Fuel Cells for a Sustainable Energy Future

1
Fuel Cellsfor a Sustainable Energy Future

• Sossina M. Haile
• Materials Science / Chemical Engineering
• California Institute of Technology

Is There a Role for Nano-materials?
I-CARES Lecture, November 19, 2009 Washington
University, St. Louis
2
The Problem of Energy

• Growing Environmental Damage
• CO2 levels from 285 to 365 ppm over last century
• Highest in 15 million years
• Diminishing Resources
• Hubberts peak analysis suggests we are past
  peak oil
• Geopolitical Tensions
• US 3 of world petroleum, 25 of consumption
• MiddleEast 65 of world petroleum reserves
• The Energy Need
• 20 TW of carbon-free energy by 2050 (today 16
  TW)
• TW 1012 W

3
Energy Solutions
Courtesy N. Lewis

• Solar
• 1.2 x 105 TW at Earth surface
• 600 TW practical

The need 20 TW by 2050
Wind 2-4 TW extractable
Biomass 5-7 TW gross all cultivatable land not
used for food
Tide/Ocean Currents 2 TW gross
Hydroelectric
Geothermal
4.6 TW gross 1.6 TW technically feasible 0.9 TW
economically feasible 0.6 TW installed capacity
12 TW gross over land small fraction recoverable
Nuclear Waste disposal 60 yr uranium supply
Fossil with sequestration 1 / yr leakage -gt lost
in 100 yrs
4
Sustainable Energy Cycle
H2O, CO2
Carbon-free Source
Solar plantBiomass
Capture
H2
Hydrides? Liquid H2?
Carbon Fuel? CH3OH,CH4
Storage
Delivery
e-
Utilization
H2O
, CO2
Fuel cell
5
Contents

• The Energy Context
• Challenges
• Possible solutions
• Fuel Cell Technology Overview
• Benefits and principle of operation
• Types of fuel cells and their characteristics
• Is there a Role for Nano-materials?
• In solid acid fuel cells
• In solid oxide fuel cells
• Addressing the Front-end of Energy

6
Fuel Cells Part of the Solution?

• High efficiency
• low CO2 emissions
• Size independent
• Various applications
• stationary
• automotive
• portable electronics
• Controlled reactions
• Zero Emissions
• Operable on hydrogen
• (if suitably produced)

Can be as high as 80-90 with co-generation
7
Fuel Cell Principle of Operation
conversion device, not energy source
best of batteries, combustion engines
e-
H
H2
O2
H2 ? 2H 2e-
½ O2 2H 2e- ? H2O
Electrolyte
Overall H2 ½ O2 ? H2O
8
Catalysis
barrier
high energy
low energy
9
Fuel Cell Operation O Electrolyte
e-
O
H2
O2
½ O2 2e- ? O
H2 O ? H2O 2e-
Electrolyte
Overall H2 ½ O2 ? H2O
10
Fuel Cell Performance

• H2 ½ O2 ? H2O
• 1.17 Volts (_at_ no current)
• voltage losses
• fuel cross-over
• reaction kinetics
• electrolyte resistance
• slow mass diffusion
• power IV
• peak efficiency at low I
• peak power at mid I

1.2
0.8
theoretical voltage
cross-over
1.0
slow reaction kinetics
0.6
0.8
Power W / cm2
0.6
0.4
Voltage V
0.4
membrane
0.2
resistance
0.2
slow mass
diffusion
0.0
0.0
0.0
0.4
0.8
1.2
1.6
Current A / cm2
11
Fuel Cell Components

• Components
• Electrolyte (Membrane)
• Transport ions
• Block electrons, gases
• Electrodes
• Catalyze reactions
• Transport
• Ions, electrons, gases
• May be a composite
• (electro)Catalyst
• Conductors
• Pore former

