Gas Transport Resistance in Polymer Electrolyte Thin Films on

Aug 24, 2015 - The oxygen transport resistances of the thin films were characterized by the limiting current density using modified 1 cm2 PEFC hardwar...
0 downloads 11 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Gas Transport Resistance in Polymer Electrolyte Thin Films on Oxygen Reduction Reaction Catalysts Hang Liu, William K. Epting, and Shawn Litster* Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States

Downloaded by FLORIDA INTL UNIV on August 31, 2015 | http://pubs.acs.org Publication Date (Web): August 31, 2015 | doi: 10.1021/acs.langmuir.5b02487

S Supporting Information *

ABSTRACT: Significant reductions in expensive platinum catalyst loading for the oxygen reduction reaction are needed for commercially viable fuel cell electric vehicles as well as other important applications. In reducing loading, a resistance at the Pt surface in the presence of thin perfluorosulfonic acid (PFSA) electrolyte film, on the order of 10 nm thick, becomes a significant barrier to adequate performance. However, the resistance mechanism is unresolved and could be due to gas dissolution kinetics, increased diffusion resistance in thin films, or electrolyte anion interactions. A common hypothesis for the origin of the resistance is a highly reduced oxygen permeability in the thin polymer electrolyte films that coat the catalyst relative to bulk permeability that is caused by nanoscale confinement effects. Unfortunately, the prior work has not separated the thin-film gas transport resistance from that associated with PFSA interactions with a polarized catalyst surface. Here, we present the first characterization of the thin-film O2 transport resistance in the absence of a polarized catalyst, using a nanoporous substrate that geometrically mimics the active catalyst particles. Through a parametric study of varying PFSA film thickness, as thin as 50 nm, we observe no enhanced gas transport resistance in thin films as a result of either interfacial effects or structural changes in the PFSA. Our results suggest that other effects, such as anion poisoning at the Pt catalyst, could be the source of the additional resistance observed at low Pt loading.

■

INTRODUCTION Polymer electrolyte fuel cells (PEFCs) are being developed as a power source for fuel cell electric vehicles because of the high power density and low operating temperature afforded by the polymer electrolytes that they use. Typically, perfluorosulfonic acid (PFSA), such as Nafion, is used for the electrolyte membrane. Because PEFC electrodes are not soaked in liquid electrolyte like many other electrochemical devices, PFSA is also dispersed throughout the porous electrode for the transport of protons to the dispersed catalyst. As Figure 1 illustrates, the Pt or Pt-alloy catalyst is prepared as carbonblack-supported nanoparticles to achieve high surface area and mass activity.1 Thin Nafion films (5−10 nm) coat the catalyst to provide the proton activity and transport for the oxygen reduction reaction (ORR) across the thickness of the cathode. A common direction in PEFC cathode development is to reduce the loading of the costly Pt catalyst.2 A recent, concerning finding is that, as the Pt loading is reduced, a previously unaccounted polarization resistance becomes significant.3−10 This apparent increase in mass transport loss at low loading reduces the power density, offsetting the cost savings from reduced catalyst material. Several groups, including many from the automotive fuel cell industry, have identified that the additional polarization resistance scales with Pt surface area.7−9 For instance, the work of Greszler et al.9 showed that polarization curves from various Pt loadings, with corrections for the established losses, all collapsed onto the same curve once the current density was scaled by electrochemically active Pt surface area. The collapsed © XXXX American Chemical Society

Figure 1. Local oxygen transport resistance within PEFC cathodes is simulated by supporting thin Nafion films on a track-etched PCM with widely spaced cylindrical 10 nm pores. The oxygen transport resistance of the PCM-supported Nafion is quantified using the limiting currrent of a modified 1 cm2 PEFC.

curve featured a notable, additional mass transport loss with low loadings that they attributed to an additional local oxygen mass transport resistance at the ionomer-coated Pt surface. In addition, that work and those of others have shown that the magnitude of additional polarization resistance is independent Received: July 6, 2015 Revised: August 18, 2015

