electrolysis systems. However, much less is known about their incorporation into a VRFB. This article describes the use of hexamethyl-p-terphenyl polybenzimidazolium (HMT-PMBI) membranes for a vanadium redox flow battery, with the membrane characteristics in acidic media being related to their performance in a single-cell VRFB setup. A change of the degree of methylation from 56 to 65, 75, and 89% leads to an increase in ionic conductivity, correlated with an increased fraction of free water in the ionomer. The corresponding increase in cell performance is, however, accompanied by a drop in capacity retention. The membrane with a degree of methylation of 65% shows balanced properties, with a 5% higher efficiency and a two times improved capacity retention compared to Nafion™ NR212 over 200 charge–discharge cycles at 200 mA cm−2.


Polymer characterization
Prior to analysis, the HMT-PMBI variants were characterized by 1 H nuclear magnetic resonance (NMR) spectroscopy for their degree of methylation.The degree of methylation was calculated according to equation S1, with x being the corresponding integration of the uncharged benzimidazole groups (~3.6 ppm) after integrating and setting the area of the charged benzimidazolium groups (~4 ppm) to 12, as described by Wright et al. 1 .

6
) + 50% The 1 H NMR spectra of the HMT-PMBI variants can be seen in Figure S1 to Figure S4.

Membrane preparation
Membranes were prepared from the HMT-PMBI materials as described in section 2.2 'Membrane preparation' of the main text with the casting conditions for a ~25 µm film given in Table S1.

Dimensional swelling and electrolyte uptake
The dimensional swelling and electrolyte uptake of the HMT-PMBI membranes with a varying degree of methylation was measured as described in section 2.4 'Dimensional swelling and electrolyte uptake' of the main text.The electrolyte uptake in terms of weight content and dimensional swelling can be seen in Table S3, with the electrolyte uptake in terms of mols per mol polymer reported in Table S4.

Ionic conductivity
The through-plane ionic conductivity of the HMT-PMBI membranes was measured as described in section 2.5 of the main text, with the results depicted in Table S5.
Table S5.Through-plane ionic conductivity of the different HMT-PMBI variants in 2 M H 2 SO 4 and a 1.6 M vanadium electrolyte (oxidation state 3.5).

Vanadium permeability
As described in section 2.6 'VO 2+ diffusion', the VO 2+ concentration was determined using a VO 2+ calibration curve (Table S5).This calibration curve was prepared by diluting a 1000 mg vanadium standard (VOSO 4 in 8.6% H 2 SO 4 , Titrisol®) with 2 M H 2 SO 4 to a vanadium concentration of 0.01 M, 0.02 M, 0.03 M and 0.04 M. Subsequently, the calibration curve was prepared by analyzing the UV-Vis absorbance of the solutions at 765 nm and plotting these versus their concentration.The VO 2+ diffusion experiments were carried out in a home-built diffusion cell, Figure S6, as described in section 2.6 'VO 2+ diffusion'.Four individual measurements were done for each HMT-PMBI membrane, Figure S7, with each test contributing to the averaged result shown in the main text.The measured diffusion coefficient corresponding to each individual test, including the averaged diffusion coefficient and its corresponding error given as the standard deviation from each individual test can be seen in Table S6.Table S6.The VO 2+ diffusion coefficients of each individual measurement and the corresponding averaged diffusion coefficient.

Chemical Stability
The chemical stability of the HMT-PMBI variants was determined as described in section 2.8 of the main text.The samples in solution at the start and during this test after 25, 55 and 90 days can be seen in Figure S8.In addition to the observed degradation of the HMT-PMBI membranes by VO 2 + as shown in Figure 5, the membranes were analyzed after the chemical stability test by Fourier-transform infrared (FTIR) spectroscopy and 1 H nuclear magnetic resonance (NMR) spectroscopy.FTIR spectra of the oxidized 75dm and 89dm films could not be measured as a result of their mechanical disintegration, with the spectra of the oxidized 56dm and 65dm being depicted in Figure S10 and Figure S11, respectively.Furthermore, the FTIR spectra of the pristine HMT-PMBI films can be seen in Figure S9.The FTIR spectra of the oxidated 56dm and 65dm both show the presence of two new peaks at 1700 cm −1 and 1830 cm −1 that cannot be assigned to the acidic electrolyte, indicating the formation of carbonyl moieties and the chemical degradation of the HMT-PMBI polymer.The signal appearing between 2000 -2700 cm −1 can be assigned to the N + -H group formed by the protonation of the benzimidazole backbone and is therefore not an indication of polymer degradation 2 .
Despite the new FTIR bands observed in the oxidized samples, only a slight peak shift is observed in the 1 H NMR spectra of all oxidized HMT-PMBI polymers, Figure S12, with no newly formed distinct peaks.The lack of new signals is hypothesized to be the result of the water solubility of most degradation products, therefore leaving only the less degraded polymer after the washing step.

Polarization curves at different states of charge
The polarization curves of Nafion™ NR212 and the HMT-PMBI were recorded as described in section 2.9 'VRFB cell cycling' of the main text.The polarization of each membrane at a state of charge (SoC) of 20, 30, 50, 70 and 90% can be seen in Figure S13-S17. -

Figure S8 .
Figure S8.The HMT-PMBI samples during the chemical stability analysis in the 1.6 M VO 2 + electrolyte, with the immersed samples at the start of the test and after day 25, day 55 and day 90.

Figure 1 Figure 1 Figure
Figure S9.FTIR spectra of the pristine HMT-PMBI samples prior to the chemical stability test.

Figure S12. 1 H
Figure S12. 1 H NMR spectra (400 MHz, DMSO-d 6 ) of the HMT-PMBI samples before and after the chemical stability test Figure S18.Cycling performance of Nafion™ NR212 and the HMT-PMBI membranes between 80 and 200 mA•cm −2 with coulombic efficiency (A), voltaic efficiency (B), energy efficiency (C) and the extrapolated energy efficiency at higher current densities (D).

Table S2 .
The detection limits of the pristine and treated 89dm XRF analysis.

Table S4 .
Water uptake, free acid uptake, bound acid uptake and ion-exchange capacity of HMT-PMBI membranes with various degrees of methylation upon immersion in 2 M H 2 SO 4 in terms of mols per mol polymer.
Polarization curves at a state of charge of 20, 30, 50, 70 and 90% of a Nafion™ NR212 membrane with a dry thickness of 51 µm.