Self-crosslinking of comb-shaped polystyrene anion exchange membranes for alkaline fuel cell application

Author’s Accepted Manuscript Self-crosslinking of Comb-shaped Polystyrene Anion Exchange Membranes for Alkaline Fuel Cell Application Wangcai Liu, Lei Liu, Jiayou Liao, Linghui Wang, Nanwen Li www.elsevier.com/locate/memsci

PII: DOI: Reference:

S0376-7388(17)30768-8 http://dx.doi.org/10.1016/j.memsci.2017.05.006 MEMSCI15240

To appear in: Journal of Membrane Science Received date: 17 March 2017 Revised date: 26 April 2017 Accepted date: 2 May 2017 Cite this article as: Wangcai Liu, Lei Liu, Jiayou Liao, Linghui Wang and Nanwen Li, Self-crosslinking of Comb-shaped Polystyrene Anion Exchange Membranes for Alkaline Fuel Cell Application, Journal of Membrane Science, http://dx.doi.org/10.1016/j.memsci.2017.05.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Self-crosslinking of Comb-shaped Polystyrene Anion Exchange Membranes for Alkaline Fuel Cell Application Wangcai Liu1, Lei Liu2, Jiayou Liao2, Linghui Wang1*, Nanwen Li 2* 1

College of Materials Science and Chemical Engineering, Ningbo University of Technology, Ningbo 315211, China.

2

State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan, 030001, China. [email protected] [email protected] *Corresponding authors.

Abstract A series of crosslinked, comb-shaped polystyrene (PS) anion exchange membranes (AEMs) having C-16 alkyl side chain were synthesized by a Cu(I)-catalyzed “click chemistry” reaction, and subsequently azide-assisted self-crosslinking for alkaline fuel cell application. The as-obtained crosslinked AEMs showed lower water uptake, and thus good dimensional stability as a result of crosslinking. The chemical stabilities of the AEMs were evaluated under severe conditions, and minor degradation was observed by measuring IEC and ion conductivity changes after stability testing. The crosslinked membranes retained their high ion conductivity even in 10 M NaOH at 80 °C for 400 h. Moreover, the PS-based AEMs shown higher hydroxide conductivity than that of the AEMs based on poly(2,6-dimethyl-phenylene oxide) (PPO) with similar architecture under the same testing condition probably due to the excellent compatibility between polystyrene backbone and alkyl side chain which may have helped to build a more efficient morphology for water uptake and thus the ion transport. These crosslinked cationic polymers were employed as membranes for alkaline fuel cells. The results indicated that the AEMs based on PPO having C-16 alkyl side chain had not initial performance, e.g. no open 1

circuit voltage (OCV). In contrast, the initial fuel cell performance with the peak power density of 88 mW/cm2 was observed for the alkaline fuel cell with xPS-65 membrane. It is assumed that the polystyrene polymer backbone may have helped to build a more efficient phase boundary between the catalyst layer and membrane probably resulted from the high water uptake though a more complete study would be needed to explain this phenomenon.

Key words: Anion Exchange Membrane; Alkaline Fuel Cells; Comb-shaped Polystyrene; Crosslinking

1. Introduction

Relative to acidic fuel cells, anion exchange membrane fuel cells (AEMFCs) have received significant interest in recent years due to the enhancement of electrode reaction kinetics, using of the non-noble metals or inexpensive metal oxides catalysts.[1-3] As a critical component of AEMFCs, the polymers with tethered organic cations generally have been demonstrated as anion exchange membranes (AEMs). The basic properties of ionic conductivity and chemical stability of AEMs control the performance ant lifetime of AEMFCs. Various polymer backbone structures such as poly(olefin)s,[4,5] polystyrenes,[6-8] poly(phenylene oxide)s,[9,10] poly(phenylene)s,[11] poly(arylene ether)s[12-17] have been chloroalkylated or bromoalkylated and subsequently quaternized by tertiary amine, pentamethylguanidine, or tertiary phosphine to obtain anion conductive copolymers having quaternary ammonium,[10] guandinium,[16] or phosphonium groups. [18]

The ionic conductivity of a AEM is related to a combination of the ion exchange capacity (IEC), it’s the hydration level, and the microphase separated morphology of the membrane.[3] Generally, high IEC values induce high ionic conductivity; however, increasing the IEC is always accompanied by consequent significant increase in water uptake which can lead to significant dimensional swelling, loss

2

of mechanical properties,[19,20] or even disintegration of the AEM, especially at elevated temperatures. Several approaches have been examined to improve anion conductivity, such as changes in the position of quaternary ammonium groups,[21-23] control of membrane morphologies by block,[24] grafting[25] or comb-shaped[10] copolymer architectures, and crosslinking of the high IEC membranes,[26-30] displaying a wide range of conductivities. For example, AEMs aminated from bromoalkyl with long alkyl chains (six carbon atoms or more) have been proven to be highly hydroxide conductive, but higher water uptake was observed than AEMs based on benzyltrimethylamine.[21] Sequenced hydrophilic and hydrophobic groups as block and grafting copolymers were also claimed to be effective for this purpose by formation of nanochannel.[23,24] However, research on this class of copolymers is still in the initial stages, and the precise control of nanoscale morphology of multiblock or grafting copolymers is often restricted due to the polydispersity of each block. In addition to the morphology controlling, the covalent crosslinking has been reported as an effective method to stabilize high IEC AEMs against high water swelling.