Membrane-Electrode Assembly (MEA)
12
Electrocatalysis Reactions
In state-of-the-art fuel cells, electrocatalysis
sets the performance Specific reaction depends on
nature of mobile species
Electro-reduction rate at cathode ltlt
electro-oxidation rate at anode
13
Reaction Pathways
S. Adler, 2007
triple-point path
electrode bulk path
electrolyte path
side view
top view
active regions
connected in 3-D to the exterior circuit
Motivates a search for mixed conductors
14
From a Single Cell to a Fuel Cell Stack

• Multiple cells
• Connect in series V nVo
• Connect in parallel I nIo
• Requires gas flows to each
• Thermal management
• Greater system complexity than batteries

Courtesy Superprotonics, Inc.
15
Fuel Cell Types
Types differentiated by electrolyte, temperature
of operation
Portable
Stationary
? Corrosive liquids
Fuel flexibility, efficiency
Easy thermal cycling
Target regime
16
Fuel Cell Electrolytes
H
O
17
Proton Transport in Superprotonic Solid Acids
H
Anion group (XO4)reorientation 10-11 seconds
X
O
Proton transfer 10-9 seconds
18
Fuel Cell Operation
H2, H2O cell O2, H2O
T 248C 8 mg Pt/cm2
Slurry deposit
T. Uda S.M. Haile, Electrochem Solid State
Lett. 8 (2005) A245-A246
10-40 mm pores, 40 porosity
Open circuit voltage 0.9-1.0 V Peak power
density 285-415 mW/cm2
19
Identifying the Rate-limiting Step
Fuel cell polarization curve
25 mm electrolyte minimal contribution to
voltage loss
O2,Pt CsH2PO4 Pt, H2
20
Identifying the Rate-Limiting Step

• Impedance measurement under uniform gas
  atmosphere
• Small a.c. perturbation

Example hydrogen electro-oxidation
H2 O ? H2O 2e-
material
material
Impedance ? Resistance Resolve different
processes with different time constants
21
Rate-limiting Step Oxygen Reduction
Impedance spectra symmetric cells
under O2 (with H2O)
under H2 (with H2O)
5 Wcm2 per electrode
0.03 Wcm2 per electrode
22
Nano to the Rescue?
Target
SEM imaging shows
CsH2PO4
micrometric
Pt/C
nanometric
Classic triple point path for electrocatalysis

• Two approaches to nanometric CsH2PO4
• Microemulsion
• Electrospray

(Aerosols pursued by Superprotonic)
23
Enhancing Cathode Electrocatalysis
CsH2PO4
micrometric
isolated electrolyte particle
catalyst
3 mm
Pt/C
nanometric
nano to the rescue?
Classic triple point path for electrocatalysis

• Approaches to nanometric CsH2PO4
• Microemulsion
• Electrospray

(Aerosols pursued by Superprotonic)
Image courtesy Superprotonic, Inc., Pasadena, CA
24
Electrospray Nano-CsH2PO4 Synthesis
w/ Costas Giapis

• . Heat ropes
• . Electrospray chamber
• . Ammeter
• . High voltage source
• . Capillary

• . Nitrogen inlet
• . Spray
• . Substrate
• . Nitrogen outlet
• 0. Resistor (1MO)

Suitable to water soluble CsH2PO4
after one week
Electrospray conditions ? 50mol H2O CH3OH
? 0.5g CDP in 100ml solution ? Chamber
temperature 120C ? Nitrogen gas flow
1000ccm ? Capillary to substrate 2.5cm
Wilm MS, Mann M, Int. J. Mass Spectrom. Ion
Proc., 136 (1994) 167-180
Extremely wide parameter space Unstable
structures, not yet nano
25
Stable Nanoscale Structures
14 days, 250C, 0.3 atm H2O
co-spray with Pt black (Pt on C, CNT)
suspend Pt in PVP 100 nm features
10 mm
1 mm
after oxygen plasma treatment
distinct PVP peak removed
2 mm
26
Stable (but poor) Electrocatalysis
press
after 10 hr test
CsH2PO4
0.09 mg Pt/cm2
2 mm
nanostructuring retained
under H2 (with H2O)
10 Wcm2
previous 0.03 Wcm2 15 mg Pt/cm2
27
Do Nano-particles Really Matter?
500 nm
1 mm
10 mm
2.2 m2/g
5.7 m2/g
0.8 m2/g
5 mm
5 mm
5 mm
Courtesy Superprotonic, Inc., Pasadena, CA
28
Confinement Effects to Control Tc?