A

DOI: 10.1021/acs.langmuir.5b02487 Langmuir XXXX, XXX, XXX−XXX

Article

Downloaded by FLORIDA INTL UNIV on August 31, 2015 | http://pubs.acs.org Publication Date (Web): August 31, 2015 | doi: 10.1021/acs.langmuir.5b02487

Langmuir

spaced 10 nm pores. The pores are small enough that Nafion chains cannot enter the pores and only forms a film on the surface. In addition, the pores are small enough to mimic the length scales of a single Pt particle coated with a Nafion film. Thus, the open area of the pore and the local oxygen flux simulates the oxygen transport at the surface of a Pt particle. Similar to Suzuki et al.,7 the transport resistance is measured by an oxygen-transport-limited current with the apparatus shown in Figure 1, with the exception that the Nafion film is not in contact with Pt. Although the technique measures the series O2 transport resistance of the PCM and Nafion, the diffusion transport resistance of the PCM is sufficiently low that the measured resistance is dominated by Nafion. We present two versions of our results. First is the O2 transport resistance experienced by the pores, including the converging, threedimensional transport to the small areas of the individual pores. In the second version, the transport resistance is geometrically corrected to the corresponding resistance if the flux was uniform over the area of the Nafion film. This second version is needed to extrapolate any interfacial resistance and to identify any non-geometrical changes in the O2 transport resistance as the films become thinner.

of the gas pressure and not due to molecular gas diffusion. The in situ measurements of Greszler et al.9 and Ono et al.8 estimate the additional, local O2 transport resistance in the range of 10− 20 s/cm. Given the proximity of the polymer film to the catalyst, it is presumed that the additional polarization resistance is due to the film. However, the observed resistance is roughly an order of magnitude larger than expected considering the oxygen permeability of bulk Nafion. It has also been shown that the polarization resistance is not consistent with microstructure transport resistances,9 such as those predicted by agglomerate models.11,12 Thus, one emergent hypothesis is that the structure of the thin Nafion film is different in such a way that it yields lower O2 permeability and higher transport resistance or that there is an interfacial effect that only becomes observable as the films become very thin (e.g., a kinetic limitation to O2 dissolution). This concept is consistent with hypotheses for the ex situ measurements of lower proton conductivity with thinner PFSA films.13,14 However, Greszler et al.9 also suggested that the apparent O2 resistance could actually be the result of first-order sulfonic acid group adsorption onto Pt that interferes with the ORR.15,16 The combined in situ experiments and simplified single-particle model in a separate work by Owejan et al.6 suggests that combined interfacial resistances at Pt|Nafion and Nafion|air interfaces could be responsible for the additional local resistance. Because of significant challenges in controlling and characterizing the Nafion films in the PEFC cathode, several studies have relied on ex situ characterizations of Nafion thin films to resolve thickness-dependent transport properties and changes in morphology, primarily for water uptake17,18 and proton conductivity.13,14 However, ex situ characterizations of oxygen transport through thin PFSA films are still lacking. Suzuki et al.7 performed ex situ characterizations of Nafion permeability as a function of film thickness. In their experiments, the Nafion thin films were deposited onto a Pt working electrode on a planar glass slide and the transport-limited ORR current was used to infer the permeability. Plotting the O2 transport resistance versus thickness showed a roughly linear variation in resistance with thickness, consistent with bulk transport properties. However, extrapolating the linear variation yielded a significant, positive y-intercept value, implying a non-zero O2 transport resistance at zero thickness. Suzuki et al. attributed the no-zero intercept to an interfacial resistance. The span of their fits to the extrapolated interfacial resistance was 17−100 s/cm. The lower bound is consistent with the in situ data of other works, while the upper bound is almost an order of magnitude higher. In any case, their work confirmed an additional interfacial resistance for a Pt electrode covered by Nafion. However, these works cannot distinguish whether this resistance is due to O2 transport into or through the Nafion or whether it is an electrochemical phenomenon at the Pt surface, because all of these studies featured a Nafion film coating a Pt working electrode. The objective of the present work is to identify whether the thin Nafion films on their own, separate from a Pt electrode surface, introduce an interfacial resistance. This paper presents a new method for simulating the local transport at the Pt particle in an ex situ experimental characterization of Nafion O2 transport resistance without the presence of the Pt|Nafion interface. As Figure 1 illustrates, we achieve this by depositing a thin Nafion film onto an inert substrate with widely spaced nanopores. Here, we have used a track-etched polycarbonate membrane (PCM) that has widely

