Effect of alkaline exchange polymerized ionic liquid block copolymer ionomers on the kinetics of fuel cell half reactions

Journal of Electroanalytical Chemistry 783 (2016) 182–187

Contents lists available at ScienceDirect

Journal of Electroanalytical Chemistry journal homepage: www.elsevier.com/locate/jelechem

Effect of alkaline exchange polymerized ionic liquid block copolymer ionomers on the kinetics of fuel cell half reactions Jacob R. Nykaza a, Yawei Li a, Yossef A. Elabd b, Joshua Snyder a,⁎ a b

Department of Chemical and Biological Engineering, Drexel University, Philadelphia, PA 19104, United States Department of Chemical Engineering, Texas A&M University, College Station, TX 77843, United States

a r t i c l e

i n f o

Article history: Received 21 July 2016 Received in revised form 10 November 2016 Accepted 12 November 2016 Available online 15 November 2016 Keywords: Alkaline exchange ionomers Fuel cells Electrocatalysis

a b s t r a c t Rotating disk electrode (RDE) half-cell experiments were used to determine the impact of a hydroxide-conducting polymerized ionic liquid block copolymer (PILBCP) ionomer on the oxygen reduction reaction (ORR) and hydrogen oxidation reaction (HOR) activity of a commercial Pt/C catalyst at three PILBCP loadings (23, 41, and 57 wt.% total solids). Increasing loadings of PILBCP resulted in reduced surface coverage of both the hydrogen (HUPD) and hydroxide/oxide (OHad/Oad) as evidenced by the cyclic voltammograms due to both a physical blocking of surface catalytic sites and a water/ion diffusional resistance imparted by the presence of the film. With the maximum loading of 57 wt.% PILBCP, a decrease of 88% for the kinetic current density (Jk) and 42% for the diffusion limited current (Id) in the ORR and a decrease of 29% for the Id in the HOR compared to bare Pt/C nanoparticles was observed. Similar trends were observed with 60 wt.% Nafion on Pt/C nanoparticles. These results indicate that while substantial, the detrimental effects of PILBCP ionomers on the half-cell reaction kinetics are no worse than those observed with the Nafion ionomer. AEMFC performance optimization for polymers with sufficient hydroxide conductivities should focus on AEM ionomer integration strategies with the goal of optimizing the triple phase boundary and limiting the interfacial resistance between the AEM and the corresponding ionomer in the catalyst layer. © 2016 Elsevier B.V. All rights reserved.

1. Introduction The continued development of alkaline exchange membranes (AEMs) with increasing ion conductivities and mechanical/chemical stability has renewed interest in alkaline chemistry-based fuel cell technologies (alkaline exchange membrane fuel cells, AEMFCs) as an alternative to proton exchange membrane fuel cells (PEMFCs). The passivity of non-noble metals at AEMFC operating/electrochemical conditions facilitates the replacement of the Pt-based catalysts required in the acidic conditions of the PEMFC, making AEMFCs a less expensive and potentially more commercially viable alternative [1–3]. Power is generated in AEMFCs, as with other fuel cell configurations, through the hydrogen oxidation reaction (HOR) (reaction 1) at the anode producing electrons and the oxygen reduction reaction (ORR) (reaction 2) at the cathode. Hydroxyl ions produced at the cathode are transported through the AEM to participate in the HOR. The AEMFC half-cell and overall reactions are listed below. Anode : 2H2 þ 4OH− →4H2 O þ 4e−

⁎ Corresponding author. E-mail address: [email protected] (J. Snyder).

http://dx.doi.org/10.1016/j.jelechem.2016.11.024 1572-6657/© 2016 Elsevier B.V. All rights reserved.

ð1Þ

Cathode : O2 þ 2H2 O þ 4e− →4OH−

ð2Þ

Overall : 2H2 þ O2 →2H2 O þ electricity þ heat

ð3Þ

To date, commercial integration of AEMFCs is limited by the moderate hydroxyl ion conductivity and low chemical stability of currently available AEM chemistries [4]. Research and development has focused on material optimization through the use of various cations [5–9] (e.g., ammonium, imidazolium, phosphonium), polymer structures [10–13] (e.g., grafted, random, block), and cation location [14,15] (e.g., alkyl chain length) in order to address these limitations. While ex situ conductivity and stability data is readily available for a broad range of AEM chemistries [5,7,16,17], few have been successfully integrated into a full cell AEMFC as there are difficulties in solubilizing the AEM polymer into an ionomer for integration into the catalyst layer [3,18– 23]. Inefficiencies associated with the ability to convert the AEM polymer or a polymer of similar chemistry into an ionomer and integrate into a catalyst layer limits ion transport between the catalytic sites at the triple phase boundary and the ion conducting membrane, manifested as an interfacial resistance [24–26]. The slow development of AEM ionomers has also resulted in little understanding of the impact of AEM polymer chemistries on the cathodic ORR and anodic HOR in

