Post on 31-Mar-2020
transcript
Membrane Development for Hybrid Sulfur Electrolysis and
Oxygen SeparationMichael A. Hickner1, Andrea Ambrosini1, Michael Hibbs1,
Kirsten M. Norman1, Cy H. Fujimoto1, Christopher J. Cornelius1, Milton E. Vernon2, Fred Gelbard2 Paul Pickard2,
1 06338 Fuels and Energy Transitions2 06771 Advanced Nuclear Concepts
Sandia National Laboratory
Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company,for the United States Department of Energy’s National Nuclear Security Administration
under contract DE-AC04-94AL85000.
This presentation does not contain any proprietary, confidential, or otherwise restricted information
Project ID #PDP32
Sulfur Based Thermochemical CyclesFor Nuclear Hydrogen Production
(Albert C. Marshall SAND2002-0513 February 2002 An Assessment of Reactor Types for Thermochemical Hydrogen Production)
The Sulfur based cycles –- Sulfur-Iodine - Hybrid Sulfur
are the focus of the current NHI research program.
These cycles are the most technically developed of the more that 200 cycles reviewed and have the potential for high efficiencies.
The Nuclear Hydrogen Initiative is investigating thermochemicalcycles as one of the promising method for hydrogen production using Generation IV reactors.
Sulfur-Iodine versus Hybrid Sulfur Cycle
Hybrid-Sulfur (thermal and electrochemical reactors)advantages disadvantages- fewer reactions - electrochemical reactor cost- no HI or I2 - electricity needed (efficiency??)
SNL Membrane Approach for Efficiency and Process Improvements
Develop new high-temperature (120-150°C) proton exchange membranes with high conductivity and low SO2crossover for efficient electrolysis
Sandia-synthesized polymer membranes have shown promise in high temperature electrochemical processes, e.g. fuel cells.
Sulfonated membranes with high temperature capability are being tested under a variety of conditions in an SO2electrolysis cell. Conditions of the process unit are being optimized and efficiency/lifetime is being measured.
Synthesize new oxygen anion conducting ceramic membranes for high temperature oxygen separation.
Sandia has proven capability in novel ceramic materials and a wide-ranging program in membrane separations.
Oxygen anion conducting ceramics are being tested for their stability in the high temperature reactor environment and their separation characteristics.
High temperature thermal reactor Proton exchange membrane electrochemical reactor
Ceramic Oxygen Separation Membranes
Characteristics:•Dense Ceramic Membrane– separates via ion conduction, not size exclusion• Self-supporting• Mixed ionic-electronic conductor– ionic component allows for conduction of oxide anion while electronic component eliminates the need for an applied potential
High pO2(sweep side)
Low pO2(feed side)Membrane
O2-
e-
O2H2SO4, O2
H2O, SO2
Perovskite ABO3
• Stable at high temperatures
• Amenable to doping and substitution by a variety of cations on both the A- and B-sites
• Can stabilize oxygen nonstoichiometries
• Mixed ionic-electronic conductivities
• Known membrane materials
Nitrate synthesis: • Nitrates of starting materials dissolved in DI H2O• Citric acid added• Sol’n heated at 90 ºC to evaporate H2O• Resulting gels dried overnight then self-ignited at 400 ºC• Powder ground up in mortar and pestle• Sintered at 1250 ºC for 24 hr
Characterization:• Powder x-ray diffraction (PXRD)• Thermogravimetric analysis (TGA)• Four-probe conductivity• Scanning electron microscopy/electron dispersive spectroscopy• Permeation measurements
Synthesis and Characterization
PXRD of La0.1Sr0.9Co0.7Mn0.3O3-δ
Space group Pm-3mSpace group Pm-3m
Rf = 4.78Chi2 = 4.44
TGA Cycle of La0.1Sr0.9Co0.7Mn0.3O3-δ
• The first part of the graph shows the weight change as the temperature is cycled between 50-850 °C, under a constant flow of O2 gas. This describes an easily reversible temperature-swing adsorption/desorption of oxygen. • The second part of the graph illustrates the reversible weight change as a function of oxygen partial pressure, by cycling the gas between O2 and Ar at a constant temperature of 850 °C. This implies that the material can transport oxygen across a membrane by pressure differential. • X-ray diffraction of the material after TGA cycles shows little/no change in structure which illustrates the stability of the structure.