■

MATERIALS AND METHODS

We prepared Nafion thin films by spin-casting Nafion dispersions onto cleaned commercial track etched PCMs with 10 nm diameter pores and a nominal pore density of 6 × 108 pores/cm2 (Structure Probe, Inc., West Chester, PA). The Nafion solution consisted of commercial Nafion dispersions (D 521 and D 2020, Ion Power, Inc., New Castle, DE) that we diluted with varying amounts of a 50:50 volume mixture of isopropyl alcohol and deionized water. Nafion dispersions with loadings of 1, 3, 4, and 20 wt % Nafion were used to prepare Nafion thin films with a large variation in thickness. The cast Nafion ionomer was dried at room temperature and was not annealed. Multiple samples were prepared and characterized at each Nafion loading. Additional samples included a Nafion 211 membrane (Ion Power) and an uncoated PCM. The oxygen transport resistances of the thin films were characterized by the limiting current density using modified 1 cm2 PEFC hardware (see Figure 1). The anode side was unchanged from the standard assembly of the end plate, current collector, and graphite flow plate. The membrane electrode assembly (MEA) of the fuel cell featured in-house-prepared 0.3 mgPt/cm2 gas diffusion electrodes based on SGL 25BC gas diffusion layers (Ion Power) and a 40 wt % Pt/C catalyst (Alfa Aesar, Ward Hill, MA). To reduce the hydrogen crossover current to negligible levels, we prepared the membrane of the MEA by hot-pressing four Nafion 115 membranes on top of each other for a total thickness of 0.5 mm. On the cathode side of the PEFC, the cathode graphite flow plate was replaced with the sample holder and an additional current collector for the cathode of the MEA. The current collector was made of 0.5 mm thick perforated 304 stainless steel having 0.84 mm diameter holes and 28% open area. The sample holder assembly consisted of an acrylic plate to distribute air flow over the PCM-supported Nafion film and gaskets around the sample perimeter to force all O2 consumed by the cathode to pass through the sample film. The Nafion film was placed on the air inlet side to maintain the Nafion water activity at that of the incoming well-mixed, humidified air. A fuel cell test stand (850e, Scribner and Associates, Southern Pines, NC) delivered humidified air and hydrogen to the measurement cell at pressures and flow rates of 1 atm and 100 standard cubic centimeters per minute (sccm), respectively. Unless otherwise specified, the sample temperature was 80 °C and the relative humidity of the air was 68%. Higher relative humidities caused condensation between the sample and the MEA that interfered with measurements. A potentiostat was used for the electrochemical characterization (SP50, Bio-Logic USA, Knoxville, TN). After an initial screening with cyclic voltammetry to identify a repeatable limiting current potential B

DOI: 10.1021/acs.langmuir.5b02487 Langmuir XXXX, XXX, XXX−XXX

Article

Downloaded by FLORIDA INTL UNIV on August 31, 2015 | http://pubs.acs.org Publication Date (Web): August 31, 2015 | doi: 10.1021/acs.langmuir.5b02487

Langmuir range, we obtained the limiting currents from the time-averaged currents for chronoamperometric constant voltage holds at 0.3, 0.25, 0.2, and 0.15 V; this procedure was repeated for each sample. The Supporting Information contains additional details on the limiting current measurements. We measured film thicknesses at each Nafion loading following the electrochemical characterization. We used a transmission electron microscope (TEM, Hitachi 7100, Hitachi Group, Tokyo, Japan) to measure the film thickness from ultramicrotomed cross-sections of the 1, 3, and 5 wt % samples. The 20 wt % film thickness was measured using nanoscale resolution X-ray radiography in Zernike phase contrast mode (UltraXRM-L200, Xradia, Inc., Pleasanton, CA). The measured oxygen transport resistance for the three-dimensional (3D) transport within Nafion to the dispersed 10 nm pores was converted to the corresponding transport resistance of uniform, onedimensional (1D) oxygen flux through a film of the same thickness using a geometric correction factor. We determined the correction factor correlation curve by 3D finite element modeling (COMSOL, version 4.3, COMSOL, Inc., Burlington, MA) of the O2 transport to a single pore in the center of a 408 × 408 nm2 area of the Nafion film (nominal area per pore) with automated parametric variation in Nafion film thickness. Because the diffusion coefficient is significantly lower in the Nafion domain relative to the pore or air, the correction is purely a geometric factor.