For

example,

the

multifunctional

groups

of

dithiol,[27]

dialdehyde,[28]

tri/tetraalkoxysilancs,[29] and tetraepoxy[30] have been widely used for AEM crosslinking. Thermal self-crosslinking and simultaneous polymerization-crosslinking has also been developed to reduce swelling and to prepare high-performance AEMs due to their controllability.[26,31]

Another critical challenge for practical application of AEMs in alkaline fuel cells is the chemical stability of AEMs under alkaline conditions.[3,20] Excellent alkaline resistance above 80 °C of AEMs has been believed to be highly required because of the possibility of diminishing the CO2 solubility in water above 80 °C, which could prevent the formation of carbonate or bicarbonate in AEMs.[3,20] Moreover, the kinetics of electrochemical reactions will be enhanced when the fuel cell operated at elevated temperature. In general, the instability of AEMs resulted from the polymer backbones session and the degradation of organic cations. Fluorinated polymer backbone such as PVDF and its derivatives, 3

which showed relatively high chemical and physical stabilities, have been reported to be easily degraded in a high-pH environment.[20] The poly (arylene ether sulfone)s backbone have also been confirmed to be degradated by the cleavage of C–O bonds.[32] Additionally, the well known degradation pathways of organic cation are beta-hydrogen (Hofmann or E2) elimination, direct nucleophilic substitution (SN2), and ylide formation at high pH conditions.[3,20] Thus, the stability of other organic cations under alkaline condition, such as benzimidazolium,[33] morpholinium,[34] pyridinium,[35] pyrrolidinium,[36] metal organic frameworks,[37] ruthenium,[38] tetrakis(dialkylamino)phosphonium,[39] and N- or/and C2-substituted imidazoliums[40] have been investigated widely to overcome the disadvantages of using the ubiquitous benzyltrimethylammonium (BTMA). Marino and Kreuer[41] recently described a class of quaternary spiroammonium compounds that exhibited improved alkaline stability as compared to the acyclic counter. Coates and coworkers[40] found that the substituent identity at each position of the imidazolium ring has a dramatic enhancement of alkaline stability. Therefore, efficient anion transport combined with alkaline-resistant cationic groups and polymer backbones are critical criteria in the design of advanced AEM materials.

Most recently, we have developed a series of comb-shaped poly(2,6-dimethylphenylene oxide) (PPO) AEMs which displayed improved anion conductivities, substantial alkaline stability, and advantageous properties for incorporation as catalyst ionomers in fuel cell devices.[42] The formed comb-shaped structures by introducing one long alkyl chain of up to 16 carbon atoms pendant to the cationic center induced microphase separation, and thus enhanced the ionic conductivities of these membranes. The long alkyl side chain has also been confirmed to be effective for improving the alkaline stability as a result of the reducing of nucleophilic attack by water or hydroxide at the quaternary ammonium (QA) moieties. However, poor mechanical properties of membranes were observed for high degree of functionalization (> 40%) probably owing to the poor compatibility between alkyl side chain and PPO 4

backbone. Herein, we therefore employed the polystyrene backbone to develop a new system that would support high IEC values and thus high hydroxide conductivity, while retaining other good material properties, including hydroxide stability. It is believed that the polymer backbone of polyolefin would be more compatible with alkyl chains than the polyaromatics. Importantly, the polystyrene backbone has been proven to be favorable for alkaline resistance of benzyltrimethylamine (BTMA) groups in AEMs according to Hickner and coworkers’ reports.[43] Combined with controllable Cu(I)-catalyzed “click chemistry” and azide-assisted self-crosslinking techniques, the crosslinked and comb-shaped polystyrene AEMs were prepared readily. A detailed investigation on the properties of these PS AEMs was performed. Their ionic conductivities, microphase-separated structure, and alkaline stability and initial H2/O2 fuel cell performance are described and compared to those of crosslinked comb-shaped PPO AEMs.

2. Experimental section

2.1. Materials

4-Vinylbenzyl chloride (VBC, 90%) (Sigma-Aldrich) was washed with 5% NaOH solution for the removal of polymerization inhibitor. 2,2’-azobis-isobutyronitrile (AIBN), N,N-dimethylhexadecylamine, propargyl and other reagents were obtained from Energy and used as received. Alkyne-termined quaternary ammonium having one alkyl side chain (16 carbon atoms) were synthesized accroding to our previous reports.[45]

2.2. Synthesis of poly(4-vinylbenzyl chloride) (PVBC) homopolymer

Typically, a mixture of monomer VBC (5mL, 35 mmol) and AIBN (0.029 g, 0.175 mmol) was added to a glass tube. After degassing with Ar gas, the reaction was carried out at 65 oC. After 4 h, the mixture 5

was precipitated into excess methanol to yield a white solid. The polymer was purified by dissolving with THF and reprecipitating in methanol. The white product was dried under vacuum at 50 oC overnight with the yield of 73%.

2.3. The synthesis of crosslinkable cPS-x copolymers The copolymers were synthesized by alkyne-azide “click chemistry” in NMP. A typical experiment of cPS-50 synthesis is as follows: To a 3-necked flask equipped with a magnetic stirring bar, PS-N3 (4.7g, 100 mmol), Alkyne-termined quaternary ammonium having one alkyl side chain bromide salt (1.94g, 50 mmol), ligand (PMDETA, 10 mmol), CuBr catalyst (5 mmol) were dissolved in 5 mL of NMP. The flask was degassed by three freeze-pump-thaw cycles, left under vacuum, and placed in a thermostated oil bath at 50 °C for 12 h. The reaction mixture was precipitated into water and then the product was washed with water three times. The crosslinkable comb-shaped copolymer cPS-50 (where the 50 refers to the degree of functionalization groups) was obtained after drying in vacuum at 80 °C for 24 h.

2.4. Membrane preparation, evaluation of crosslinking efficiency and ion exchange The membranes (thickness ~70 μm) having crosslinkable azide groups were prepared by solution casting in NMP at 75 °C. Subsequently, the crosslinking of membranes was carried out in a vacuum oven at 135 °C for 18 h. The gel fraction was employed to determine the degree of crosslinking of the membranes, and was obtained by subjecting the membranes to NMP at 80 °C for 24 h under argon to effect removal of the soluble polymer fraction. The gel fraction value was calculated from the residual mass of the sample by the equation: gel fraction=(Wd/Wi)*100%, where Wi is the initial weight of dried membranes and Wd is the weight of the dried insoluble fraction of membranes after extraction. The

6

crosslinked AEMs xPS-x in the hydroxide form (x Alkyne-termined quaternary ammonium having one alkyl side chain, thus the degree of crosslinkable sites is 100-x) were obtained by ion exchange with 1 M NaOH solution at room temperature for 48 h.

