Simulating Stimulating Interfaces Applications in Adsorption and Catalysis C. Heath Turner and Xian Wang Department of Chemical and Biological Engineering The University of Alabama Kah Chun Lau Department of Chemistry, George Washington University Brett I. Dunlap Code 6189, Naval Research Laboratory, Washington D. C. Financial Support Provided by the Office of Naval Research
19
Embed
Simulating Stimulating Interfaces Applications in Adsorption and Catalysis
Simulating Stimulating Interfaces Applications in Adsorption and Catalysis. C. Heath Turner and Xian Wang Department of Chemical and Biological Engineering The University of Alabama Kah Chun Lau Department of Chemistry, George Washington University Brett I. Dunlap - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Simulating Stimulating InterfacesApplications in Adsorption and Catalysis
C. Heath Turner and Xian WangDepartment of Chemical and Biological Engineering
The University of Alabama
Kah Chun LauDepartment of Chemistry, George Washington University
Brett I. DunlapCode 6189, Naval Research Laboratory, Washington D. C.
Financial Support Provided by the Office of Naval Research
Solid Oxide Fuel Cells
Performance Characteristics…• Mainly for stationary applications with an output from 100 W to 2 MW.
• They typically operate at temperatures between 700 and 1,000°C.
• Efficiency can be as high as 90% (when the off-gas is used to fire a secondary gas turbine).
• Due to the high operating temperature, no need for expensive catalyst.
• SOFC are not poisoned by CO and this makes them highly fuel-flexible. So far they have been operated on methane, propane, butane, fermentation gas, gasified biomass, etc.
• Thermal expansion demands a uniform and slow heating process at startup (typically, 8 hours or more).
…we will focus on the cathode interface
SIMULATION SIZES and METHODS
TIME (s)
LENGTH(m)
Classical Methods
Mesoscale Methods
Continuum Methods
10-10 10-710-9 10-8 10-6 10-5 10-310-4
10-16
10-14
10-12
10-10
10-8
10-6
10-4
10-2
100
Semi-Empirical Methods
Ab Initio Methods
Kinetic Monte Carlo
Understanding the Performance of Solid-Oxide Fuel Cells
Determining how fuel cell materials respond at the atomic level to the global operation dynamics, and locally, due to interconnected micro or nanostructures, is a looming challenge.
Many variables to consider…
…how are they all correlated?…how is SOFC performance affected?
• Temperature
• Partial Pressure of the Gases
• Bias Voltage
• Geometry/Surface Area
• Cathode/Anode Materials
• Electrolyte Material, Dopant Level
Difficult to isolate experimentally, but given an appropriate model, simulations can lend some help
THE MODEL• We focus on the cathode-side of the electrolyte interface.
the anode side and the associated reactions are ignored
• The electrolyte is yttrium-stabilized zirconia (YSZ) the dopant level can vary, which affects the oxygen vacancy fraction
• The oxygen partial pressure can vary (PO2) this affects the O2 adsorption rate
• The temperature (T) can vary this affects the O2 adsorption/desorption equilibrium, the elementary
chemical kinetics, and the ionic diffusivity in the YSZ
• The bias voltage can vary this affects the oxygen incorporationreactions at the SOFC interface
• Current can be measured by monitoring the flux of ions through the YSZ.
• The electric double-layer can be incorporated into the model.
• Concentration profiles, surface coverage, and electrochemical information can be extracted.
Z0 Z1 ... Zi ... ZN
Cat
hod
e x
y
z
Anode
O2- FluxO2 (g)
Present WorkVappl- +
Zr4+
Y3+
O2-
Oxide-ion vacancyUnitCell
THE MODEL
• Structure corresponds to YSZ (100) surface of the bulk cubic fluorite (Fm3m) crystal…thermodynamically stable phase of zirconia at high T.
• A fixed Fm3m lattice parameter of 5.14 Å for 9 mol % YSZ is assumed.
• The YZr and Zr ions are fixed during the oxide ion vacancy migration.
• Other thermally, chemically, or electrically induced chemical or morphological changes are not included.
• YSZ: conducts only oxygen ions, electronic insulator.
• Neglect electrode details…assume electrode/YSZ TPB accounts for 1% of total surface area (used to normalize the calculated current).
O2 adsorption
O2 dissociation to O+O
O+O association to O2
O/O2 diffusion
Transfer of O in/outof the YSZ
Diffusion of O through the YSZ
CURRENTCURRENT
Structural Details
K. C. Lau, C. H. Turner, B. I. Dunlap, Solid State Ionics 179, 1912 (2008).
Default Simulation Parameters:• PO2 = 0.30 atm
• T = 1073 K• V = -0.50 V• relative permittivity = 40• D = ~10 nm• Area = ~32 nm2
• 9 mol % YSZ
Kinetic Monte Carlo Simulations
Electrostatic Interactions
Gas Adsorption
Bulk YSZ
• Total potential (VT) = electrode (Ve) + space-charge (Vsc)• Ve = electric potential from electrode (evenly distributed along YSZ)• Vsc = local space-charge (due to distribution of charges within YSZ)
K. C. Lau, C. H. Turner, B. I. Dunlap, Solid State Ionics 179, 1912 (2008).
• Poisson equation of electrostatics: 3D 1D• “Parallel Plate Capacitor”
• Assume uniform charge distribution within each plane and use Gauss’ law.