the supplier. Although SEM imaging could not verify the small 10 nm pore diameter specification as a result of charging artifacts and beam damage, they roughly appear on the order of 10 nm. To verify our measurement accuracy, we measured the oxygen transport resistance of the NR 211 Nafion membrane and the uncoated PCM and compared them to the expected value. The O2 transport resistance RO2 of the films is calculated from the limiting current I according to the expression RO2 = 4cO2AF/I, where A is the film area, F is Faraday’s constant, and cO2 is the oxygen concentration in the delivered gas. Figure 3a

■

RESULTS AND DISCUSSION In this work and similar to prior work on oxygen transport in thin films, we report only the total transport resistance of the Nafion films and membranes rather than the diffusivity reported in prior characterization of thick Nafion membranes. This is because the calculation of a diffusion coefficient requires an additional characterization of the oxygen solubility to resolve the oxygen concentration in Nafion at the air interface. Fortunately, lumping the solubility and diffusivity together as a total transport resistance does not affect the expected proportional scaling of the resistance with thickness with constant diffusivity nor does it affect the expectation of zero resistance at the extrapolated limit of zero thickness if there is no additional interfacial resistance. We performed scanning electron microscopy (SEM) imaging of the samples after electrochemical testing to ensure both integrity of the film as well as the specified pore density of the commercial PCM film. Figure 2 shows the coated and uncoated

Figure 3. (a) Log−log plot of oxygen transport resistance of the Nafion-coated PCM and Nafion membrane19−22 (NR 211 from the present study) as a function of the thickness. The corrected curve (solid) are the values when resistance is corrected for uniform 1D flux through a film of the same thickness using the correction factors in Figure 4e. The gray dashed line indicates a slope of 1 for the expected proportionality between resistance and thickness. (b) Linear plot of the same uncorrected and corrected oxygen transport resistance values, including the uncoated PCM and an inset near the zero thickness intercept. The range of interfacial resistance estimation from prior work is shown at thicknesses that they were estimated from in experiments: Suzuki et al.7 (hatched bar), Ono et al.8 (black bar), and Greszler et al.9 (white bar).

Figure 2. SEM images of the 1 wt % Nafion-coated (left) and uncoated (right) surfaces of the PCM. The uncoated image shows the sparse distribution of 10 nm through-hole pores.

shows the transport resistances of these membranes. Our NR 211 measurement shows good agreement with prior literature values given the significant variance among those values,19−22 where reported values have been converted to the corresponding transport resistance quantity (RO2) reported here. The measured transport resistance of 8.5 × 103 s/m for the uncoated PCM is shown in Figure 3b and agrees well with the theoretical resistance of 8.4 × 103 s/m for Knudsen diffusion through the 10 nm pores. The theoretical value for the bare PCM based on Knudsen diffusion is given by the equation RO2 = τL/(εdpore(8RT/9πMO2)1/2), where ε is the porosity with a value of 0.047% based on pore density, τ is the estimated tortuosity of 1.07 based on a typical 15° angle off of normal for

sides of the Nafion-coated PCM. The film shown is the lowest Nafion solution loading of 1 wt %, which yields the thinnest films studied here. We observe complete coverage and some apparent roughness in the film, but low-resolution TEM imaging of large sample cross-sections show the thickness variation is only on the order of tens of nanometers. Any prolonged scanning time resulted in beam damage that exposed the pores of the PCM underneath the thin Nafion film. The uncoated image highlights the sparse distribution of the highly aligned, through-hole pores in the PCM. From counting pores in the SEM images, we found the average pore density of three samples to be 6 × 108 pores/cm2, matching the specification of C