2.5. Characterization and measurements 1

H NMR spectra were measured at 400 MHz on a Bruker AV 400 spectrometer using CDCl3 or

DMSO-d6 as the solvent. The membranes in the bromide form were immersed in 100 mL of 0.1 M NaNO3 standard for 24 h. The solutions were then titrated with a standardized AgNO3 solution using K2CrO3 as an indicator to obtain the titrated gravimetric IEC values.

Water uptake was measured after drying the membrane in hydroxide form at 60 °C under vacuum for 12 h. The dried membrane was immersed in water and periodically weighed on an analytical balance until a constant mass was obtained, giving the mass-based water uptake. Membrane densities and volumetric IECv (meq./cm3) were determined according to previously reported methods based on membrane water uptake using the following equation.

IECv 

1

 polymer

IECw (1) WU ( wt %)  100   water

where IECw is the gravimetric IEC (meq./g) and ρ (g/cm3) is the density. Conductivity (σ, Scm-1) of each membrane sample (size: 1 cm × 4 cm) was obtained using =d/LsWsR (d is the distance between reference electrodes, and Ls and Ws are the thickness and width of the membrane, respectively). The membrane impedance was measured over the frequency range from 100 mHz to 100 kHz by two-point probe alternating current (AC) impedance spectroscopy using an 7

impedance/gain-phase analyzer (Biologic, FR). The hydroxide conductivity measurements under fully hydrated conditions in the longitudinal direction were carried out with the cell immersed in water which was degassed and blanketed with flowing Ar.

Small-angle X-ray scattering curves of unstained dry bromine form membranes were obtained using a Rigaku (formerly Molecular Metrology) instrument equipped with a pinhole camera with an Osmic microfocus Cu Ka source and a parallel beam optic. Typical counting times for integration over a

Measurements were taken under vacuum at ambient temperature on dry samples. Scattering intensities were normalized for background scattering and beam transmission.

2.6. Fabrication of membrane/electrode assemblies (MEA)

After mixing 46.4 wt% Pt/C catalysts (Johnson Matthey Co.) with de-ionized water, 1-propanol and ionomer solution (AS-4, Tokoyama Corp.) to form a well-dispersed catalyst ink, the ink was coated onto surface of SGL 25BC carbon paper (HCP120, HESEN) using an air spray gun to obtain a catalystcoated substrate (CCS) for the electrodes. The Pt loading and ionomer content in the catalyst layer were designed to be ~0.50 mg/cm2 and ~20 wt.%, respectively. Anode CCS, xPS-65 membrane (Tokuyama Corporation, Japan) and cathode CCS were assembled together in a cell fixture to form a membrane/electrode assembly (MEA). The electrode size was 2.25 cm  2.25 cm (~5 cm2).

2.7. Fuel cell performance evaluation

Fue cell testing was carried out on a commercial fuel cell testing system (Smart 2, WonATech, Korea). The 5 cm2 MEAs were mounted in a 2.25  2.25 cm test fixture containing two graphite blocks with machined single serpentine flow channels and two gold-coated current collector plates. The fuel cell 8

performance was measured at 60 °C. The anode/cathode humidifier temperatures were controlled to be 5 °C higher than the cell temperature in order to achieve full humidification (RH = 100%). Fully humidified hydrogen was supplied to the anode at 500 SCCM as indicated in the data, while fully humidified oxygen was supplied to the cathode at 200 SCCM. The MEA was activated by operating at high current density (potentiostatic cell discharge at 100 mV) until the current density increased to a maximum and became constant after 30 min to 1 h. After full activation, the fuel cell polarization curve was measured under galvanostatic mode, i.e. holding the fuel cell at serial constant currents (for example, 50, 100 and 250 mA/cm2, etc.) for 3 min. The cell voltage as a function of current density was recorded using fuel cell testing software.

3. Results and Discussion

3.1. Synthesis and characterization of crosslinked comb-shaped copolymers

x

(a)

135 oC

N Cl

Cl

100-x

Click

NaN3

AIBN

N

N

N3

N3 Br

14

N

H3C

CH3

O

CH3

O 100-x

x

CH3 N3

CH3

CH3

CH3

O

or

O 60

40 N N

5 CH3

CH3

O

135 oC

O 45

50

N3

Br

N N

N

Br

N

14

N

N

CH3

O

14

CH3

(b)

xPS-x

H3C

H3C

PPO-40

xPPO-50

Scheme 1. Synthesis of crosslinked comb-shaped xPS-x, xPPO-50, PPO-40 by click chemistry and

9

azide-assisted self-crosslinking.

Poly (4-vinylbenzyl chloride) (PVBC) homopolymer (Mn=112,000, Mw=158000, PDI=1.41) was firstly synthesized by radical polymerization using AIBN as inititor (Scheme 1). Subsequently, the azide-functionalized PS-N3 homopolymers were obtained readily with a high yield of 99 % by treatment of the PVBC with sodium azide at 60 oC overnight, as shown in Scheme 1. The chemical shift corresponding to benzylic methylene shifted from 4.42 ppm to 4.14 ppm. The Menshutkin reaction with high-yielding was employed to prepare the clickalbe precursor of alkyne-functionalized quaternary ammonium based on N,N-dimethylhexadecylamine. A series of crosslinkable cPS-x (x refers to the degrees of functionalization) were synthesized using Cu(I)-catalyzed azide alkyne click chemistry (CuAAC) at 50 oC for 12 hours. Thus, the degree of crosslinkable sites could be calculated to be 100-x. Moreover, the uncrosslinked PS-50 with degree of functionalization of 50 % was synthesized by copolymerization with styrene and 4-vinylbenzyl chloride, subsequently azidation and click reaction. As the control experiments for fuel cell application, the comb-shaped and crosslinked comb-shaped AEMs based poly(2,6-dimethyl phenylene oxide) (PPO) were also synthesized by Cu(I)-catalyzed azide alkyne click chemistry (CuAAC) accroding to previous reports (Scheme 1).