J.R. Nykaza et al. / Journal of Electroanalytical Chemistry 783 (2016) 182–187

AEMFCs, where both have a major role in defining the peak power density and performance of the fuel cell. Previous work in our laboratory has focused on improving the ionic conductivity, chemical stability, and mechanical properties of AEMs by exploring polymerized ionic liquid block copolymers (PILBCPs) [14,15, 27–30]. PILBCPs are a new class of block copolymer, which combines the benefits of both PILs and block copolymers. PILs possess unique properties, such as high solid-state ionic conductivity, high chemical, electrochemical, and thermal stability, and a widely tunable chemical platform, where significant changes in physical properties have been observed with subtle changes in chemistry [27,31]. Block copolymers are known to self-assemble into well-defined nanostructures with long-range order where both morphology and domain size are tunable [32–35]. When PILs are incorporated into the block copolymer, the resulting PILBCP possesses numerous orthogonal properties in the solid-state, such as excellent mechanical properties from the nonionic polymer and high ion transport from the PIL and block copolymer morphology. We have recently synthesized a PILBCP, poly(MMA-bMUBIm-OH), comprised of a PIL component (MUBIm-OH = 1-[(2methacryloyloxy) undecyl]-3-butylimidazolium hydroxide) and a nonionic component (MMA = methyl methacrylate) [30]. The family of polymerized ionic liquid block copolymers containing the PILBCP used here has demonstrated sufficient chemical stability at AEMFC relevant conditions [5,7,36]. Membrane electrode assemblies (MEA) were fabricated and tested in a single-cell stack with this PILBCP as both the solidstate membrane separator and the ionomer in the catalyst layers [30]. A peak power density of approximately 30 mW cm−2 (60 °C with H2/O2 at 25 psig (172 kPa) backpressure) was achieved. The resulting AEMFC performance was considerably lower than expected given the high hydroxide conductivity of approximately 56 mS cm−1 at 60 °C and 95% RH for the PILBCP [30]. There are two likely sources of the observed low AEMFC power density: (1) physical/chemical blocking of catalytic sites by the PILBCP (poor triple phase boundary geometry) adversely affecting the kinetics of the anodic and cathodic reactions through both an inactivation of blocked catalytic sites and an ionic/reactant diffusional resistance imparted by the polymer film and/or (2) interfacial resistance between the ion conducting species in the AEM and ionomer within the catalyst layer as a consequence of sub-optimal ionomer integration, catalyst layer fabrication and MEA assembly. In this work we attempt to address the first of these limiting factors by using rotating disk electrode (RDE) half-cell experiments to determine the impact of the PILBCP, poly(MMA-b-MUBIm-OH), on the ORR and HOR activity of a commercial Pt/C nanoparticle catalyst. The effect of PILBCP on both the ORR and HOR was investigated at three PILBCP loadings within the catalyst layer at 23, 41, and 57 wt.% of total solid. The PILBCP results were then compared to Nafion coated Pt/C nanoparticles to determine if Nafion exerts a similar impact on half-cell reaction kinetics.

2. Experimental 2.1. Materials Acetonitrile (ACN, anhydrous, 99.8%), 2-propanol (IPA, electronic grade, 99.999%), sulfuric acid (H2SO4, ACS reagent, 95–98%), and potassium hydroxide (KOH, semiconductor grade, 99.99%) were used as received from Sigma-Aldrich. Nitric acid (HNO3, ACS plus, 70%) was used as received from Fisher Scientific. Perchloric acid (HClO4, omnitrace ultra, 65–71%) was used as received from EMD Millipore. Millipore (Milli-Q Synthesis A10) water with resistivity ≈ 18.2 MΩ cm was used to make all electrolyte solutions. Research grade (99.999%) oxygen (O2), hydrogen (H2), and argon (Ar) were used as received from Airgas. Liquion solution LQ-1105 1100 EW (5 wt.% Nafion) was used as received from Ion Power. The PILBCP was prepared according to literature and the synthetic procedure and properties can be found elsewhere [14,30].