DOI: 10.1021/acs.langmuir.5b02487 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Downloaded by FLORIDA INTL UNIV on August 31, 2015 | http://pubs.acs.org Publication Date (Web): August 31, 2015 | doi: 10.1021/acs.langmuir.5b02487

PCM pores, L is the PCM thickness, R is the universal gas constant, T is the temperature, and MO2 is the molecular mass of oxygen. Thus, it is clear that the high diffusion coefficient in the gas diffusion layer of the PEFC MEA and the high Pt roughness factor of the cathode (>100 cmPt2/cmfilm2) mean that the transport resistance of the MEA is negligible compared to that of the PCM, and RO2 of the PCM is, in turn, negligible compared to the Nafion films to be measured. The ratio of the cathode Pt surface area to the film measured is on the order of 200; therefore, the impact of a 5−10 nm Nafion film on the catalyst of the MEA is negligible. Figure 4 shows example images from the nano-computed tomography (nano-CT) (Figure 4a) and TEM (Figure 4c)

In comparison to the corrected data, the uncorrected plots present a higher resistance because of the concentrated flux at each of the 10 nm pores. In addition to the plotted data, the uncorrected data can be converted to an equivalent active site (i.e., pore) area value to yield a resistance specific to simulated active sites rather than the entire membrane area. We do this by multiplying the uncorrected RO2 values by the PCM porosity, yielding values of 9.5, 13, 27, and 45 s/m, respectively, for the four thicknesses given above. For comparison, the transport resistance of the ∼400 μm thick fibrous gas diffusion layer is on the order of 0.1 s/m. The first goal of this work was to determine if there is a significant interfacial resistance to oxygen transport through Nafion films in addition to the diffusion across the film. A second goal was to determine whether the Nafion films exhibit thickness-dependent transport properties. To achieve these goals, our analysis has to eliminate the role of the 3D diffusion path through Nafion from the air to the pore that will artificially increase resistance with thinner films. Figure 4 shows the results of the 3D finite element method (FEM) simulation of oxygen transport to a single pore that are used to compute a correction factor for uniform 1D transport. For the 2 μm film result in Figure 4b, the concentration contours show flux crowding at the 10 nm pore and a transition to the uniform 1D transport through the thickness. For the 200 nm case shown in Figure 4d, the O2 contours show that the flux through the entire film thickness is concentrated over the area of the pore, increasing the effective transport resistance. The correction factor plotted in Figure 4e is calculated by the ratio of the effective diffusivity to the bulk diffusivity, Kgeom = Deff/D, where Deff is calculated from the area-averaged flux at the air interface, thickness, and concentration difference, Deff = ṅO2L/ΔcO2. Toward the limit of zero thickness, the correction approaches the PCM pore volume fraction of 4 × 10−4, whereas correction slowly approaches unity with increasing thickness because the O2 transport resistance of the bulk diffusion outweighs the effective contact resistance of the PCM interface. The correction is significant even for 40 μm thick Nafion films. In addition to the directly measured, uncorrected RO2 values, Figure 3 also shows the O2 transport resistance when corrected for each thickness using the Kgeom factors from Figure 4e. The corrected RO2 values are lower and exhibit a more linear trend with a negligible non-zero resistance y intercept. The gray dashed line in the log−log plot of Figure 3a shows a slope of unity for ideal thickness scaling with constant diffusivity, which aligns well with the Nafion-coated PCM and the membrane measurements of this work and prior work. This shows that our measurements are consistent with prior work and also demonstrates that we do not observe a significant thickness dependence for the oxygen permeability of Nafion. The negligible non-zero resistance intercept when extrapolating the corrected data to zero thickness in Figure 3b shows that there is no observable interfacial resistance over the range of thicknesses that we have characterized. Any additional interfacial resistance that prior work suggests would have been observable as a significant non-zero y intercept in the inset of Figure 3b. In addition, the slight positive curvature of the corrected RO2 data in Figure 3a indicates a slightly reduced diffusivity within thinner films, if any change at all. The curvature is minor enough that we do not conclude that there is a significant thickness effect, only that there is no y intercept