10

Fig. 1. 1H NMR of PS-N3 and PS-50 in DMSO-d6. Generally, the “click” reaction is quantitative. After the click reaction, the chemical shift of benzylic methylene protons shifted from 4.45 to 4.85 ppm, as shown in Fig. 1. The absence of benzylic methylene protons both on the azide (H5‘) and 1,2,3-triazole ring (H5) suggested the successful “click” reaction and existence of residual thermally crosslinkable azide groups. The degree of crosslinkable sites of cPS-50 were calculated to be around 52 % from the integral ratio of methylene protons H5 to H5‘, which was agreed well with the theoretical value of 50 %. Moreover, the appearance of the proton H6 in 1,2,3-triazole at 8.83 ppm and the additional signals of the methylene protons H7 linked to the quaternary ammonium (QA) unit at 5.62 ppm further confirmed the successful 'click' reaction. Excellent solubility in pure NMP, DMF and DMAc was observed for all of the crosslinkable cPS-x copolymers. Light yellow-colored, flexible and transparent membranes with a thickness of ~40 µm were obtained by casting from NMP solution. However, the membrane based on PPO polymer backbone become very brittle or even break into pieces during membrane formation when the degree of functionalization was higher than 40 %. Similar behavior was found by previously reported comb-shaped PPO AEMs.[12,42] This results suggested the polystyrene backbone shown a good compatibility with the long alkyl side chain which would significantly affect the conductivity and fuel cell performance as will discussed below. After the solvent evaporation, the as-obtained membranes were thermally treated at 135 oC for 18 h in vacuum for crosslinking to get crosslinked xPS-x membrane. Accroding to our previous reports,[45] the reactive nitrene intermediate which generated from the decomposition of azide would reacted with C-H bond in polymer through a radical process, and thus to give the crosslinked membranes. The crosslinked membranes could not be dissolved in water, methanol or n-propanol suggesting high degrees of crosslinking. The gel fraction of xPS-x membranes in NMP was more than 85 %, which further indicated that crosslinking occured at 135 oC. After crosslinking, the azide band at 2120 cm-1 in the FT11

IR spectrum (Fig. 2) did not disappeared completely suggesting the uncomplete reaction of the azide groups during thermal treatment. It is assumed that the crosslinking reaction would be difficult to proceed under the solid membrane. However, the residual azide groups have been proved to show a negligible effect on the membrane properties.[46] The calculated ion exchange capacity (IECw) from proton NMR of crosslinkable cPS-x was in the range of 1.40 to 1.85 meq.g-1 (Table 1) and are also very close to the experimental values determined from titration. Thus, the titrated IECw values of xPS-x membranes were used in the following discussion.

Fig. 2. FT-IR spectra of PS-50 and xPS-50 membranes.

3.2. Morphological structure of xPS-x membranes

The small-angle X-ray scattering (SAXS) was used to characterize the morphological structure of the crosslinked xPS-50 and xPPO-50 membranes. A clear ionomer peak for crosslinked comb-shaped xPS-50 was observed, as shown in Fig. 3. This result indicated a nonophase separation between hydrophilic and the hydrophobic domain in this crosslinked membranes which is similar with our previous report comb-shaped poly(2,6-dimethyl phenylene oxide) anion exchange membrane.[42]

12

Additionally, the xPPO-50 AEM also displayed a similar ionomer peak in its SAXS result (Fig. 3). The result indicated that the xPS-x membrane had a similar morphological structure with the xPPO-x membrane in spite of their different polymer backbone thanks to their similar polymer architecture. Thus, a periodic structure at a length scale of d=2π/qmax, where qmax is the peak maximum was confirmed in this crosslinked comb-shaped polystyrene AEM. The corresponding d-spacing value was determined to be around the range of 3.1 nm, which roughly corresponds to the length of the extended aliphatic side chains. Nevertheless, the fact of no second order peak suggested that the arrangement of the phase separated domains was only locally correlated, and no long-range ordered structures are formed.

Intensity

xPS-50 xPPO-50

1

2

3

4

5

-1

q (nm )

Fig. 3. Small angle X-ray scattering (SAXS) of dry xPS-50 and xPPO-50 membranes in bromine form.

3.3. Water uptake and swelling ratio of crosslinked xPS-x membranes

Table 1. IEC, water uptake, and hydroxide conductivity of un/crosslinked membranes.

13

a

sample

IECwa

IECwb

IECv

WU (wt %) c

λ

△l

Conductivity (HCO3-, mS/cm) c

Conductivity (OH-, mS/cm) c

xPS-35

1.40

1.42

1.27

10.5

4.0

2.5

2.4

10.2

xPS-50

1.66

1.62

1.32

21.2

7.0

3.8

6.5

26.3

xPS-65

1.85

1.81

1.43

26.0

8.0

6.1

11.5

48.7

PS-50

1.68

1.65

1.23

32.6

11.0

3.9

9.8

42.5

xPPO-50

1.56

1.51

1.27

18.2

6.6

2.2

3.2

14.5

PPO-40

1.50

1.49

1.12

25.8

9.6

2.7

7.2

28.0

calculated from 1H NMR; b titrated values; c measured at 20 oC in water. Generally, higher gravimetric IECw membranes showed higher water uptake (WU) due to their

increased ion content. As shown in Fig. 4, the water uptake of xPS-x membranes increased linearly with increaseing IECw ranging from 10.5 % to 26 % at 20 oC. The λ of xPS-x membranes, e.g. number of absorbed water molecules per quaternary ammonium (QA) group, was calculated to be in the range of 4 – 8.0. These values were lower than that of the uncrosslinked PS-50 membrane (λ=11). The AEMs based on PPO displayed similar water uptake behavior, e.g. lower water uptake was observed for crosslinked membrane, as listed in Table 1. Even at elevated temperature, the xPS-65 membrane showed a reasonable water uptake of 58 % in spite of its high IEC value of 1.81 meq./g and lower degree of crosslinking. However, the PS-50 and PPO-40 membranes even have the lower IEC value shown a excessive water swelling at 80 oC without crosslinking, as shown in Fig. 4. Unlike the previous reported comb-shaped AEMs prepared directly by Menshutkin reaction using N,N-dimethylhexadecylamine[42] in which lower water uptake even at elevated temperature was observed, the clicked comb-shaped AEMs displayed significant dependence on temperature of water. It is assumed that the triazole groups provided more sites for water and/or hydroxide to form more effective and continuous hydrogenbonding networks, as confirmed by our previous report. Thus, the crosslinked xPS-x membranes exhibited lower dimensional swelling behavior (Table 1), endowing it with better dimensional stability as a result of the lower water uptake. Considering the remarkable agreement between the high solvent