• High operating temperature combined with
  dehydration tendency
• Implies energetically expensive humidification
• Lower Tc by nanoscale confinement?
• Confine within linear pores of anodic aluminum
  oxide
• Coat with SiO2 to prevent reaction
• Variable pore diameter and spacing achievable
• In-house prepared AAO
• Interaction with SiO2 can also enhance
  conductivity

29
(No Transcript)
30
Silica coating of membrane
Pull vacuum dry
Anneal for 1 hr at 150 C
50 5 1 ethanol TEOS 1M HCl
Infiltrate with CsH2PO4
Dissolve Al2O3 in H3PO4
2 min sonication
30 min hydrolysis
Characterization SEM TEM
Hydrolysis
Condensation
31

• Wall Thickness 16 nm
• Pore Size 32nm

Silica coating of membrane (SEM TEM)
32
Infiltration by CsH2PO4
CsH2PO4

• Melt Under High Humidity
• (280C, 1hr, Ar (30sccm))

Top face
Bottom face
33
From Breakthrough to Product
2007
Calum
2001
1 mg Pt for 200 mW 2.5 g Pt for 60 W bulb 100
in Pt
Dane
1 cm2 of fuel cell area 36 mg Pt for 10 mW 43 g
Pt for 60 W bulb 9,000 in Pt
2008
20 cell stack 60W net
Tom Friedman talking to his wife on an SAFC
powered cell phone
MVI_1924.avi
34
Intermission

• Solid Acid Fuel Cells
• Performance limited by electrocatalysis
• Nanostructuring may lead to higher power output
• Several routes
• Electrospray scalable, stable with additives,
  direct electrode fabrication (deposition onto
  current collector)
• Preliminary measurements
• Excellent stability
• Not yet the high performance expected
• Nanoconfined electrolyte for thermal stability
• Effect yet to be confirmed
• Now on to SOFCs

35
State-of-the-Art SOFCs
Component Materials
cathode (air electrode)
(La,Sr)MnO3
Zr0.92Y0.08O2.96 yttria stabilized zirconia
(YSZ)
electrolyte
Ni YSZ composite
anode (fuel electrode)

• Cathode typically exhibits highest losses
• Prepared as thinnest component
• Murky literature on impact of cathode composition
  on O2 electroreduction rates
• General but not detailed mechanistic understanding

36
New Cathodes for Solid Oxide Fuel Cells
(Ba0.5Sr0.5)(Co0.8Fe0.2)O2.4

• Traditional cathodes
• A3B3O3 perovskites
• Poor O2- transport
• Limited reaction sites
• Our approach
• High O2- flux materials
• Extended reaction sites
• A2B4O3 perovskites

almost 1 in 5 vacant
electrode bulk path
triple-point path
O2
O2
Oad
Oad
cathode
2e-
cathode
O2-
2e-
Oad
O2-
O2-
electrolyte
electrolyte
37
Oxygen Flux and Non-stoichiometry
Ba,Sr
(Ba0.5Sr0.5)(Co0.8Fe0.2)O3
Co,Fe
O

• 1 in 6 oxygen sites is vacant!!
• No sign of vacancy ordering
• Co must be 3 and 2