Figure 4. (a) X-ray radiography characterization of the 20 wt % Nafion film thickness. Red dashed lines show Nafion interfaces from the Zernike phase contrast image. (b) Simulation of O2 diffusion in the 2 μm film from the 20 wt % Nafion coating. (c) TEM measurement of the 3 wt % Nafion film thickness. (d) Simulation of O2 diffusion in the 0.2 μm film from the 20 wt % Nafion coating. (e) Correction factors for converting measured resistance to the corresponding value for uniform 1D flux based on a parametric computational study of O2 diffusion through films of varying thickness.

thickness characterizations of the Nafion films being studied. The nominal thicknesses of the Nafion films at the 1, 3, 5, and 20 wt % solutions were 50 nm, 200 nm, 500 nm, and 2 μm, respectively. Table S1 of the Supporting Information lists the details of the thickness measurements. Thus, the thinnest layer is well within the range of thicknesses that Page et al. observe for significant confinement effects in Nafion thin films.17 Figure 3 shows the uncorrected transport resistance RO2 of the Nafioncoated PCM films as a function of thickness as well as the corrected resistance, which will be discussed later. The horizontal error bars illustrate the uncertainty in the thickness, and vertical error bars show the standard deviation of multiple resistance measurements for multiple samples at each thickness. D

DOI: 10.1021/acs.langmuir.5b02487 Langmuir XXXX, XXX, XXX−XXX

Article

Downloaded by FLORIDA INTL UNIV on August 31, 2015 | http://pubs.acs.org Publication Date (Web): August 31, 2015 | doi: 10.1021/acs.langmuir.5b02487

Langmuir

thin as 50 nm and the diffusion coefficient in these thin films does not significantly depart from that of bulk Nafion.

and that the slight positive curvature supports this conclusion even further. Thus, the Nafion films that we characterized here would not introduce the additional local resistance reported in the prior work. For comparison, Figure 3b shows the range of the reported additional Nafion interfacial resistance reported by Suzuki et al.,7 Greszler et al.,9 and Ono et al.8 in the vertical bars. The horizontal axis position of those bars indicates approximately the thinnest film evaluated in those studies. The ex situ work of Suzuki et al. showed a large range for their estimate; however, the lower bound of their interfacial O2 transport resistance is notably larger than our measured resistance at 200 nm. Recall, their work evaluated the resistance for a Nafion film deposited on the surface of a Pt working electrode. The other two estimates are from in operando characterizations of operating PEFCs with 5−10 nm Nafion films coated on a carbon-supported Pt catalyst. In contrast, this work measured RO2 in Nafion films that were not in direct contact with a Pt working electrode, and no additional resistance was observed. We now discuss the origin of the additional local oxygen resistance reported in the prior work in light of the present findings. In the prior work, we observe that the lower bound of ex situ measurements by Suzuki et al.7 agrees well with the in operando measurements of Greszler et al.9 and Ono et al.8 The origin of the wide range of Suzuki et al. data is unknown. The Suzuki et al. films were spin-coated, as were the samples herein; therefore, it is not evident that the spin coating is the origin of our distinct result. One could question whether the additional resistance of the in operando films is due to significant Nafion structural changes in thinner 5−10 nm films. However, the agreement of the ex situ lower bound by Suzuki et al. with the in operando results plus the fact that they measured that resistance and higher values using linear fits from thicker films (165−800 nm) suggest that their additional resistance does not originate from a property change with thinner films. This is consistent with our film thickness range of 50−2000 nm, given the roughly proportional relationship between the film thickness and the oxygen transport resistance. Furthermore, we can compare the present work to prior work on the film thickness effects for proton conductivity13,14 and water uptake.17,18 In those cases, significantly lower proton conductivity and water uptake are observed as a result of rearrangement of the ionomer chains at interfaces that increases the elastic modulus17 and disrupts the conductive ion channels.14 The contrast with the present work suggests that the oxygen permeability of thin PFSA films may be less sensitive to polymer chain arrangement. On the basis of the discussion above, the most significant difference between the prior ex situ and in operando characterizations of an additional Nafion interfacial resistance and our current findings of no additional resistance is the presence of a Pt|Nafion interface in the prior work. Mathematical analysis by Greszler et al. indicated that the additional local transport resistance observed in the experiments could also manifest from a slow, first-order chemical adsorption step in the oxygen reduction reaction. However, it is not clear from the prior or present work if the additional resistance is related to an adsorption mechanism or an additional, unknown transport resistance that evolves at the Pt|Nafion interface. Thus, the key conclusions from this work are that, when there is no Pt|Nafion interface, there is no additional interfacial O2 transport resistance in Nafion films as