183

2.2. Electrochemical measurements PILBCP and Nafion coated Pt/C nanoparticle catalysts were electrochemically characterized in a three-electrode cell with a Pt mesh (Alfa Aesar) counter electrode and a Ag/AgCl (BASi) reference electrode. The reference electrode was calibrated against a hydrogen reference and found to have an offset of 0.97 V and 0.27 V at 25 °C for 0.1 M KOH and 0.1 M HClO4, respectively. All potentials listed are referenced to the reversible hydrogen electrode (RHE). Prior to any electrochemical experiments, all glassware was cleaned by soaking in a solution of concentrated 1:1 H2SO4:HNO3 for at least 8 h followed by rinsing and boiling in Millipore water. Pt/C thin films on glassy carbon (GC) disks (5 mm diameter, 0.196 cm2, HTW GmbH) were synthesized by drop casting from a catalyst ink in which Pt/C (40 wt.% Pt, Fuel Cell Store HiSPEC 4000) was dispersed in H2O at a concentration of 1 mgcatalyst mL−1. Prior to loading with catalyst, the GC disks were polished to a mirror finish using progressively finer diamond paste down to 0.05 μm (Buehler). The GC disks were then sonicated in Millipore water to remove contaminants. The appropriate volume (7.35 μL) of catalyst ink to achieve a loading of 15 μgPt cm− 2 was pipetted onto the GC disks and dried under a flow of argon to form a uniform layer. Catalyst thin films were approximately 10 μm in thickness with a uniform coating on the disk; any nonuniform films were rejected. Catalytic activity is directly dependent on the quality of the catalyst layer on the disk [38,39]. A similar amount (7.35 μL) of PILBCP solution (0.25, 0.5, and 1.0 wt.% in ACN) or Nafion solution (1.0 wt.% in IPA) was then drop cast onto the catalyst layer and allowed to dry under a flow of argon. Cyclic voltammetry (CV) was performed in Ar purged electrolyte by cycling between 0.18 and 1.07 V vs. the reversible hydrogen electrode (RHE) (Autolab PG Stat 302N) at 50 mV s− 1 for at least 30 cycles or more until the CV curve reached a steady state. All potentials are referenced vs. RHE. The polymer-coated catalysts were then transferred to O2 saturated electrolyte for ORR activity measurements or H2 saturated electrolyte for HOR activity measurements. Using a Pine Instruments rotator (AFMSRCE), the GC disk was rotated at 1600 rpm while running linear sweep voltammetry from 0.18 to 1.07 V vs. RHE at 20 mV s−1. Ohmic iR drop was compensated for through the process described in ref. [40]. 3. Results and discussion The chemical structure of the PILBCP, poly(MMA-b-MUBIm-OH), used here is shown in Fig. 1. This PILBCP was synthesized through an anion exchange metathesis reaction. The bromide form of the conducting PILBCP as reported in [14], poly(MMA-b-MUBIm-Br), was used to coat the Pt/C nanoparticles due to its favorable solubility in organic solvents compared to other hydrophilic ions, such as bicarbonate (i.e., HCO− 3 ) [30]. The uniform solution of poly(MMA-b-MUBIm-Br) in acetonitrile resulted in evenly distributed films of the PILBCP on the Pt/C nanoparticles. The PILBCP films were then ion exchanged to the hydroxide form, poly(MMA-b-MUBIm-OH), by cycling the potential between 0.18 and 1.07 V vs. RHE in Ar purged 0.1 M KOH electrolyte until the CV reached steady state. The large excess of electrolyte in the electrochemical cell results in sufficient dilution of any bromide ions after complete anion exchange, ensuring that bromide ions do not interfere with electrochemical experiments. However, as a precaution, the electrolyte was replaced with fresh 0.1 M KOH for the kinetic measurements after completion of the anion exchange. Fig. 2 contains the CVs for commercial 40 wt.% Pt/C as a function of PILBCP content. With the addition of the PILBCP to the Pt/C film, adsorption features associated with hydrogen (between 0.5 and 0.1 V vs. RHE) and hydroxide/oxide (between 0.7 and 1.0 V vs. RHE) are significantly suppressed. This depression in current is a consequence of both the physical blocking of the platinum sites by the PILBCP, lowering the number of electrochemically active sites

184

J.R. Nykaza et al. / Journal of Electroanalytical Chemistry 783 (2016) 182–187

][m

[

]n

O

O

O

O

( )

9

N N

OH

Fig. 1. Chemical structure of polymerized ionic liquid diblock copolymer (PILBCP): poly(MMA-b-MUBIm-OH).

with increasing polymer content and limiting the amount of water/ electrolyte interfaced with the Pt/C catalyst, as well as an ionic diffusional resistance imparted by the polymer film. At polymer contents beyond 41 wt.%, the CVs indicate nearly complete coverage as little

0.6

0 wt. % PILBCP 23 wt. % PILBCP 41 wt. % PILBCP 57 wt. % PILBCP

Current (mA cm

geo

-2

)