14

resistance and low swelling ratio, a strong covalent crosslinking network was formed by the azideassisted crosslinking which was further confirmed by the gel fraction in NMP (> 85 %) as discussed above. These results indicated that the crosslinking could restrict the water absorption of membranes efficiently.

o

80 C, Excessive Swelling

60

Water Uptake (wt %)

50

o

80 C PS-50 PPO-40 xPS-x

40 o

20 C

30 xPPO-50

20 10 1.4

1.5

1.6

1.7

1.8

1.9

IECw (meq./g)

Fig. 4. Liquid water uptake of membranes as a function of (a) IECw and (b) temperature. The volumetric IEC (IECv, meq./cm3), refers to the molar concentration of QA groups per unit volume of the hydrated material, was further calculated to detailed comparsion of the water uptake and hydrated ion content among the membranes. Normally, the volumetric IECv of membranes increased with increasing IECw if the membrane did not swellen excessively in water. As shown in Fig. 5, the IECv of crosslinked xPS-x membranes increased from 1.27 to 1.43 meq./cm3 as IECw increased from 1.42 to 1.81 meq./g. However, the volumetric IECv of crosslinked xPS-x membranes decreased with increasing IECw at high temperature of 80 oC. For the membranes measured at 20 °C, the increased quaternary ammonium group concentration of the dry polymer was retained after equilibration with water, while hydration of membrane at high temperature of 80 oC resulted in swelling and dilution of the 15

ion concentration. However, these values were still range from 1.17-1.23 meq./cm3 which are much higher than that of uncrosslinked PS or PPO AEMs (lower than 0.5 meq./cm3). This results indicated the crosslinked membranes did not show excessive swelling in liquid water even at elevated temperature, suggesting the crosslinking could restrict the water absorption more effectively while the membrane PS50 and PPO-40 without crosslinking swellen excessively in water at elevated temperature.

o

20 C 1.4

3

IECv (meq./cm )

xPS-x xPPO-50 1.2

o

80 C

PS-50

PPO-40

1.0

o

80 C, Excessive Swelling 0.8 1.4

1.5

1.6

1.7

1.8

1.9

IECw (meq./g)

Fig. 5. Volumetric IECv of crosslinked and uncrosslinked comb-shaped copolymer membranes in water at 20 and 80 oC as a function of gravimetric IECw.

3.4. Ionic Conductivity

16

Hydroxide Conductivity (S/cm)

normalized hydroxide conductivity

xPS-x o 80 C

100

80

60

xPS-x o 20 C

PS-50 o

20 C 40

PPO-40 o

20 C

o

xPPO-50, 80 C

20 o

xPPO-50, 20 C 0 1.4

1.5

1.6

1.7

1.8

6.4

(b)

xPS-x

5.6 4.8

PS-50

4.0 3.2

PPO-40 2.4

xPPO-50

1.6

1.9

1.4

IEC (meq./g)

1.5

1.6

1.7

1.8

1.9

IEC (meq./g)

Fig. 6. (a) Hydroxide conductivity and (b) λ-normalized hydroxide conductivity of crosslinked and uncrosslinked comb-shaped membranes as a function of IECw.

Similar to the water uptake, the hydroxide conductivity of the xPS-x membranes increased with increasing IECw because of the increase in the water content which increased the local mobility of water. The crosslinked xPS-x membranes displayed the hydroxide conductivity at 20 oC ranging from 10.2 to 48.7 mS/cm (Table 1). When comparing the hydroxide conductivity of crosslinked comb-shaped xPS-x and uncrosslinked PS-x membranes, the xPS-x membranes showed much lower conductivity compared to PS-x membranes under the same conditions probably due to their lower water uptake. For example, the xPS-50 membrane with IECw=1.62 meq./g showed the hydroxide conductivity of 26.3 mS/cm, which was almostly one times lower than that of the PS-50 membrane (42.5 mS/cm) with IECw of 1.65 meq./g. The highest hydroxide conductivity of 101.2 meq./g was achieved for crosslinked xPS-65 membranes having IECw of 1.81 meq./g at 80 oC, as shown in Fig. 6a. However, the uncrosslinked membranes PS-50 and PPO-40 were excessive swelling at 80 oC, and thus conductivity measurement can not be carried out under this condition. After normalizing the membrane hydroxide conductivity on the basis of the water uptake, the crosslined xPS-x membranes still showed comparable or even higher λ-normalized hydroxide conductivity values than the uncrosslinked PS-50 membrane and other reported 17

crosslinked AEMs, as shown in Fig. 6b. This result indicates that the well-developed microphase separation of comb-shaped copolymer membranes could efficiently utilize water to facilitate hydroxide transport in spite of their crosslinking architecture. Furthermore, the comb-shaped AEMs based on PS showed higher hydroxide conductivity and even λ-normalized hydroxide conductivity than that of PPO AEMs with the same comb-shaped architecture, as shown in Fig. 6. It is assumed that the polystyrene which showed a good compatibility with alkyl chain which induced a good ionic domains for anion transport of AEMs though further investigation needed to be done to explain this phenomenon. Moreover, Fig. 7 showed the temperature dependence on the bicarbonate conductivity of crosslinked xPS-x and xPPO-50 AEMs. The bicarbonate conductivity steadily increased with temperature and exhibited the highest conductivity of 22.3 mS cm-1 for a xPS-65 membrane having a highest IEC value

Bicarbonate Conductivity (mS/cm)

of 1.81 meq./g at 80 oC.

xPS-35 xPS-50 xPS-65 xPPO-50

25

20

15

10

5

0 20

40

60

80

o

Temperature ( C)

Fig. 7. Temperature dependence of conductivity for the AEMs in the HCO3- form.