■

CONCLUSION By supporting thin PFSA films on inert, nanoporous supports, we have measured the oxygen transport resistance of films mimicking those present in PEFC cathodes. Our parametric study of the O2 transport resistance as a function of the film thickness showed no indication of an interfacial resistance. The observed resistance was roughly proportional to thickness, indicating no significant changes in the O2 diffusion coefficient at thicknesses as thin as 50 nm. This result differs significantly from characterizations of O2 transport in thin films with polarized Pt interfaces, which have shown significant interfacial resistances. This contrast suggests that an electrochemical effect, such as an interaction between the acid site of an ionomer and the polarized catalyst, may play a significant role in this apparent interfacial resistance. In addition to the immediate implications for PEFCs, these results are also an important new data set for the general understanding of thickness effects in polymer electrolyte films.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b02487. Limiting current measurements to determine flux through thin films and Nafion film thickness measurements (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This material is based on work supported by the National Science Foundation under Grant 1053752. The authors also thank Joseph Suhan, Yu-Ting “Tim” Hsu, and Pratiti Mandal for their assistance in electron imaging.

■

ABBREVIATIONS USED PFSA, perfluorosulfonic acid; PEFC, polymer electrolyte fuel cell; ORR, oxygen reduction reaction; PCM, polycarbonate membrane; MEA, membrane electrode assembly; TEM, transmission electron microscopy; SEM, scanning electron microscopy

■

REFERENCES

(1) Litster, S.; McLean, G. PEM fuel cell electrodes. J. Power Sources 2004, 130, 61−76. (2) Gasteiger, H. A.; Kocha, S. S.; Sompalli, B.; Wagner, F. T. Activity benchmarks and requirements for Pt, Pt-alloy, and non-Pt oxygen reduction catalysts for PEMFCs. Appl. Catal., B 2005, 56, 9−35. (3) Yoon, W.; Weber, A. Z. Modeling Low-Platinum-Loading Effects in Fuel-Cell Catalyst Layers. J. Electrochem. Soc. 2011, 158, B1007− B1018. (4) Nonoyama, N.; Okazaki, S.; Weber, A. Z.; Ikogi, Y.; Yoshida, T. Analysis of Oxygen-Transport Diffusion Resistance in Proton-

E

DOI: 10.1021/acs.langmuir.5b02487 Langmuir XXXX, XXX, XXX−XXX

Article

Downloaded by FLORIDA INTL UNIV on August 31, 2015 | http://pubs.acs.org Publication Date (Web): August 31, 2015 | doi: 10.1021/acs.langmuir.5b02487

