The Challenge of Mass Transport Resistance in
Catalyst Layers and Its Impact to Performance
at High Current Density
Di-Jia Liu
Argonne National Laboratory
Presented at DOE CATALYSIS WORKING GROUP MEETING
Wednesday, July 27, 2016
Argonne National Laboratory
Mass Transport Limited Region in Fuel Cells
2
0
0.5
1.0
Theoretical Voltage
Mas
s Tr
ansp
ort
Lim
ited
Reg
ion
Ohmic Region
Current Density K
inet
ic R
egio
n
Ce
ll V
olt
age
Fuel Cell Polarization Curves
Availability → Accessibility → Activity
Reaction Temperature
Cata
lyti
c c
on
vers
ion
Kinetic
Regime
Intra-
particulate
Diffusion
Regime
Bulk
Diffusion
Regime
Heterogeneous Catalytic Conversion
Key Factors Hampering Fuel Cell Performance at
High (or low) Current Domain
Lack of Activity – low turn-over-frequency (TOF)
Lack of Accessibility – active site not near triple-phase boundary
Lack of Accessibility – mass transport resistance
Lack of Availability – insufficient active site density
Lack of Reactant – mass transfer rate < conversion rate
3
Examples of Performance Differences at High Current
Region between PGM and PGM-free Catalyst Electrodes
4
0 500 1000 1500 2000 2500 3000 3500 4000
0.2
0.4
0.6
0.8
1.0
Non-PGM
BASF
Non-PGM
BASF
Current Density (mA cm-2)
Ce
ll p
ote
nti
al (V
)
0
200
400
600
800
1000
1200
Po
wer
Den
sit
y (
mW
cm
-2)
One-bar Oxygen
0 200 400 600 800 1000 1200 1400 16000.0
0.2
0.4
0.6
0.8
1.0
non-PGM
BASF
NON-PGM
BASF
Current Density (mA cm-2)
Ce
ll P
ote
nti
al (V
)
0
100
200
300
400
500
600
700
Po
wer
Den
sit
y (
mW
cm
-2)
One-bar Air
Condition: Pair or PO2= PH2 = 1 bar (back pressure = 7.3 psig) fully humidified; T = 80 °C; N-211 membrane; 5 cm2 MEA; PGM-free cathode catalyst = 3.5 mg/cm2, anode catalyst = 0.3 mgPt/cm2.
Commercial BASF MEA, cathode catalyst = 0.35 mgPt/cm2 , Anode catalyst = 0.2 mgPt/cm2
Challenges in Catalytic Structure using Conventional
Carbon Support
5
Deff = D/; = porosity
= (path l/straight l)2
Gas diffusion inside electrode
1/Deff = 1/Dmicro + 1/Dmeso + 1/Dmacro
Macro Meso
Micro
Catalyst
Micro→meso→macro tiered structure held by vdW force
Mass transfer through propagation of different porosities (Knudsen vs. molecular diffusions)
Charge/heat transfer through particle contact percolation
Challenges in Electrode Architecture using
Conventional Carbon Support
6
Conventional Membrane Electrode Assembly
Electrolyte Anode
GDL Cathode
GDL
H2 e-
e-
H+ H2
e-
e-
O2
H2O
H2
Nonoyama, et. al. JES 158 (4) B416, (2011)
7
Aligned Carbon Nanotube (ACNT) as Catalytic
Support for PEMFC
ACNT MEA
Electrolyte Anode Cathode
H + O2
e-
H2
e-
H+
H2O
Better catalyst utilization
Better support stability
Better electrical & thermal conductivity
Better water management
Better mass transport
Functionalized ACNT as PGM-free catalyst
Advantages of ACNT based MEA
J. Yang, G. Goenaga, A. Call and D.-J. Liu*, Electrochem. Solid-State Lett, (2010) 13 (6) B55-B57,
J. Yang, D.-J. Liu*, N. Kariuki and L, X. Chen, Chem. Comm. 3 (2008) 329 – 331
D.-J. Liu*, J. Yang, N. Kariuki, G. Goenaga, A. Call, and D. Myers, (2008), ECS Trans, 16 (2) 1123-1129