3.5. Alkaline stability

18

Generally, the Hoffmann elimination of β-hydrogen, direct nucleophilic substitution, or nitrogen ylide formation has been believed to be the degradation pathways of tetraalkylammonium ions under alkaline conditions in the AEMs.[3] Therefore, the alkaline stability of the crosslinked AEMs was evaluated in 2 M, 5 M and 10 M NaOH at 80 °C for 400h. In our previous reports, the comb-shaped AEMs based on PPO exhibited excellent alkaline stability in 1 M NaOH at 80 °C [10,42] and maintained their toughness, flexibility and more than 80 % original hydroxide conductivity even after 2000 h stability testing. Unlike testing in the hot DI water in which the uncrosslinked PS-50 and PPO-50 swellen excessively, these membrane did not become dissolved in hot NaOH aqueous probably due to the rapid crosslinking during stability testing. Thus, the reduced viscosity and molecular weight changes of all of the membranes were not able to be obtained due to this crosslinking after stability testing, as observed in our previous reports.45

However, the IEC and bicarbonate conductivities changes of these membrane were recorded after stability testing. As shown in Fig. 6a, the bicarbonate conductivities of crosslinked xPS-x and xPPO-50 membranes remainted at ~ 90 % of their initial values after the test of 400 h even at high NaOH concentration of 10 M at 80 oC, while the conductivity of the uncrosslinked PS-50 and PPO-40 membanes decreased to ~ 80 % of the initial value (Fig. 8a). This degradation behavior indicated that the crosslinked membranes have much better alkaline stability than the uncrosslinked membranes, and probably have sufficient durability for AEM fuel cell application. The IEC value of crosslinked membranes decreased little after the alkaline stability testing, which further confirms their excellent alkaline stability, as shown in Fig. 8b.

19

100

(a)

Crosslinked

Loss in Conductivity (%)

90 80

Uncrosslinked 70 xPS-35 xPS-50 xPS-65 PS-50 xPPO-50 PPO-40

60 50 40 30 0

2

4

6

8

10

Concentration of NaOH (mol/L)

100 (b)

Crosslinked

90

Loss in IEC (%)

80

Uncrosslinked

70

xPS-35 xPS-50 xPS-65 PS-50 xPPO-50 PPO-40

60 50 40 30 0

2

4

6

8

10

Concentration of NaOH (mol/L)

Fig. 8. The changing trend in (a) bicarbonate conductivity and (b) IEC values of AEMs after immersion in NaOH solution at 80 °C for 400 h.

3.6. Fuel cell performance

In our previous reports, the comb-shaped PPO AEMs have shown good initial performance in AEMFC single cells as the ion-conductive material in the catalyst layer. We hypothesized that quaternized copolymers that can have good performance and lifetime in the catalyst layers will also serve as good membranes. Thus, our original intention was to employ these highly alkaline stable and ion-conductive comb-shaped PPO AEMs as polymer electrolyte membranes in AEMFC. In spite of numerous attempts, the PPO AEMs displayed very poor feasibility for AEMFC testing. After membrane 20

electrode assembling, no open circuit voltages (OCVs) could be observed. The inability to obtain the initial fuel cell performance may be attributed to the incompatibility between PPO backbone and the numerous pendant long alkyl chains the C16 chain which have affected the water transport in the interface between the catalyst layer and membrane,42 though a more complete study would be needed to explain this phenomenon. However, the polystyrene backbone showed good compatibility with alkyl chain as discussed above, and thus was chosen for novel comb-shaped AEMs for fuel cell application. Interestingly, the open circuit voltage (OCV) was obtained successfully after membrane electrode assembly when the xPS-50 was used as membrane. The OCV are close to the theoretical value of about 1.1 V, indicating that the interface between membrane and catalyst layer do not affect the catalyst function of Pt significantly. Fig. 9 showed the polarization curves of an H2/O2 AEMFC with crosslinked comb-shaped copolymers xPS-50 as the membrane. The peak power density for the AEMFC with the xPS-65 membrane was 88 mW cm-2. The result suggested that the polystyrene polymer backbone may have helped to build a more efficient phase boundary between the catalyst layer and membrane than that of PPO backbone.

Cell voltage (V)

1.0

80

0.8

60

0.6 40 0.4 20 0.2 0

0.0 0

50

100

150

200

250

300

2

Current density (mA/cm )

21

2 Power density (mW/cm )

100

1.2

Fig. 9. Polarization curves and power density curves of an AEMFC with the xPS-65 membrane at 60 ° C with (a) H2/O2 flow rate of 500/200 cm3/min.

4. Conclusion

In summary, we have designed and synthesized a series of crosslinked, comb-shaped polystyrene (PS) anion exchange membranes (AEMs) having C-16 alkyl side chain that allow tuning membrane properties for H2/O2 alkaline fuel cell application. Tough and transparent membranes were obtained in spite of their high degree of functionalization probably due to the excellent compatibility between polystyrene backbone and long alkyl chains. In contrast, the comb-shaped PPO AEMs become very brittle when the degree of functionlization was more than 40 %. After azide-assisted self-crosslinking via thermal treatment at 135 oC, the as-obtained AEMs exhibited lower water uptake, and thus good dimensional stability. Excellent chemical stabilities of these AEMs were observed under severe conditions as confirmed by minor changing in IEC, ion conductivity. Higher hydroxide conductivities for the crosslinked comb-shaped polystyrene membranes were observed than the PPO-based AEMs with similar crosslinked, comb-shaped architecture. It is assumed that the excellent compatibility between polystyrene backbone and alkyl side chain have helped to build a more efficient morphology for ion transport. Thus, the AEMs based on PPO having C-16 alkyl side chain had not initial performance, e.g. no open circuit voltage (OCV) when it was employed as membrane in fuel cell. However, the initial fuel cell performance with the peak power density of 88 mW/cm2 was observed successfully for the H2/O2 alkaline fuel cell with xPS-65 membrane probably due to a more efficient phase boundary between the catalyst layer and membrane which built from polystyrene polymer backbone.

Acknowledgements

22

Financially supported by the National Science Foundation of Ningbo (No. 2015A610059 and No. 2015A610029). The authors also acknowledge the financial support from National Natural Science Foundation of China (No. 21474126 and 21504101 ) and Hundred Talents Program of the Chinese Academy of Science.