• The effective field influences the migration of the ions (when moving to neighboring plate, only)
• Eeff = E0 + 0.5(q)VT
Kinetic Monte Carlo Simulation: Reaction Events
41% coverage 47% coverage
PO2 = 0.21 atmT = 800 K
= - 0.60 V
0
10
20
30
40
50
0.08 0.09 0.10
Vacancy Fraction
An
gst
rom
s
I = 107 mA/cm2 I = 35 mA/cm2
VO˙˙ PROFILE
PO2 = 0.21 atmT = 800 K
= - 0.40 V
Reaction Steps in the KMC Mechanism:• Adsorption of O2(g): O2(g) + (*) → O2* k1• Desorption of O2*: O2* → (*) + O2(g) k2• Diffusion of O2*: O2* → O2* k3• Dissociation of O2*: O2* → O-* + O-* k4• Dimerization of O-*: O-* + O-* → O2* k5• Diffusion of O-*: O-* → O-* k6• Incorporation of O-*: O-* + VO˙˙→ O- - k7• Diffusion of VO˙˙: VO˙˙ → VO˙˙ k8
i i
ii r
rP
i ir
numberrandomt
)ln(
Choose an event proportional to the rate of the event:
Increment the clock by t:
0
1
2
3
4
5
6
7
8
9
0.04 0.06 0.08 0.10 0.12 0.14
Vacancy Fraction
Le
ng
th (
nm
)
-50
-25
0
25
50
-1 0 1Voltage (V)
Cu
rrent
(mA
/cm2)
• Kinetic parameters taken from previous experiments and calculations.
V = -1.0, T = 1073 K, 9% YSZ
0.01
0.1
1
10
100
1000
0.000001 0.0001 0.01 1 100
Partial Pressure O2 (atm)C
urre
nt (
mA
/cm
2 )
V = -0.50, T = 1073 K, 9% YSZ
0.01
0.1
1
10
100
0.00001 0.0001 0.001 0.01 0.1 1
Partial Pressure O2 (atm)
Cur
rent
(m
A/c
m2 )
RESULTS
• Is the current affected by the O2 partial pressure?
• Is the trend consistent with experimental observations?
• Is the same trend observed at LOWER voltages?
Kinetic Monte Carlo Simulation of a Solid-Oxide Fuel Cell
K. C. Lau, C. H. Turner, B. I. Dunlap, Chem Phys Lett (2009), accepted.
Kinetic Monte Carlo Simulation of a Solid-Oxide Fuel Cell
• Is the current affected by the relative permittivity of the electrolyte?
• Does the Y-dopant level strongly affect the current?
• Is this consistent with the experiments?
• Is the same trend observed at lower voltages?
Kinetic Monte Carlo Simulation of a Solid-Oxide Fuel CellRESULTS
• What is happening at the atomic level at the cathode interface?
…concentration profiles of the individual species can give us some insight.
0.8
1.0
1.2
1.4
1.6
10 20 30 40 50
Rel
ativ
e V
acan
cy C
on
cen
trat
ion
BULK YSZ
INTERFACE
V = -1.0
V = -0.8
V = -0.6
V = -0.4
V = -0.2
V = -0.0
Electric Double-Layer
0.6
0.8
1.0
1.2
10 20 30 40 50
Re
lati
ve
Va
ca
nc
y C
on
ce
ntr
ati
on
BULK YSZINTERFACE
V = 1.0
V = 0.8
V = 0.6
V = 0.4
V = 0.2
V = 0.0
Electric Double-Layer
…concentration profiles are consistent when the voltage is reversed.
Kinetic Monte Carlo Simulation of a Solid-Oxide Fuel CellRESULTS
• Can we model the frequency-response characteristics of the fuel cell?
…apply an alternating bias voltage.
As the frequency increases, the electric double-layer begins to diminish…the voltage frequency becomes larger than the dynamics of the charge accumulation.
0.0
0.5
1.0
1.5
2.0
2.5
1.00E+03 1.00E+05 1.00E+07 1.00E+09 1.00E+11
Am
plit
ud
e R
ati
o
current
(A/cm2)
effective voltage(V)
-90
-60
-30
0
30
1.00E+03 1.00E+05 1.00E+07 1.00E+09 1.00E+11
Applied Voltage Frequency
Ph
as
e A
ng
le
effective voltage
current
-2.5
-1.5
-0.5
0.5
1.5
2.5
simulation data…
Kinetic Monte Carlo Simulation of a Solid-Oxide Fuel CellRESULTS
(a)
(b)
(c)
Rp
Cdl
Re
Element Freedom Value Error Error %Rp Free(? 1.501E8 3.7752E6 2.5151Cdl Free(? 7.282E-15 1.7872E-16 2.4543Re Free(? 2.0242E6 52052 2.5715
Chi-Squared: 0.018252Weighted Sum of Squares: 0.4563
Data File: D:\fuel_cells\newcode-data\t1073-50\1.datCircuit Model File: D:\fuel_cells\newcode-data\t1073-20v05\3r2c.mdlMode: Run Fitting / Freq. Range (0.001 - 1000000000)Maximum Iterations: 100Optimization Iterations: 0Type of Fitting: ComplexType of Weighting: Calc-Modulus
The relative oxide-ion vacancy concentrations of the YSZ layers
Kinetic Monte Carlo Simulation of a Solid-Oxide Fuel CellRESULTS
(a)
(b)
(c)
Rp
Cdl
Re
Element Freedom Value Error Error %Rp Free(? 1.501E8 3.7752E6 2.5151Cdl Free(? 7.282E-15 1.7872E-16 2.4543Re Free(? 2.0242E6 52052 2.5715
Chi-Squared: 0.018252Weighted Sum of Squares: 0.4563
Data File: D:\fuel_cells\newcode-data\t1073-50\1.datCircuit Model File: D:\fuel_cells\newcode-data\t1073-20v05\3r2c.mdlMode: Run Fitting / Freq. Range (0.001 - 1000000000)Maximum Iterations: 100Optimization Iterations: 0Type of Fitting: ComplexType of Weighting: Calc-Modulus