J. Yang and D.-J. Liu*, Carbon 45 (2007) 2843–2854.
D.-J. Liu, J. Yang and D. Gosztola, ECS Transc. (2007) 5(1) 147-154
Straight Spiral
Branched Wavy
8
Process of Fabricating ACNT as Catalyst Support of
MEA
Fabricating ACNT as catalyst support in MEA
“Method of fabricating electrode catalyst layers with directionally oriented carbon support for proton exchange membrane fuel cell”, Liu
& Yang, US Patent 8,137,858
“Catalytic Membranes for Fuel Cells”, Liu & Yang, US Patent 7,927,748
“Aligned carbon nanotube with electro-catalytic activity for oxygen reduction reaction”. Liu, Yang & Wang, US Patent 7767616
“Method of fabricating electrode catalyst layers with directionally oriented carbon support for proton exchange membrane fuel cells”,
Liu, Yang & Wang, US Patent 7,758,921
Forming MEA Growing ACNT Seeding Catalyst
9
Catalyzing ACNT with Pt via Dry and Wet Chemistries
-1.0E-01
-8.0E-02
-6.0E-02
-4.0E-02
-2.0E-02
0.0E+00
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
E (V vs SHE)
I / A
mg
-1 P
t
Commercial Pt/C
Pt/ACNT-4c
TEM Study RDE Study
50 nm
Pt/ACNT prepared through CVD process demonstrated very good
electrochemical activity even with highly hydrophobic surface
Bifunctional solvents needed to avoid nanotube bundling
CVD processes enable >20 wt% Pt incorporated in ACNT
Highly dispersed Pt with particle sizes from 2 nm ~6 nm was observed
Relatively uniform distribution through depth of ACNT layer was observed
10
Cross-section Characterization of ACNT-MEA
SEM Images of ACNT-MEA
ACNT Cathode Layer
Nafion® Membrane Layer
Ink-based Anode Layer
11
Single Cell Test
0
0.2
0.4
0.6
0.8
1
1.2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Current Density, A/cm2
Cel
l V
olt
ag
e, V
olt
s
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Po
wer
Den
sity
, W
/cm
2
Argonne ACNT-MEA
Benchmark - MEA
Single cell test demonstrated improved polarization using ACNT based MEA
5 cm2 Single Cell, T = 75 C,
PH2= PAir=1 Bar,
H2 flow=100 ml/min,
Air flow=400 ml/min
Support Stability Significantly Improved during
Potential Cycling
12
0
20
40
60
80
100
0 100 200 300
Potential cycles
Per
cen
t of
EC
SA
as
of
ori
gin
al,
% ACNT MEARef MEA
Cell cycling potentials 0.5 V to 1.4 V
Improved electrocatalytic surface area retention is due to more stable graphitic ACNT
13
Synthesize TM/N/C active site over carbon nanotubes
ACNT as PGM-free Catalysts
0 1 2 3 4 5 6
ACNT-1
ACNT-2
R (Å)
FT
(
·k2)
Fe-Fe
Fe-N
0 1 2 3 4 5 6
ACNT-1
ACNT-2
R (Å)
FT
(
·k2)
Fe-Fe
Fe-N
A. Titov, P. Zapol, P. Kral, D.-J. Liu, & L. Curtiss, J. Phys. Chem. C. 113, 21629 (2009)
Study N-ligated TM active site structure by XAS and DFT calculation
Shell N R (Å)
Fe-N 4.2 1.95
+ NH3 M +
CH3
CH3 N
N
M
N
N
TM precursor
HC’s N-dopant Active site CNT
J. Yang, D.-J. Liu, N. Kariuki, L. Chen, Chem. Comm. 3, 329 (2008)
14
Probing Active Site Redox Mechanism by In Situ X-ray
Absorption Spectroscopy
It
X-ray
I0
If
APS
Wo
rkin
g
Re
fere
nce
Co
un
ter
In-Situ Electrochemical Cell
O2
D.-J. Liu, J. Yang and D. Gosztola, ECS Transc. 5(1) 147 (2007).