-0.9399% (-0.5692mg)
-0.9574% (-0.5798mg)
0.9484% (0.5743mg)
-0.06516% (-0.03946mg)
0.8816% (0.5339mg)
0.9010% (0.5456mg)
0
200
400
600
800
1000
Tem
pera
ture
(°C
)
98.0
98.5
99.0
99.5
100.0
100.5
Wei
ght (
%)
0 200 400 600 800 1000 1200 1400
Time (min) Universal V2.5H TA Instruments
Temperature cycle 50 – 850 ºC under flowing O2
Pressure cycle between Ar and O2 at 850 ºC
Ar Ar Ar
O2 O2 O2
Four-Probe Conductivity of LSCM
• Conductivity is several orders of magnitude better than YSZ • Conductivity ↑ as temperature ↑ and pO2 ↑• Large magnitude implies electronic conductivity contribution• Ionic contribution between 0.2 – 0.4 S/cm at 850 °C
0.001
0.01
0.1
1
10
100
1000
400 500 600 700 800 900 1000Temperature (C)
Log
Con
duct
ivity
(S/c
m)
LSCM1973LSCM1991YSZLSCF2882
vs. Temperature (in air)
0.001
0.01
0.1
1
10
100
1000
1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00
Log pO2 (atm)Lo
g C
ondu
ctiv
ity (S
/cm
)
LSCM1973LSCM1991YSZLSCF2882
vs. pO2 (at 900 °C)
Garino, SNL
Oxygen Permeation Unit
Constructed of Inconel 600 Ni alloyConstructed of Inconel 600 Ni alloy
membranemembrane
Mass flow controller (air)
Mass flow controller (air)
thermocouplethermocouple
to µGCto µGC
Permeation Unit - Design
air
air outto micro GC
He O2
thermocouple tube furnace
Cu O-ring seal
¾ ″½ ″
⅛ ″⅛ ″
membrane
Permeation of LSCM Membranes
0.000
0.200
0.400
0.600
0.800
1.000
1.200
1.400
1.600
780 800 820 840 860 880 900 920 940 960
Temperature (C)
Oxy
gen
Perm
eatio
n (m
l/min
.cm
2 )
LSCM1973LSCM1982LSCM1991
Eltron Research Inc.
Mini H2SO4 Decomposition Reactor
F. Gelbard, SNL
Goal: To test the stability of the membrane under “reactor” conditions
10 20 30 40 50 60Two-Theta (deg)
0
500
1000
1500
Inte
nsity
(Cou
nts)
[AA2-58-2_postSO2.RAW] LSCM1982 post-H2SO4 decomp, 3h/850C 89-5717> (La0.6Sr0.4)CoO3 - Lanthanum Strontium Cobalt Oxide74-2035> Celestine - SrSO4
34-0328> SrS2 - Strontium Sulfide
Post-H2SO4 Decomposition
• Microscopy of cross-section reveals corrosion layer of approx. 5 µm • XRD (above) shows formation of SrSO4 and possibly SrS• Not yet known if corrosion caused by exposure to H2SO4, SO2, or both• Membrane can be regenerated by heating to 1300 °C under O2
Summary• The La0.1Sr0.9Co1-yMnyO3-δ (LSCM) family shows promise for use as
ceramic high-temperature oxygen separation membranes• The materials are robust under varying pO2 and show reversible oxygen
sorption properties at 850 ºC• Permeation measurements show the membranes are oxygen permeable• Preliminary stability tests show at least some corrosion occurs upon
exposure to the H2SO4 decomposition stream at 850 ºC
Ongoing Work• Ongoing high temperature permeation studies (SNL)• Continued structural elucidation• Determine extent of corrosion during H2SO4 decomposition and possible
mitigation steps• Continue membrane development (density, processing, scale-up)• Testing on actual decomposition reactor
High Temperature Polymeric Proton Conducting MembranesSulfonated Diels-Alder Poly(phenylene) - SDAPP
nSO3H
HO3S• Thermal Stability• Good Chemical Stability• Chemical Diversity• Compositional Control• Ion Conductivity• Morphology
Polyphenylenes are a chemically, thermally, and mechanically stable backbone upon which to build a library of membranes (with both cation and anion fixed sites) for application to a large array of membrane-based processes such as fuel cells, water desalination, electrodialysis, etc.
Fujimoto, C.H, Hickner, M.A., Cornelius, C.J., Loy, D.A. Macromolecules 2005, 38(12), 5010-5016.
Electrolyzer Schematic
SO2 H2O
flow field flow fieldMEMBRANE
H+
H2H2O
H2O
Membrane Electrode Assembly Membrane Electrode Assembly –– MEAMEAMembraneMembraneCatalyst layersCatalyst layersGas diffusion layersGas diffusion layers
e-
SOSO22 + 2 H+ 2 H22O O + electric power+ electric power →→ HH22SOSO44 + H+ H22 + heat+ heatMembrane is critical for:• low SO2 crossover• efficient water transport• high temperature operationSNL Membrane PerformanceTarget: 0.5 mA/cm2 at 0.6V
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
4400 4650 4900 5150 5400 5650 5900 6150
Time (s)
Tota
l Cur
rent
(A)
100°C110°C
120°C3.5A
4.0A4.3A
Total cell current at 0.7 V cell potential of SDAPP 2.2 meq/g electrolysis cell with time at 100°C, 110°C, and 120°C cell temperature.