0.4 0.2 0 -0.2 -0.4 -0.6 0

0.2

0.4

0.6

0.8

1

1.2

E (V vs. RHE) Fig. 2. Half-cell cyclic voltammograms (CVs) of Pt/C coated with various PILBCP thicknesses. PILBCP catalyst layer contents: 0 wt.% (red), 23 wt.% (blue), 41 wt.% (green), and 57 wt.% (black). CVs recorded in Ar purged 0.1 M KOH at room temperature with a sweep rate of 50 mV s−1. Loading on the glassy carbon disk in all cases was 15 μgPt cm−2. Potentials were corrected for iR drop (45 Ω) within the electrolyte. Currents are reported as current density per geometric area of the disk. Arrows indicate direction of increasing PILBCP thickness.

to no current associated with Pt-hydroxide/oxide formation or hydrogen underpotential deposition (H UPD ) was observed. Polymer content dependent electrochemically active surface areas (ECSA) as measured by HUPD are listed in Table 1. It is important to note that no extra Faradaic processes other than those that typically occur on Pt, H UPD and Pt oxidation, were observed, indicating that there is no specific adsorption, reversible or irreversible, of charged molecular species associated with the PILBCP on the Pt surface. This is an expected result, yet one that is not often discussed when considering the advantages of AEM vs. PEM polymers and devices. ORR operational potentials for an AEMFC are mostly above the potential of zero charge for Pt regardless of pH [41]. The catalyst is positively charged, repelling the cation/charged species of the PILBCP and other AEM cations/charged species and preventing any active site blocking through the specific adsorption of ions. This is in contrast to PEM polymers, such as Nafion, where the negatively charged sulfonate head groups are shown to specifically adsorb onto the surface of Pt, blocking catalytic sites and lowering the specific and mass based ORR activities [42,43]. Fig. 3(a) contains the ORR polarization curves in O2 saturated 0.1 M KOH at room temperature for 40 wt.% Pt/C nanoparticles as a function of PILBCP content. The ionomer content in AEMFCs range from 20 to 40 wt.% solids, the loadings tested here (23–57 wt.%) cover that range [21,30,44]. ORR polarization curves can be separated into two regimes: (1) the diffusion limited regime (below 0.8 V vs. RHE) and (2) the mixed kinetic/diffusion limited regime (above 0.8 V vs. RHE). In the diffusion limited regime the reaction rate surpasses the rate of transport of reactant to the catalytic surface, characterized by a potential independent current known as the diffusion limited current (Id). In the mixed kinetic/diffusion limited regime Koutecky-Levich analysis can remove any limitations associated with reactant transport and extract the kinetic current density (Jk) which is directly associated with the potential dependent rate of the reaction. The ORR diffusion limited currents (Id) and mass normalized activities (Jk), measured at 0.9 V vs. RHE, for each PILBCP loading are summarized in Table 2. Both the Id and Jk values for the bare Pt/C nanoparticles are similar to literature values [45,46]. The addition of PILBCP to the Pt/C nanoparticles impacts both the reaction kinetics, as well as the diffusion-limited current. Comparing the bare Pt/C (0 wt.%) to the lowest PILBCP content, 23 wt.%, there is a decrease in Jk by 76% (0.54 to 0.13 mA μg−1 Pt ), as well as a decrease in Id by 3% (1.07 to 0.98 mA). This continues to decrease with increasing PILBCP content to 88% and 42% for Jk and Id, respectively, for the highest PILBCP content of 57 wt.%. The relatively high initial decrease in Jk is expected as even with a low PILBCP coverage, the loss of Pt active sites due to physical blocking by the polymer is considerable. The additional decrease in Jk with increasing PILBCP content is attributed to further coating of available platinum sites resulting in a lower electrochemically active surface area (ECSA) and a compounding diffusional resistance of ions and reactant gas from the bulk electrolyte. The Id is found to decrease with increasing PILBCP content as the increased content is manifested as both increases in Pt site coverage and thickness of the film covering the catalyst. A diffusional barrier is imposed by the polymer for the transport of reactant oxygen as well as OH− anions from the bulk electrolyte

Table 1 Hydrogen underpotential deposition (HUPD) values from cyclic voltammograms (CVs). Room temperature 0.1 M KOH with a sweep rate of 50 mV s−1; catalyst loading on glassy carbon disk is 15 μgPt cm−2. Sample [wt.%]a

HUPD [cm2 Pt (μg Pt)−1]

PILBCP - 0 PILBCP - 23 PILBCP - 41 PILBCP - 57 Nafion - 0 Nafion - 60

0.39 0.11 0.038 0.034 0.30 0.21

a

PILBCP in 0.1 M KOH and Nafion in 0.1 M HClO4.