Langmuir Exchange-Membrane Fuel Cells. J. Electrochem. Soc. 2011, 158, B416− B423. (5) Weber, A. Z.; Kusoglu, A. Unexplained transport resistances for low-loaded fuel-cell catalyst layers. J. Mater. Chem. A 2014, 2, 17207− 17211. (6) Owejan, J. P.; Owejan, J. E.; Gu, W. Impact of Platinum Loading and Catalyst Layer Structure on PEMFC Performance. J. Electrochem. Soc. 2013, 160, F824−F833. (7) Suzuki, T.; Kudo, K.; Morimoto, Y. Model for investigation of oxygen transport limitation in a polymer electrolyte fuel cell. J. Power Sources 2013, 222, 379−389. (8) Ono, Y.; Ohma, A.; Shinohara, K.; Fushinobu, K. Influence of Equivalent Weight of Ionomer on Local Oxygen Transport Resistance in Cathode Catalyst Layers. J. Electrochem. Soc. 2013, 160, F779−F787. (9) Greszler, T. A.; Caulk, D.; Sinha, P. The Impact of Platinum Loading on Oxygen Transport Resistance. J. Electrochem. Soc. 2012, 159, F831−F840. (10) Choo, M.-J.; Oh, K.-H.; Park, J.-K.; Kim, H.-T. Analysis of Oxygen Transport in Cathode Catalyst Layer of Low-Pt-Loaded Fuel Cells. ChemElectroChem 2015, 2, 382−388. (11) Sun, W.; Peppley, B. A.; Karan, K. An improved twodimensional agglomerate cathode model to study the influence of catalyst layer structural parameters. Electrochim. Acta 2005, 50, 3359− 3374. (12) Epting, W. K.; Litster, S. Effects of an agglomerate size distribution on the PEFC agglomerate model. Int. J. Hydrogen Energy 2012, 37, 8505−8511. (13) Siroma, Z.; Kakitsubo, R.; Fujiwara, N.; Ioroi, T.; Yamazaki, S.-i.; Yasuda, K. Depression of proton conductivity in recast Nafion (R) film measured on flat substrate. J. Power Sources 2009, 189, 994−998. (14) Paul, D. K.; Karan, K. Conductivity and Wettability Changes of Ultrathin Nafion Films Subjected to Thermal Annealing and Liquid Water Exposure. J. Phys. Chem. C 2014, 118, 1828−1835. (15) Masuda, T.; Fukumitsu, H.; Kondo, T.; Naohara, H.; Tamura, K.; Sakata, O.; Uosaki, K. Structure of Pt(111)/Ionomer Membrane Interface and Its Bias-Induced Change in Membrane Electrode Assembly. J. Phys. Chem. C 2013, 117, 12168−12171. (16) Subbaraman, R.; Strmcnik, D.; Stamenkovic, V.; Markovic, N. M. Three Phase Interfaces at Electrified Metal-Solid Electrolyte Systems 1. Study of the Pt(hkl)-Nafion Interface. J. Phys. Chem. C 2010, 114, 8414−8422. (17) Page, K. A.; Kusoglu, A.; Stafford, C. M.; Kim, S.; Kline, R. J.; Weber, A. Z. Confinement-Driven Increase in Ionomer Thin-Film Modulus. Nano Lett. 2014, 14, 2299−2304. (18) Kusoglu, A.; Kushner, D.; Paul, D. K.; Karan, K.; Hickner, M. A.; Weber, A. Z. Impact of Substrate and Processing on Confinement of Nafion Thin Films. Adv. Funct. Mater. 2014, 24, 4763−4774. (19) Broka, K.; Ekdunge, P. Oxygen and hydrogen permeation properties and water uptake of Nafion(R) 117 membrane and recast film for PEM fuel cell. J. Appl. Electrochem. 1997, 27, 117−123. (20) Peron, J.; Mani, A.; Zhao, X.; Edwards, D.; Adachi, M.; Soboleva, T.; Shi, Z.; Xie, Z.; Navessin, T.; Holdcroft, S. Properties of Nafion (R) NR-211 membranes for PEMFCs. J. Membr. Sci. 2010, 356, 44−51. (21) Zhang, L.; Ma, C. S.; Mukerjee, S. Oxygen permeation studies on alternative proton exchange membranes designed for elevated temperature operation. Electrochim. Acta 2003, 48, 1845−1859. (22) Sambandam, S.; Parrondo, J.; Ramani, V. Estimation of electrode ionomer oxygen permeability and ionomer-phase oxygen transport resistance in polymer electrolyte fuel cells. Phys. Chem. Chem. Phys. 2013, 15, 14994−15002.

F

DOI: 10.1021/acs.langmuir.5b02487 Langmuir XXXX, XXX, XXX−XXX