References

[1]

J.R. Varcoe, R.C.T. Slade, Prospects for alkaline anion-exchange membranes in low temperature fuel cells, Fuel Cells 2 (2005) 187–200.

[2]

Y. Wang, J. Qiao, R. Baker, J. Zhang, Alkaline polymer electrolyte membranes for fuel cell applications, Chem. Soc. Rev. 42 (2013) 5768–5787.

[3]

J. R. Varcoe, P. Atanassov, D. R. Dekel, A. M. Herring, M. A. Hickner, P. A. Kohl, A. R. Kucernak, W. E. Mustain, K. Nijmeijer, K. Scott, T. Xu, L. Zhuang, Anion-exchange membranes in electrochemical energy systems, Energy Environ. Sci. 7 (2014) 3135-3191.

[4]

N. J. Robertson, H. A. Kostalik, T. J. Clark, P. F. Mutolo, H. D. Abruna, G. W. Coates, Tunable high performance cross-Linked alkaline anion exchange membranes for fuel cell applications, J. Am. Chem. Soc. 132 (2012) 3400–3404.

[5]

M. Zhang, J. Liu, Y. Wang, L. An, M. D. Guiver, N. Li, Highly stable anion exchange membranes based on quaternized polypropylene, J. Mater. Chem. A 3 (2015) 12284–12296.

[6]

J. R. Varcoe, R. C. T. Slade, E. L. H. Yee, S. D. Poynton, D. J. Driscoll, D. C. Apperley, Poly(ethylene-co-tetrafluoroethylene)-derived radiation-grafted anion-exchange membrane with properties specifically tailored for application in metal-cation-free alkaline polymer electrolyte fuel cells, Chem. Mater. 19 (2007) 2686–2693.

[7]

Q. H. Zeng, Q. L. Liu, I. Broadwell, A. M. Zhu, Y. Xiong, X. P. Tu, Anion exchange

23

membranes

based

on

quaternized

polystyrene-block-poly(ethylene-ran-butylene)-block-

polystyrene for direct methanol alkaline fuel cells, J. Membr. Sci. 349 (2010) 237–243. [8]

Y. Luo, J. Guo, C. Wang, D. Chu, Quaternized poly(methyl methacrylate-co-butyl acrylate-covinylbenzyl chloride) membrane for alkaline fuel cells, J. Power Sources 195 (2010) 3765–3771.

[9]

L. Wu, G. Zhou, X. Liu, Z. H. Zhang, C. Li, T. Xu, Environmentally friendly synthesis of alkaline anion exchange membrane for fuel cells via a solvent-free strategy, J. Membr. Sci. 371 (2011) 155–162.

[10]

N. Li, T. Yan, Z. Li, T. Thurn-Albrecht, W. H. Binder, Comb-shaped polymers to enhance hydroxide transport in anion exchange membranes, Energy Environ. Sci. 5 (2012) 7888-7892.

[11]

M. R. Hibbs, C. H. Fujimoto, C. J. Cornelius, Synthesis and characterization of poly(phenylene)based anion exchange membranes for alkaline fuel cells, Macromolecules 42 (2009) 8316–8321.

[12]

J. Yan, M. A. Hickner, Anion exchange membranes by bromination of benzylmethyl-containing poly(sulfone)s, Macromolecules 43 (2010) 2349–2356.

[13]

M. R. Hibbs, M. A. Hickner, T. M. Alam, S. K. McIntyre, C. H. Fujimoto, C. J. Cornelius, Transport properties of hydroxide and proton conducting membranes, Chem. Mater. 20 (2008) 2566–2573.

[14]

J. Wang, S. Li, S. Zhang, Novel hydroxide-conducting polyelectrolyte composed of an poly(arylene ether sulfone) containing pendant quaternary guanidinium groups for alkaline fuel cell applications, Macromolecules 43 (2010) 3890-3896.

[15]

D. S. Kim, A. Labouriau, M. D. Guiver, Y. S. Kim, Guanidinium-functionalized anion exchange polymer electrolytes via activated fluorophenyl-amine reaction, Chem. Mater. 23 (2011) 3795 – 3797.

[16]

Q. Zhang, S. Li, S. Zhang, A novel guanidinium grafted poly(aryl ether sulfone) for high-

24

performance hydroxide exchange membranes, Chem. Commun. 46 (2010) 7495-7497. [17]

S. Gu, R. Cai., T. Luo, Z. Chen, M. Sun, Y. Liu, G. He, Y. Yan, A soluble and highly conductive ionomer for high-performance hydroxide exchange membrane fuel cells, Angew. Chem. Int. Ed. 48 (2009) 6499 –6502.

[18]

P. Hinksman, D.H. Isaac, P. Morrissey, Environmental stress cracking of poly(vinylidene fluoride) and welds in alkaline solutions, Polym, Degrad. Stab. 68 (2000) 299-305.

[19]

J. Pan, S. Lu, Y. Li, A. Huang, L. Zhuang, J. Lu, High-performance alkaline polymer electrolyte for fuel cell applications, Adv. Funct. Mater. 19 (2009) 1–8.

[20]

G. Merle, M. Wessling, K. Nijmeijer, Anion exchange membranes for alkaline fuel cells: a review, J. Membr. Sci. 377 (2011) 1-35.

[21]

H. Dang, E. A. Weiber, P. Jannasch, Poly(phenylene oxide) functionalized with quaternary ammonium groups via flexible alkyl spacers for high-performance anion exchange membranes, J. Mater. Chem. A, 3 (2015) 5280–5284.

[22]

W. Lee, A. D. Mohanty, C. Bae, Fluorene-based hydroxide ion conducting polymers for chemically stable anion exchange membrane fuel cells, ACS Macro Lett. 4 (2015) 453−457.

[23]

J. Wang, S. Gu, R. Xiong, B. Zhang, B. Xu, Y. Yan, Structure–property relationships in hydroxide-exchange membranes with cation strings and high ion-exchange capacity, ChemSusChem 8 (2015) 4229 – 4234.