J.R. Nykaza et al. / Journal of Electroanalytical Chemistry 783 (2016) 182–187

1

(b)

0 wt. % PILBCP 23 wt. % PILBCP 41 wt. % PILBCP 57 wt. % PILBCP

-2

-1

geo

)

4

Current (mA cm

-2

geo

)

0

Current (mA cm

(a)

0 wt. % PILBCP 23 wt. % PILBCP 41 wt. % PILBCP 57 wt. % PILBCP

185

-2 -3 -4

2

0

-2

-5 -6

0

0.2

0.4

0.6

0.8

1

-4 -0.2

0

0.2

0.4

0.6

0.8

1

E (V vs. RHE)

E (V vs. RHE)

Fig. 3. Half-cell reactions for PILBCP coated Pt/C catalyst. (a) Oxygen reduction reaction (ORR) and (b) hydrogen oxidation reaction (HOR) for Pt/C catalyst layer with PILBCP contents of 0 wt.% (red), 23 wt.% (blue), 41 wt.% (green), and 57 wt.% (black). Curves recorded in O2 and H2, respectively, saturated 0.1 M KOH at room temperature with a sweep rate of 20 mV s−1 and a rotation rate of 1600 rpm. Loading on the glassy carbon disk in all cases was 15 μgPt cm−2. Potentials were corrected for iR drop (45 Ω) within the electrolyte. Currents are reported as current density per geometric area of disk. Arrows indicate direction of increasing PILBCP thickness.

to the catalyst surface and is a function of polymer content in the catalyst layer and consequently ionomer film thickness, resulting in a decrease in both Jk and Id in the half-cell, as shown in Table 2. Fig. 3(b) contains the HOR polarization curves in H2 saturated 0.1 M KOH at room temperature for 40 wt.% Pt/C nanoparticles as a function of PILBCP content. The HOR diffusion limited currents (Id) for each sample are summarized in Table 2. The HOR Id value for the Pt/C nanoparticles is similar to literature values [47]. Similar to the ORR results, the addition of PILBCP to the Pt/C nanoparticles impacts the diffusion of reactants from the bulk electrolyte to the catalyst surface as quantified by the Id values listed in Table 2. Note, the kinetic current is not reported for the HOR, as even in alkaline electrolyte the activity of Pt/C nanoparticles is such that it is difficult to deconvolute the kinetic limited and diffusion limited regions of the polarization curve. Comparing the uncoated Pt/C (0 wt.% PILBCP) to a PILBCP loading of 23 wt.%, a decrease in Id current by 4% (0.52 to 0.50 mA) was observed. The Id continues to decrease with increasing PILBCP content as shown in Fig. 3 and Table 2. As with the ORR, Id decreases due to the additional diffusional resistance, gaseous and ionic, imparted by the ionomer coating on the Pt/C catalyst. The lower proportional drop in Id with PILBCP catalyst layer content for HOR in comparison to ORR is due to the smaller H2 molecule and its more facile permeation through the PILBCP ionomer film.

Table 2 Oxygen reduction reaction (ORR) mass activities (Jk) measured at 0.9 V vs. RHE, ORR diffusion limited currents (Id), and hydrogen oxidation reaction (HOR) diffusion limited currents measured at room temperature in O2 and H2, respectively, saturated 0.1 M KOH with a sweep rate of 20 mV s−1 and a rotation rate of 1600 rpm; catalyst loading on glassy carbon disk was 15 μgPt cm−2. PILBCP [wt.%]

0 23 41 57 a b c

0.9 V vs. RHE. 0.45 V vs. RHE. 0.6 V vs. RHE.

ORR

HOR

a Jk [mA μg−1 Pt ]