0.0
0.4
0.8
1.2
1.6
7.100 7.110 7.120 7.130 7.140 7.150 7.160
E (keV)
Norm
ali
zed
Ab
sorp
tio
n
E = 1.0 volt
E = 0.55 volt
E = 0.0 volt
1s 3d
1s 4p
-4
-2
0
2
4
6
8
-10010030050070090011001300
E/mV (vs. NHE)
i/m
A
Ar Saturated
O2 Saturated
TM oxidation state and coordination structure under different polarization
potentials revealed O2 binding and conversion kinetics during ORR
R (Å) 0 1 2 3 4 5 6
RD
F A
mp
litu
de
(a
.u.)
E = 1.0 v
E = 0.55 v
E = 0.0 v
Fe-N/O bindings
Fe electronic transitions
NN
Fe+3
NN
NN
Fe+2
NN
O=O
e-H+
NN
Fe+2
NN
O=OH
e-H+
Step 1 Step 2
Design Non-PGM Catalyst using “Support-free” Precursors
15
Catalytic Activity Turn-Over-Freq. x Site Density
• Different transition metals & organic ligands
• Different metal-ligand coordination
• Carbon “Support-free”
• High & uniformly distributed active site density
Conventional ANL’s MOF/POP approach
From carbon supported to “support-free” catalyst
Achievable Current Density @ 0.8 V
(Cathode loading @ 4mg∙cm-2 / 1 bar O2)
- 170 mA/cm2 for 1% Fe, or - 340 mA/cm2 for 2% Fe loading
Critical assumption: a) TM site atomically dispersed & fully utilized b) TOF = 1/10 of Pt (2.5 e-/site.s)
* Gasteiger, et al. Applied Catalysis B: Environmental 56 (2005) 9 R. Jasinski, Nature, (1964) Ma, Goenaga, Call and Liu,
Chemistry: A Euro. J (2011)
“Reality-check” on Non-PGM Activity
i (A/cm2) = 1.6x10-19 x TOF (e-/site∙s) x SD (cm-3) x τ (cm)* From 2D to 3D - Breaking away from 50-year tradition of supported
square-planar precursor
16
MOFs as Precursors for non-PGM Catalysts
“Non-Platinum Group Metal Electrocatalysts Using Metal Organic Framework Materials and Method of Preparation”, D.-J. Liu, S. Ma, G. Goenage, US Patent 8,835,343
.
Synthesis of MOF-based non-PGM Catalyst (Why “burn a perfectly good crystal”?)
TM SBU
MOF Catalyst
Solvothermal Reaction
Heat Activation
Organic Ligand
Advantages of MOF based non-PGM Catalyst
– Highest precursor density for active site conversion
– Well-defined coordination between metal (SBU) & ligand
– Porous 3-D structure with high SSA and uniform micropores
– Large selection of existing MOF compositions
ANL Non-PGM Catalyst IP Portfolio
Developed Fe/N/C decorated CNT catalyst for ORR and demonstrated FeN4 active site structure
– Chem. Comm. 3, 329 (2008) – US Patent 7,927,748 – US Patent 7767616 – US Patent 8,137,858 – US Patent 7,758,921
Developed metal-organic framework (MOF) based non-PGM catalyst and started “support-free” approach
– Chemistry: A European Journal, 17 2063 (2011)
– US Patent 8,835,343 – US Patent Application 20150180045
Developed binary MOF non-PGM catalyst, achieved measured volumetric current density and power density
– Chemical Science, 3 (11), 3200 (2012)
Aligned Carbon
Nanotube
Co-Zeolitic Imidazolate
Framework (ZIF)
Fe-ZIF/ZIF-8 binary MOF
System
ANL Non-PGM Catalyst IPs (continued)
18
Developed non-PGM catalyst using porous organic polymer (POP), achieved peak power density of 730 mW/cm2
