Effect of Ionomer on
Electrode Performance Durability
Christina Johnston, Zhongfen Ding, Baeck Choi,
and Yu Seung Kim
Los Alamos National Laboratory
MS D429, MPA-11
Los Alamos, NM 87545
Modern proton-exchange membrane fuel cells require
stable performance during long-term operation [1]. The
insufficient durability of Pt-based electrodes remains a
major barrier to commercialization. Although most
strategies to increase electrode durability focus on
preventing the loss of electrochemically-active surface
area (ECSA), we demonstrated recently that performance
loss after potential cycling could be reduced by changing
the dispersing solvent used to prepare the electrode [2, 3].
Because the ECSA did not vary much between electrodes,
the improvement in durability was interpreted as an
increase in the robustness of the ionomer and/or ionomer
interfaces. As a continuation of the strategy of
manipulating the ionomer to improve durability, we report
the effect of varying the ionomer side chain structure.
The cathode electrode for each membrane electrode
assembly (MEA) was prepared from commercial or
house-made polymer dispersion, and a carbon-supported
Pt catalyst (20 wt.%, ETEK). LANL-standard Pt-based
electrode [4] and Nafion
®
212 were used for the anode
electrode and proton exchange membrane, respectively.
Two types of ionomer were compared: short side chain
(SSC) perfluorinated sulfonic acid (PFSA) (Aquivion™,
Solvay Solexis) and long side chain (LSC) PFSA
(Nafion
®
, Ion Power). MEAs were fabricated by LANL
standard decal method [4]. Initial fuel cell performance
and cell durability during potential cycling [5] were
measured.
Figure 1 compares H
2
/air polarization curves of initial
and after 30,000 potential cycles under H
2
/N
2
conditions
for SSC and LSC ionomer bonded electrodes. Electrodes
prepared from SSC ionomer exhibited slightly better
initial fuel cell performance in both kinetic and ohmic
region than the electrode prepared from LSC ionomer,
likely because the higher proton conductivity of low EW
SSC ionomer improves the rate of ORR and lowers
resistive losses. After 30,000 potential cycles, fuel cell
performance decreased in both SSC and LSC-bonded
electrodes. The SSC-bonded electrode exhibited greater
ECSA loss than the LSC-bonded electrode (particularly
during first 1000 potential cycles) which resulted in
slightly better kinetic performance for LSC. Nevertheless,
the performance loss using SSC-ionomer bonded
electrode was much smaller than that using LSC ionomer-
bonded electrode at lower voltages than 0.80 V. Increased
resistance of the aqueous-dispersed LSC-bonded
electrode indicates that disintegration of electrodes
occurred.
Further performance durability improvement of
electrode using LSC and SSC ionomers could be obtained
by changing other variables such as EW or dispersion
solvents, and these factors will be discussed at the
meeting. Combining specific observations from this data
and previous work, especially AC impedance, XRD, and
microscopy measurements, some general concepts
regarding the effect of potential cycling on the ionomer
structure, the ionomer/catalyst interface, and the
electrode/membrane interface will be further developed.
Figure 1. H
2
/air polarization curves of MEAs prepared
from aqueous SSC and LSC dispersion before and after
30000 potential cycling. (H
2
/N
2
potential cycling
condition [5]: cycling potential: 0.6-1.0 V, rate: 50
mV/sec at 80
o
C under fully humidified conditions; H
2
/air
fuel cell test condition: at 80
o
C under fully humidified
conditions, anode/cathode pressure: 30 psig).
Table 1. ECSA and fuel cell performance change of LSC
and SSC PFSA bonded electrode.
LSC
PFSA
SSC
PFSA
SSC
PFSA*
% ECSA loss after 30k 71.3 75.0 66.3
Current density
at 0.9V (A/cm
2
)
Initial 0.042 0.044 0.046
After 30k 0.010 0.005 0.020
Current density
at 0.4V (A/cm
2
)
initial 1.64 1.74 1.77
After 30k 1.28 1.53 1.70
* House-made dispersion
Acknowledgment
The authors thank Drs. C.F. Welch, Andrea Labouriau,
M. Hawley, R.P. Hjelm, and E.B. Orler (Los Alamos
National Laboratory) for instrumental characterization.
We also thank US DOE Fuel Cell Technologies Program,
Technology Development Manager Dr. Nancy Garland,
for financial support.
References
[1] R. Borup, J. Meyers, B. Pivovar, Y. S. Kim, R.
Mukundan, N. Garland et al., Chem. Review, 107 (10),
3904-3951 (2007).
[2] C. M. Johnston, K. Lee, Z. Ding, T. Rockward et al.
Manuscript submitted. (2010).
[3] C. M. Johnston, K. Lee, T. Rockward, A. Labouriau,
N. Mack, and Y. S. Kim, ECS Trans. 25 (1), 1617 (2009)
[4] M. Wilson, S. Gottesfeld, J. Electrochem. Soc. 139,
L28-L30 (1992).
[5] U.S. DOE, Cell Component Accelerated Stress Test
Protocols for PEM Fuel Cells, 2007.
Initial performance
Current density (A cm
-2
)
0.0 0.5 1.0 1.5 2.0
cell voltage (V)
0.0
0.2
0.4
0.6
0.8
1.0
HFR(Ohm cm
2
)
0.0
0.1
0.2
0.3
0.4
0.5
SSC electrode (EW=830)
LSC electrode (EW=1000)
After 30,000 cycles
Current density (A cm
-2
)
0.0 0.5 1.0 1.5 2.0
cell voltage (V)
0.0
0.2
0.4
0.6
0.8
1.0
HFR(Ohm cm
2
)
0.0
0.1
0.2
0.3
0.4
0.5
SSC electrode (EW=830)
LSC electrode (EW=1000)
Abstract #736, 218th ECS Meeting, © 2010 The Electrochemical Society