Id [mA]b

Id [mA]c

0.54 0.13 0.09 0.06

1.07 0.98 0.84 0.62

0.52 0.50 0.43 0.37

The impact of the PILBCP AEM polymer on ORR and HOR reaction rates is obvious and significant as demonstrated in Fig. 3. Considering this evidence alone, one could conclude that the low AEMFC performance exhibited in ref. [30] is directly related to the adverse impact of the PILBCP ionomer on catalyst availability/utilization, related to active site blocking and diffusional resistances imposed by the polymer film, consequently reducing the anodic and cathodic reaction rates and resulting in low peak power density. However, if we conduct similar experiments with 33 and 60 wt.% Nafion on Pt/C nanoparticles in 0.1 M HClO4, see Fig. 4 and Table 3, we see a similar trend to that observed with the PILBCP ionomer where the addition of polymer to the catalyst layer results in a significant decrease in activity due to Pt site blocking and an increased resistance to the transport of ions and reactant gases from the bulk electrolyte to the catalyst surface. Additionally, it has been shown that the sulfonate anion on the Nafion side chains can specifically adsorb to Pt, blocking active sites and further reducing the ECSA [42]. Nafion is the most ubiquitously used ionomer and membrane in PEMFCs [13–15] with demonstrated peak power densities greater than 1 W cm−2 [48]. In other words, Nafion adversely impacts the ORR and HOR rates to the same degree as the PILBCP used in this study with an even greater impact on the ORR kinetics due to the sulfonate adsorption, yet PEMFCs with Nafion ionomer in the catalyst layer are able to operate at high power densities [49,50]. As a consequence of the combination of this impact and the necessary ionic properties of the Nafion ionomer, a balance must be found in terms of ionomer content where an optimal value exists in which ionic transport is maximized and diffusional resistances are minimized. For Nafion this value is found experimentally to be about 33 wt.% Nafion in the catalyst layer [49,50]. Similar MEA catalyst layer development needs to be completed for these new AEM ionomers to find the optimum triple phase boundary geometry and structure maximizing both ionic and reactant transport and minimizing the interfacial resistance between catalyst layer and AEM. The correspondence between the half-cell results for the PILBCP and Nafion is an indication that the observed impact of the ionomer on the anodic and cathodic reactions, while needing to be addressed in the future, is not likely the main source of low performance observed for the PILBCP based AEMFC tested in ref. [30]. These results strongly suggest that the interface between the PILBCP ionomer in the catalyst layer and the PILBCP AEM is likely the limiting factor. This

186

J.R. Nykaza et al. / Journal of Electroanalytical Chemistry 783 (2016) 182–187

1

0.5

(a)

60 wt. % Nafion 0 wt. % Nafion

(b)

60 wt. % Nafion

0

33 wt. % Nafion

geo

Current (mA cm

geo

Current (mA cm

-2

)

-2

)

0 wt. % Nafion

0

-1 -2 -3 -4 -5

-0.5

0

0.2

0.4

0.6

0.8

1

1.2

-6

0

0.4

0.6

0.8

1

E (V vs. RHE)

E (V vs. RHE)

(c)

60 wt. % Nafion 33 wt. % Nafion 0 wt. % Nafion

Current (mA cm

geo

-2

)

4

0.2

2

0

-2

-4 -0.2

0

0.2

0.4

0.6

0.8

1

1.2

E (V vs. RHE) Fig. 4. Half-cell reactions for Nafion coated Pt/C catalyst. (a) Cyclic voltammograms (CVs) of 0 wt.% Nafion (red) and 60 wt.% Nafion (black) catalyst layer contents recorded in Ar purged 0.1 M HClO4 at room temperature with a sweep rate of 50 mV s−1. (b) Oxygen reduction reaction (ORR) and (c) hydrogen oxidation reaction (HOR) curves for 0 wt.% Nafion (red), 33 wt.% Nafion (blue), and 60 wt.% Nafion (black) recorded in O2 and H2 saturated, respectively, 0.1 M HClO4 at room temperature, with a sweep rate of 20 mV s−1 and a rotation rate of 1600 rpm. Loading on the glassy carbon disk in all cases was 15 μgPt cm−2. Potentials were corrected for iR drop (25 Ω) within the electrolyte. Currents are reported as current density per geometric area of disk. Arrows indicate direction of increasing Nafion thickness.

issue is potentially related to the difficulties in solubilizing typical AEM polymer chemistries, including the PILBCP studied here, into an ink formulation resulting in poor ionomer dispersions in the catalyst layer and limited contact between the ionomer and AEM. Table 3 Oxygen reduction reaction (ORR) mass activity numbers (Jk), ORR diffusion limited currents (Id), and hydrogen oxidation reaction (HOR) diffusion limited currents measured at room temperature in O2 and H2, respectively, saturated 0.1 M HClO4 with a sweep rate of 20 mV s−1 and a rotation rate of 1600 rpm; catalyst loading on glassy carbon disk was 15 μgPt cm−2. ORR

HOR

Nafion [wt.%]

a Jk [mA μg−1 Pt ]

Id [mA]b

Id [mA]c

0 33 60

0.31 0.004 0.07

1.14 1.03 0.51

0.55 0.46 0.43

a b c

0.9 V vs. RHE. 0.45 V vs. RHE. 0.6 V vs. RHE.

4. Conclusions Rotating disk electrode half-cell experiments were used to determine the impact of a PILBCP ionomer on the ORR and HOR activity of a commercial Pt/C catalyst. CVs in Ar purged electrolytes indicated that the PILBCP reduced the adsorption features of both hydrogen (HUPD) and hydroxide/oxide (OHad/Oad) due to ionic diffusional resistances imparted by the polymer film as well as physical blocking of the platinum sites where polymer loadings beyond 41 wt.% were found to completely cover the Pt/C nanoparticles. The CVs also indicated that there was no specific adsorption, reversible or irreversible, of ions associated with the PILBCP to the Pt surface. The addition of PILBCP to the Pt/C nanoparticles adversely impacted the reaction kinetics (Jk) of the ORR, as well as the diffusion limited current (Id) of both the ORR and HOR. A decrease of 88% and 42% for Jk and Id, respectively, with the highest PILBCP content of 57 wt.% compared to bare Pt/C was observed for the ORR, while a decrease of 29% for Id was observed for the HOR with the same PILBCP loading. While discouraging at face value, nearly identical half-cell reaction impact was observed with Nafion