38
Vacancy Diffusion Coefficient
100x higher than other materials
similar enhancement in ksurf
39
Fuel Cell Operation
H2, 3 H2O fuel cell Air
0.71 cm2
ceria electrolyte Ni anode
gt 1 W/cm2 at 600C!!!
1.3 cm
Now anode is rate-limiting component!
40
Hydrogen electro-oxidation
Cell Metal Sm-doped ceria Metal
very different metals, similar behavior
open Pt closed Au
connected to electron concentration?
electronic conductivity?
-¼
W. C. Chueh, W. Lai and S. M. Haile, Solid State
Ionics (2008).
41
Hydrogen electro-oxidation

• H2 O ? H2O 2e-
• Independent of metal
• Depends on pO2 like n
• Hypothesis
• Reaction on ceria
• Rate-limiting step connected to electron
  concentration
• Metal not required
• Most likely
• Both surface and TPB paths
• Architecture determines dominant pathway

surface reaction
H2
H2O
M
M
ceria
O
e
electron migration
W. Lai and S. M. Haile, J. Amer. Cer. Soc. 88,
29792997 (2005).
Electrolyte path dominates
42
Electrocatalysis at Metal or Oxide?
650 C, pH2 0.049 atm, pH2O 0.020 atm, pO2
1.2 ? 10-25
lTPB 5 m/cm2
16 kW cm2
6 kW cm2
7 Wcm2
normalized to SDC/YSZ interfacial area
1,000x less resistive
Metal need not be exposed! Electron transport is
not rls
43
Target Anode Architecture
nanowire forest
2-d metallic network
ceria nano-coating
metal nanowire
ceria electrolyte

• Triple points not required
• Flexibility in choice of metal
• Oxide coating stabilizes structure
• Ceria catalytic for carbon containing fuels

44
Routes to 2-D Metallic Networks

• Self-assembling sacrificial templates

(4) Remove polysterene template
(1) Monolayer of PS beads 2 mm
(2) Oxygen plasma etch
(3) Metal deposition (evaporation)
Base for nanowire growth
45
Fabrication Flexibility
500 nm beads
790 nm beads

• Smaller starting bead size
• 500 790 nm
• Deposition method spin coating
• Final pore sizes 250 500 nm

• Larger starting bead size
• 1.0 3.2 microns
• Deposition method
• spin coating water wash
• Final pore sizes 1 2.5 microns

2 mm beads
2 mm beads
3.2 mm beads
3.2 mm beads

• Metals
• Ni, Au, Ti, Ti/Au, Cu, Pt

Perfection in ordering not required
46
Intermission

• Ceria based solid oxide fuel cells
• BSCF is an exceptional cathode electrocatalyst
• Connected to high oxygen non-stoichiometry
• High oxygen diffusivity surface rxn constant
• Surface reaction limited
• Ceria as a hydrogen electroxidation catalyst
• Higher activity than Pt
• Insight suggests atypical anode geometry
• Still a role for triple phase boundaries?
• Now, on to the front end of energy

47
A Convenient Thermochemical Cycle
Thermal Reduction
Oxidation
TH
MO2-d
H2O, CO2 (N2)
MO2
H2, CO, CH4, etc.
O2
TL
Metal oxide releases/incorporates oxygen No phase
change, large nonstoichiometry range Rapid
kinetics bulk diffusion, surface reaction
Ideal candidate ceria
48
Predicted Oxygen Release / Fuel Production
inert gas
TH 1500C d 0.05
Theoretical fuel production 7 mL H2 (STP) /
gram of ceria / cycle
Predict 3.5 mL O2 g-1
air
TL 800C d 0.00
49
Fuel Production
H2O dissociation
800 ºC
CO2 dissociation
800 ºC
No solid carbon formation!
pH2O 0.064 atm, FRtot 380 sccm g-1SDC
pCO2 0.032 atm, FRtot 300 sccm g-1SDC
Complete utilization of ceria non-stoichiometry
for fuel production
SDC samaria doped ceria
50
Fuels of Choice from H2O CO2?
500 C
500 C
Add Ni catalyst
pH2O 0.064 atm pCO2 0.032 atm FRtot 10 sccm
g-1SDC
due to transient carbon deposition
51
Conclusions