[24]

M. Tanaka, K. Fukasawa, E. Nishino, S. Yamaguchi, K. Yamada, H. Tanaka, B. Bae, K. Miyatake, M. Watanabe, Anion conductive block poly(arylene ether)s: synthesis, properties, and application in alkaline fuel cells, J. Am. Chem. Soc. 133 (2011) 10646–10654.

[25]

J. Ran, L. Wu, T. Xu, Enhancement of hydroxide conduction by self-assembly in anion conductive comb-shaped copolymers, Polym. Chem. 4 (2013) 4612-4620. 25

[26]

N. Li, L. Wang, M. A. Hickner, Cross-linked comb-shaped anion exchange membranes with high base stability, Chem. Commun. 50 (2014) 4092-4095.

[27]

D. Stoica, L. Ogier, L. Akrour, F. Alloin and J. F. Fauvarque, Novel anion-exchange organic– inorganic hybrid membranes: preparation and characterizations for potential use in fuel cells, Electrochim. Acta 53 (2007) 1596–1603.

[28]

C. C. Yang, S. J. Chiu, W. C. Chien and S. S. Chiu, Novel anion-exchange organic–inorganic hybrid membranes: Preparation and characterizations for potential use in fuel cells, J. Power Sources, 195 (2010) 2212–2219.

[29]

Y. H. Wu, C. M. Wu, T. W. Xu, F. Yu and Y. X. Fu, Novel anion-exchange organic–inorganic hybrid membranes: Preparation and characterizations for potential use in fuel cells, J. Membr. Sci. 321(2008) 299–308.

[30]

J. F. Zhou, M. Unlu, I. Anestis-Richard, P. A. Kohl, Crosslinked, epoxy-based anion conductive membranes for alkaline membrane fuel cells, J. Membr. Sci. 350 (2010) 286–292.

[31]

S. Gu, R. Cai, Y. Yan, Self-crosslinking for dimensionally stable and solvent-resistant quaternary phosphonium based hydroxide exchange membranes, Chem. Commun. 47 (2011) 2856-2858.

[32]

C. G. Arges, V. Ramani1, Two-dimensional NMR spectroscopy reveals cation-triggered backbone degradation in polysulfone-based anion exchange membranes, PNAS, 110 (2013) 2490-2495.

[33]

A. G. Wright, T. Weissbach, S. Holdcroft, Poly(phenylene) and m-terphenyl as powerful protecting groups for the preparation of stable organic hydroxides, Angew. Chem. Int. Ed. 55 (2016) 4818 –4821.

[34]

C. G. Morandi, R. Peach, H. M. Krieg, J. Kerres, Novel morpholinium-functionalized anion26

exchange PBI–polymer blends, J. Mater. Chem. A 3 (2015) 1110-1120. [35]

J. Miyake, K. Fukasawa, M. Watanabe, K. Miyatake, Effect of ammonium groups on the properties and alkaline stability of poly(arylene ether)-based anion exchange membranes, J. Polym. Sci. Part A: Polym. Chem. 52 (2014) 383-389.

[36]

F. Gu, H. Dong, Y. Li, Z. Sun, F. Yan, Base stable pyrrolidinium cations for alkaline anion exchange membrane applications, Macromolecules 47 (2014) 6740-6747.

[37]

M. Sadakiyo, H. Kasai, K. Kato, M. Takata, M. Yamauchi, Design and synthesis of hydroxide ion–conductive metal–organic frameworks based on salt inclusion, J. Am. Chem. Soc. 136 (2014) 1702-1705.

[38]

M. L. Disabb-Miller, Y. Zha, A. J. DeCarlo, M. Pawar, G. N. Tew, M. A. Hickner, Water uptake and ion mobility in cross-linked bis(terpyridine)ruthenium-based anion exchange membranes, Macromolecules 46 (2013) 9279.

[39]

K. J. T. Noonan, K. M. Hugar, H. A. Kostalik, E. B. Lobkovsky, H. D.

bru a, G. W. Coates,

Phosphonium-functionalized polyethylene: a new class of base-stable alkaline anion exchange membranes, J. Am. Chem. Soc. 134 (2012) 18161-18164. [40]

K. M. Hugar, H. A. Kostalik, G. W. Coates, Imidazolium cations with exceptional alkaline stability a systematic study of structure–stability relationships, J. Am. Chem. Soc. 137 (2015) 8730–8737

[41]

M. G. Marino, K. D. Kreuer, Alkaline stability of quaternary ammonium cations for alkaline fuel cell membranes and ionic liquids, ChemSusChem 8 (2015) 513 – 523.

[42]

N. Li, Y. Leng, M. A. Hickner, C. Wang, Highly stable, anion conductive, comb-Shaped copolymers for alkaline fuel cells, J. Am. Chem. Soc. 135 (2013) 10124−10133.

[43]

S.

.

u ez, M. A. Hickner, Quantitative 1H NMR analysis of chemical stabilities in anion-

27

exchange membranes, ACS Macro Lett. 2 (2013) 49−52. [44]

N. Li, M. D. Guiver, W. H. Binder, Towards high conductivity in anion-exchange membranes for alkaline fuel cells, ChemSusChem 6 (2013) 1376 – 1383.

[45]

S. He , L. Liu, X. Wang, S. Zhang, M. D. Guiver, N. Li, Azide-assisted self-crosslinking of highly Ion conductive anion exchange membranes, J. Membr. Sci.509 (2016) 48–56.

[46]

Q. Ge, J. Ran, J. Miao, Z. Yang, T. Xu, Click chemistry finds its way in constructing an ionic highway in anion-exchange membrane, ACS Appl. Mater. Interfaces 7 (2015) 28545–28553.

Highlights 

Self-crosslinking of comb-shaped polystyrene with one long alkyl chain has been developed for as alkaline fuel cells operable anion exchange membrane (AEM).



The AEMs exhibited suppressed swelling, excellent alkaline stability while high OH-conductivities were observed.



The as-obtained AEMs showed good initial fuel cell performance while the no open circuit voltage was observed for comb-shaped poly(2,6-dimethyl-phenylene oxide) (PPO) AEMs.

28