– Angew. Chem. Int. Ed. 52 (32), 8349 (2013) – US Patent 9012344 – US Patent Application 20150194681
Developed “one-pot” synthesis approach for low-cost production of a broader range of MOF-base non-PGM catalyst
– Advanced Materials, 26, 1093, (2014) – US Patent Application 20130273461 – US Patent Application 20150056536
Developed the first nano-network non-PGM catalyst, achieved peak power density of 870 mW/cm2
– Proceedings of National Academy of Sciences, 112(34), 10629 (2015)
– US Patent 9,350,026
Fe/Co–POP
“One-pot” Synthesized
ZIFs
Nanofibrous Network
Catalytic Mass/Charge Transport Improvement in
Graphtic Nanonetwork Architecture
19
Conventional Support O2
H+
MOF Catalyst
Nanofibrous Support
O2
H+
Nano- network
Electrospinning Conversion to catalyst
Fuel cell fabrication e-
H+
O2 H2O
Process of Preparing PGM-free Catalyst with Porous
Nano-network Architecture
J. Shui, C. Chen, L. R. Grabstanowicz, D. Zhao and D.-J. Liu, Proceedings of National Academy of Sciences, U. S. A. 2015, vol. 112, no. 34, 10629 D.-J. Liu, J. Shui, C. Chen, “Nanofibrous electrocatalysts”, US Patent 9,350,026
Unique Nanofiber Morphology – High Surface Area
Nearly Exclusively Distributed in Micropore Region
21
Black Pearl
Ketjen Black
Fe/N/CF
BET (m2 g-1) 1432 798 809
Surface Area of Pores ≤ 2nm (m2 g-1)
867 398 780
Surface Area of Pores ≤ 50nm (m2 g-1)
960 507 788
S≤2nm/S≤50nm 90% 78% 99%
Total Volume in Pores ≤ 80 nm (cm³ g-1)
0.955 0.508 0.330
Volume of Pores ≤ 2nm (cm³ g-1)
0.371 0.167 0.271
Volume of Pores ≤ 50nm (cm³ g-1)
0.766 0.463 0.323
V≤2nm/V≤50nm 48% 36% 84%
High specific surface area (BET = 700 ~ 1251 m2/g)
High micro-porosity (Micro-pore vol./Total pore vol. > 80%)
Excellent PGM-free Catalytic Activity Measured by RDE
22
E1/2 = 0.82 V
Ne 4
New Electrode Architecture Led to Very Good
Performance at Single Cell Levels
23
Nanofibrous Network Morphology is Sensitive to
Composition of Electro-spin Mixture
24
14% 0%
39% 24%
Morphology Impacts Electrode Architecture and
Fuel Cell Performance
25
New Nanofibrous Network Showed Promise for
Durability Improvement
26
RDE shows excellent activity retention under multi-potential (0.6 to 1.0 V)
cycling in acidic media
Fuel cell accelerated stress test (AST) shows similar decay rate to that of
commercial Pt/C MEA
Prospects of Porous Nanonetwork as New Support
for PGM Catalyst
Efficient mass transport
Better thermal/electronic conductivity
Stable graphitic network backbone
High specific surface area
Tailored surface compositions to promote catalyst-support interaction
27
Acknowledgement
28
Fuel Cell Research - Shengqian Ma, Dan Zhao, Shengwen Yuan, Jianglan Shui, Gabriel Goenaga, Junbing Yang, Chen Chen, Heather Barkholtz, Lina Chong, Lauran Grabstanowicz, Alex Mason, Brianna Reprogle, Sean Comment, Zachary Kaiser, Debbie Myers
US DOE, US DOE Office of Fuel Cell Technologies and Office of Science. The use of Advanced Photon Source and Electron Microscopy Center are supported by Office of Science, U. S. Department of Energy under Contract DE–AC02–06CH11357.