• Continued fuel cell development will largely
  depend on enhanced electrocatalysis, less so on
  new electrolytes
• Optimize architecture as well as material
  properties
• Mixed ion and electron conductors play a key role
• BSCF cathode almost eliminates cathode
  overpotential, but longevity may be a problem
• Ceria as an anode component greatly lowers anode
  overpotential
• The search is on for mixed H/e- conductors for
  solid acid cathodes
• Variable-stoichiometry, mixed conductors (ceria)
  may also play an important role in solar fuel
  production
• Rapid kinetics, high cyclability, fuels of
  choice, acceptable efficiency
• Much remains to be done!

52
Summary Conclusions

• Sustainable energy is the grand challenge of
  the 21st century
• Solutions must meet the need, not the hype
• Fuel cells can play an important role
• Nanostructured electrodes
• Possible at warm temperatures w/solid
  electrolytes
• May solve electrocatalysis challenges
• Solid acid fuel cells cathode is rate-limiting
• Solid oxide fuel cells anode is rate-limiting
• Still plenty of need for fundamental research

The stone age didnt end because we ran out of
stones. -Anonymous
53
The People Agencies

• Current Students

Eugene
Drew
Ayako
Kenji
Evan
Chatr
Áron
Rob
Chi-khai
Mary
William
Danien
Tae-Sik

• Current Post-docs

• Former, who contributed to results

Zongping
Tetsuya
Marion
Yoshi
Yong
Calum
Jianhua
Wei
Jaemin
Dane

• NSF, Stanford GCEP

• Collaborators Goodwin, Giapis

54
Microemulsion Approach
Water-in-oil droplets (w/o) as nanoreactors
Tail Hydrophobic
Head Hydrophilic
Non-ionic surfactant Brij30 Polyethylene
glycol dodecyl ether CH3(CH2)mCH2(OCH2CH2
)nOH Organic solvent Heptane
Surfactant
Micelle
Hydrophilic head
Reversed (inverted) Micelle
complex phase behavior
Organicsolvent
Aqueous phase
Hydrophilic head
Organic solvent (oil)
Hydrophobic tail
Hydrophobic tail
Oil
Water
Aqueous phase 3 10 nm in diameter
water-in-oil microemulsion phase
oil-in-water microemulsion phase
Clear, thermodynamically stable, isotropic liquid
55
Microemulsions as Nanoreactors
Surfactant
Oil
Nanoparticle
Water Reactant 1
Challenges for CsH2PO4 water soluble
product sensitive to impurities high
Na present
Droplet fusion
Water Reactant 2
56
Non-aqueous Route to CsH2PO4
Cs-acetate (ethanol) H3PO4 (ethanol) ? CsH2PO4
(solid) acetic acid (ethanol)
Phase Diagram Study
Confirm outside of microemulsion
Surfactant
(H2O)
8 nm droplets (DLS)
Is ethanol a co-surfactant??
aqueous
Oil
57
It works!
Cs 0.1 M H3PO4 0.2 M
0.8 M solutions
Sort of .
100 nm
300 nm
2 mm
Regions of highly monodisperse, 60 nm
particles Other regions of large particles ?
growth due to moisture? Not readily scaled for
fuel cell electrodes ? surfactant removal??
58
Routes to Metallic Nanowires
Metal (Cu) nanowires on SDC
oxidize, then reduce
CuO nanowires on SDC

• Literature oxidation of Cu ? CuO nanowires
• Here
• Thermal evap Cu onto polycrystalline ceria
• Oxidize Cu to obtain CuO nanowires
• 425 ?C
• Reduce in H2 plasma
• Gentle retains morphology
• Harsh changes morphology
• More general
• Anodic alumina