J.R. Nykaza et al. / Journal of Electroanalytical Chemistry 783 (2016) 182–187

ionomer loadings of 33 wt.% and 60 wt.% in the thin film catalyst layer, yet Nafion based PEMFCs operate with peak power densities greater than 1 W cm−2 [48]. The correspondence between the AEM and PEM ionomer half-cell tests is an indication that while detrimental, the impact of the PILBCP polymer is not likely the sole source of the low peak power density observed with the PILBCP AEM/ionomer fuel cell tested in ref. [30]. This result points to the interface between the ionomer and the AEM as well as the distribution of PILBCP ionomer within the catalyst layer, triple phase boundary, as the likely sources of the poor performance. Future efforts will focus on specific identification/quantification of the interfacial resistance through in situ impedance spectroscopy of the MEA in a single-cell FC stack as well as the development of molecular level insight through IR spectroscopy of the catalyst/ionomer interface as a function of applied potential to aid the optimization of PILBCP based AEMFCs. Acknowledgements This work is supported in part by the U.S. Army Research Office under Grant no. W911NF-14-0310. References [1] G. Couture, A. Alaaeddine, F. Boschet, B. Ameduri, Prog. Polym. Sci. 36 (2011) 1521–1557. [2] Y.A. Elabd, M.A. Hickner, Macromolecules 44 (2011) 1–11. [3] J.R. Varcoe, R.C.T. Slade, Fuel Cells 5 (2005) 187–200. [4] 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, Energy Environ. Sci. 7 (2014) 3135–3191. [5] K.M. Meek, Y.A. Elabd, Macromolecules 48 (2015) 7071–7084. [6] P. Cotanda, G. Sudre, M.A. Modestino, X.C. Chen, N.P. Balsara, Macromolecules 47 (2014) 7540–7547. [7] Y. Ye, Y.A. Elabd, Macromolecules 44 (2011) 8494–8503. [8] O.M.M. Page, S.D. Poynton, S. Murphy, A.L. Ong, D.M. Hillman, C.A. Hancock, M.G. Hale, D.C. Apperley, J.R. Varcoe, RSC Adv. 3 (2013) 579–587. [9] C. Qu, H.M. Zhang, F.X. Zhang, B. Liu, J. Mater. Chem. 22 (2012) 8203–8207. [10] J.R. Varcoe, R.C.T. Slade, Electrochem. Commun. 8 (2006) 839–843. [11] N.J. Robertson, H.A. Kostalik, T.J. Clark, P.F. Mutolo, H.D. Abruna, G.W. Coates, J. Am. Chem. Soc. 132 (2010) 3400–3404. [12] Y. Luo, J. Guo, C. Wang, D. Chu, J. Power Sources 195 (2010) 3765–3771. [13] K.M. Lee, R. Wycisk, M. Litt, P.N. Pintauro, J. Membr. Sci. 383 (2011) 254–261. [14] J.R. Nykaza, Y. Ye, Y.A. Elabd, Polymer 55 (2014) 3360–3369. [15] J.R. Nykaza, Y. Ye, R.L. Nelson, A.C. Jackson, F.L. Beyer, E.M. Davis, K. Page, S. Sharick, K.I. Winey, Y.A. Elabd, Soft Matter 12 (2016) 1133–1144. [16] K.J.T. Noonan, K.M. Hugar, H.A. Kostalik, E.B. Lobkovsky, H.D. Abruna, G.W. Coates, J. Am. Chem. Soc. 134 (2012) 18161–18164.