Cell Conditions:
Increased Temperature Promotes Better PerformanceHigh Temperature Enabled by SNL membranes
• 10 cm2 cell
• SDAPP 2.2 meq/g membrane batch
• 2 mg Pt/cm2 Pt Black anode and cathode
• Dry SO2 gas anode, 100 sccmconstant SO2 flow rate with 15 psig backpressure
• Preheated liquid water cathode, 3 mL/min constant H2O flow rate with 15 psig backpressure
Performance increase of SNLMembrane Electrolysis Cell at 0.7 V
0
50
100
150
200
250
300
350
400
450
500
80 90 100 110 120 130
Temperature (°C)
Cur
rent
Den
sity
@ 0
.7V
(mA
/cm
2 )
Current density at 0.7 V cell potential of SDAPP 2.2 meq/gelectrolysis cell as a function of temperature
Galvanostatic Performance of SNL Membrane at 120°C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
9150 9250 9350 9450 9550 9650 9750 9850
Time (s)
Cel
l Pot
entia
l @ 5
00 m
A/c
m2 (V
)
target0.6V
Cell potential at 500 mA/cm2 and 120°C for SNL SDAPP 2.2 meq/g electrolysis cell.
The average potential of the electrolysis cell for 10 minutes is 0.75 V. The previous best performance of this membrane was 0.83 V at 500 mA/cm2 and 80°C as measured by University of South Carolina.
Performance and Lifetime of SNL (4-141-D) Membranes
1
2
3
4
5
6
0 5 10 15 20
Time (Hours)To
tal C
urre
nt (A
)
Higher current densities are achieved at elevated temperatures with 0.4 mA/cm2 at 120°C. Extended stable performance for 16 hours at 0.8V and120°C is also demonstrated. The SDAAP 2.2 meq/g (4-141-D batch) electrolysis cell was tested with 100 sccm dry SO2 and and 6.4 ml/min H2O(l)flow rates and 15 psig backpressure.
0
0.1
0.2
0.3
0.4
0.5
80 90 100 110 120 130
0.8 V0.7 V
Temperature (oC)
Cur
rent
Den
sity
(A/c
m2 )
Decreased SO2 Crossover Using SNL Membranes –less process loss, higher efficiency
SO2 crossover - process loss- parasitic H2 consumption- elemental sulfur buildup on the
cathode, block reaction sites
SO2 flux to cathode is lower for Sandia membranes even though SNL membranes are thinner.
Steady-state SO2 flux for Sandia membrane SDAAP 2.2 meq/g (4-141-D) and Nafion 212 (Lynntech MEA).
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
50 70 90 110 130
Temperature (C)
Sandia SDAAP 2.2
Lynntech (N212)
Flux
(μ
mol
es S
O2*
cm-2
*s-1
)
60 μm thick
90 μm thick
Potential H2 Losses from SO2 Crossover
Hydrogen consumption at cell cathode from SO2 Crossover:SO2 + 2 H2 → S(solid) + 2 H2O
Temperature Crossover FluxTheoretical Maximum
H2 ConsumptionCell Hydrogen
Production
Theoretical Maximum %H2
Loss
C μmol SO2/cm2*s μmol H2/cm2*s μmol H2/cm2*s %
90 0.397 0.794 1.68 47.1
100 0.322 0.644 1.81 35.5
110 0.272 0.544 2.02 26.9
120 0.176 0.352 2.10 16.8
More efficient hydrogen production is achieved at 120°C cell operating temperature.
The number noted above are the maximum amount of H2 lost. In reality, ~ 5% penalty is observed due to counter water flux at high currents.
Summary• SNL membranes have shown promise in SO2 electrolyzer tests.• High temperature, up to 120°C, and long run-time performance has been
demonstrated with SNL membranes.• SNL membranes have approximately 50% less SO2 permeability than
Nafion membranes.• Batch-to-batch repeatability needs improvement.
Ongoing Work• Repeated scaled-up synthesis of the polymer.• Large film casting.• Higher temperature variants of SNL polymers being tested.• Additional SO2 crossover measurements.
Acknowledgements
Our project partners at:Savannah River National Laboratory – William Summers,
David Hobbs, Hector Colon-MercadoUniversity of South Carolina – John Weidner, John Staser
Dr. Richard Mackay (Eltron Research, Inc)Dr. Margaret Welk (Permeation measurements)Mr. Gary Jones (Permeation unit construction)
Prof. Alexandra Navrotsky & Dr. R.G. Iyer(UC Davis, Calorimetry)
Dr. Bonnie McKenzie (SEM)