Cu
SDC
(furnace oxidation)
2 mm
CuO
SDC
Cu nanowires on Cu on SDC
(gentle H2 plasma)
Cu
2.31 mm
SDC
10 nanowires/mm2 ? 140 area
Base for ceria coating
59
Hydrogen electro-oxidation
Impedance spectra Me ceria Me
fit to
open Pt closed Au
-¼
very different metals, similar behavior
connected to electronic conductivity?
W. C. Chueh, W. Lai and S. M. Haile, Solid State
Ionics (2008).
60
Hydrogen electro-oxidation

• Independent of pH2 and pH2O, beyond setting pO2

open const. pH2 0.022 atm closed const. pH2O
0.16 atm
-¼
DH 2.71 ? 0.05 eV
61
Observations of Climate Change

• Evaporation rainfall are increasing
• More of the rainfall is occurring in downpours
• Corals are bleaching
• Glaciers are retreating
• Sea ice is shrinking
• Sea level is rising
• Wildfires are increasing
• Storm flood damages are much larger

62
(No Transcript)
63
Greenland Ice Sheet Melt Extent
64
More Observations of Global Climate Change
1910
Grinnell Glacier and Grinnell Lake Glacier
National Park
Coral Bleaching
1977
65
Future Scenarios
Courtesy John Seinfeld
Most optimitistic scenario
66
Future Scenarios
Courtesy John Seinfeld
Highly optimitistic scenario stabilize at 380 ppm
(aerosols)
67
World Energy Consumption
2005 totals 490 Q-Btu, 515 EJ, 16TW 2030
projections 720 Q-Btu, 760 EJ, 24TW
86 fossil
81
Source US Energy Information Administration
68
World Energy Consumption
How good is the EIA at making projections??
(annual)
Source US Energy Information Agency
Coal, actual
69
Fossil Fuel Supplies
Source US Energy Information Administration
Rsv Reserves (90) Rsc Resources (50)
56-77
287-345
70
Reserves History for American Coal
Courtesy David Rutledge
Coal Commission
(based on surveys by Marius Campbell
of the USGS)
4,045 years
Paul Averitt (USGS)
2,136 years
1,433 years
Bureau of Mines/EIA (based on Paul
Averitts surveys)
270 years
236 years
368 years
Hubbert Peak type of analysis suggests 90
depletion by 2076
71
US Energy Imports/Exports 1949-2004
Source US Energy Information Administration
Imports
Exports
35
6
Total
30
5
25
4
Total
20
Quad BTU
Coal
3
Quad BTU
15
2
10
Petroleum
1
Petroleum
5
0
0
1950
1960
1970
1980
1990
2000
1950
1960
1970
1980
1990
2000

• 65 of known petroleum reserves in Middle East
• 3 of reserves in USA, but 25 of world
  consumption

Net
35
30
25
20
Quad BTU
15
10
5
0
1950
1960
1970
1980
1990
2000
72
Environmental Outlook
Global CO2 levels
2008 385 ppm Projections 500-700 ppm by 2020

• Anthropogenic
• Fossil fuel (75)
• Land use (25)

Industrial Revolution
Source Oak Ridge National Laboratory
73
Environmental Outlook
Intergovernmental Panel on Climate Change, 2001
http//www.ipcc.ch N. Oreskes, Science 306,
1686, 2004 D. A. Stainforth et al, Nature 433,
403, 2005
74
Caltech Center for Sustainable Energy Research
Electricity
Photovoltaic and photolysis power plants
H2O, CO2
Fuel H2 or CH3OH
Fuel cell power plant
H2O, CO2
Harry Atwater Harry Gray Sossina Haile Nathan
Lewis
Electric power, heating
75
Fuel Cell Design Philosophy

• Solid electrolyte and warm temperature
• Combines benefits of rapid catalysis and ease
• Why a solid?
• Non-corrosive, mechanically confined
• Potential for nanostructured electrodes
• Why warm temperatures?
• Avoid Pt catalysts, relaxes material constraints
• Potential for nanostructured electrodes
• SOFCs and PEMs
• High temperature
• Wet polymer
• Additionally, require a robust electrolyte
• Fewer engineering work-arounds for operation

? particle coarsening