187

[17] Y. Wang, J. Qiu, J. Peng, L. Xu, J. Li, M. Zhai, J. Membr. Sci. 376 (2011) 70–77. [18] J. Fang, Y. Yang, X. Lu, M. Ye, W. Li, Y. Zhang, Int. J. Hydrog. Energy 37 (2012) 594–602. [19] J. Qiao, J. Zhang, J. Zhang, J. Power Sources 237 (2013) 1–4. [20] Y. Leng, L. Wang, M.A. Hickner, C.-Y. Wang, Electrochim. Acta 152 (2015) 93–100. [21] N. Li, Y. Leng, M.A. Hickner, C.-Y. Wang, J. Am. Chem. Soc. 135 (2013) 10124–10133. [22] D.S. Kim, C.H. Fujimoto, M.R. Hibbs, A. Labouriau, Y.-K. Choe, Y.S. Kim, Macromolecules 46 (2013) 7826–7833. [23] M.H. Robson, K. Artyushkova, W. Patterson, P. Atanassov, M.R. Hibbs, Electrocatalysis 5 (2014) 148–158. [24] X. Lin, Y. Liu, S.D. Poynton, A.L. Ong, J.R. Varcoe, L. Wu, Y. Li, X. Liang, Q. Li, T. Xu, J. Power Sources 233 (2013) 259–268. [25] X. Lin, X. Liang, S.D. Poynton, J.R. Varcoe, A.L. Ong, J. Ran, Y. Li, Q. Li, T. Xu, J. Membr. Sci. 443 (2013) 193–200. [26] J. Ran, L. Wu, J.R. Varcoe, A.L. Ong, S.D. Poynton, T. Xu, J. Membr. Sci. 415–416 (2012) 242–249. [27] K.M. Meek, S. Sharick, Y. Ye, K.I. Winey, Y.A. Elabd, Macromolecules 48 (2015) 4850–4862. [28] K.M. Meek, Y.A. Elabd, J. Mater. Chem. A 3 (2015) 24187–24194. [29] Y. Ye, S. Sharick, E.M. Davis, K.I. Winey, Y.A. Elabd, ACS Macro Lett. 2 (2013) 575–580. [30] J.R. Nykaza, R. Benjamin, K.M. Meek, Y.A. Elabd, Chem. Eng. Sci. 154 (2016) 119–127. [31] Y. Ye, J.-H. Choi, K.I. Winey, Y.A. Elabd, Macromolecules 45 (2012) 7027–7035. [32] R.L. Weber, Y. Ye, A.L. Schmitt, S.M. Banik, Y.A. Elabd, M.K. Mahanthappa, Macromolecules 44 (2011) 5727–5735. [33] M.D. Green, D. Salas-de la Cruz, Y. Ye, J.M. Layman, Y.A. Elabd, K.I. Winey, T.E. Long, Macromol. Chem. Phys. 212 (2011) 2522–2528. [34] J.-H. Choi, Y. Ye, Y.A. Elabd, K.I. Winey, Macromolecules 46 (2013) 5290–5300. [35] L.D. McIntosh, T. Kubo, T.P. Lodge, Macromolecules 47 (2014) 1090–1098. [36] K.M. Meek, J.R. Nykaza, Y.A. Elabd, Macromolecules 49 (2016) 3382–3394. [38] Y. Garsany, I.L. Singer, K.E. Swider-Lyons, J. Electroanal. Chem. 662 (2011) 396–406. [39] Y. Garsany, O.A. Baturina, K.E. Swider-Lyons, S.S. Kocha, Anal. Chem. 82 (2010) 6321–6328. [40] D. van der Vliet, D.S. Strmcnik, C. Wang, V.R. Stamenkovic, N.M. Markovic, M.T.M. Koper, J. Electroanal. Chem. 647 (2010) 29–34. [41] E. Gileadi, S.D. Argade, J.O.M. Bockris, J. Phys. Chem. 70 (1966) 2044–2046. [42] R. Subbaraman, D. Strmcnik, V. Stamenkovic, N.M. Markovic, J. Phys. Chem. C 114 (2010) 8414–8422. [43] R. Subbaraman, D. Strmcnik, A.P. Paulikas, V.R. Stamenkovic, N.M. Markovic, ChemPhysChem 11 (2010) 2825–2833. [44] F. Zhang, H. Zhang, C. Qu, J. Mater. Chem. 21 (2011) 12744–12752. [45] Y. Garsany, J. Ge, J. St-Pierre, R. Rocheleau, K.E. Swider-Lyons, J. Electrochem. Soc. 161 (2014) F628–F640. [46] K. Shinozaki, J.W. Zack, S. Pylypenko, R.M. Richards, B.S. Pivovar, S.S. Kocha, Int. J. Hydrog. Energy 40 (2015) 16820–16830. [47] W. Sheng, H.A. Gasteiger, Y. Shao-Horn, J. Electrochem. Soc. 157 (2010) B1529–B1536. [48] X. Wang, F.W. Richey, K.H. Wujcik, R. Ventura, K. Mattson, Y.A. Elabd, Electrochim. Acta 139 (2014) 217–224. [49] J.D. Snyder, Y.A. Elabd, J. Power Sources 186 (2009) 385–392. [50] G. Sasikumar, J.W. Ihm, H. Ryu, Electrochim. Acta 50 (2004) 601–605.