The Effects of Hydrocarbons on NOxReduction over Fe-based SCR Catalyst
Maruthi Devarakonda1, Russell Tonkyn1, Diana Tran1, Darrell Herling1, Jong Lee1, Josh Pihl2, Stuart Daw2
1Pacific Northwest National Laboratory2Oakridge National Laboratory
Directions in Engine-Efficiency and Emissions ResearchDetroit, MI
September 30th, 2010
2
Motivation
• Selective Catalytic Reduction (SCR) effective over a wide range of temperatures.
• Higher NOx reduction performance required to meet more stringent emission standards
• Cooler exhaust temperatures due to advanced combustion technologies.
• Factors that can affect NOx conversion at low temperatures include: Inhibition of NH3 storage at low temperatures
Hydrocarbon slip from DOC/DPF during cold-start
3
Objectives
• Investigate the effects of hydrocarbons on various SCR reaction steps under controlled lab reactor conditions.
• Develop kinetic models to characterize competitive adsorption and inhibition, and to describe the impact on SCR performance quantitatively.
4
Outline
• Lab Reactor Experiments
• DRIFT Experiments
• Hydrocarbon Storage & Inhibition Modeling
• Conclusions & Future Work
5
Reactor Test Setup
Fe-zeolite catalyst400/6.5
160g/L washcoat loading
6
Test Conditions
7
NOx Reaction Pathways
In addition to NH3 adsorption and desorption on SCR catalyst surface,
NH3 oxidation: 2NH3 + 3/2O2 → N2 + 3H2O
NO oxidation: NO + 1/2O2 ↔ NO2
Standard SCR: 4NH3 + 4NO + O2 → 4N2 + 6H2O
NO2-SCR: 4NH3 + 3NO2 → 7/2N2 + 6H2O
Fast SCR: 2NH3 + NO + NO2 → 2N2 + 3H2O
8
20
40
60
80
100
200 300 400 500 600Temperature (oC)
Con
vers
ion
(%)
NOx (with Dodecane) NOx (w/o HC)NH3 (with Dodecane) NH3 (w/o HC)NOx (with Toluene) NH3 (with Toluene)
Feed Conditions
175 ppm NO
175 ppm NO2
350 ppm NH3
14% O2
2% H2O
50 ppm toluene (350 C1)
29 ppm dodecane (350 C1)
29k h-1
Effect of Hydrocarbon on NOx Reduction
• No effect of ethylene, propane
• Detrimental effects of toluene, dodecane
9
Feed Conditions
350 ppm NOx
350 ppm NH3
14% O2
2% H2O
50 ppm toluene (350 C1)
29k h-1
Effect of Toluene on NOx Reduction
• More pronounced effect on Standard SCR
• No effect on NO2-SCR
10
350 ppm NO, 14% O2, 25 ppm toluene, 2% H2O
• More pronounced inhibition effect during temp-up ramp
• Severe inhibition in the presence of H2O
Effect of Toluene on NO Oxidation
11
Feed Conditions
175 ppm NO
175 ppm NO2
350 ppm NH3
14% O2
2% H2O
29 ppm dodecane (350 C1)50
60
70
80
90
100
200 300 400 500 600Temperature (oC)
Con
vers
ion
(%)
NOx, w/o dodecaneNOx,w/ Dodecane
• Decreased NOx reduction during temp-down ramp
• More pronounced effect on Standard SCR
• No effect on NO2-SCR
Effect of Dodecane on NOx Reduction
12
Outline
• Lab Reactor Experiments
• DRIFT Experiments
• Hydrocarbon Storage & Inhibition Modeling
• Conclusions & Future Work
13
In Situ DRIFT Setup
• Diffuse reflectance IR FT spectroscopy (DRIFT) Adsorbates on the particle surfaces
under SCR reaction conditions
• Typical experiment: Under 14% O2, 5% CO2, 4.5% H2O:
o Heat to 500 C to clean surfaceo Cool to experiment temperatureo Take background spectrum
Turn on 350 ppm NOx Turn on 350 ppm NH3
Turn on 350 ppm [C1] HC Continue scanning after turning off HC
HeatElement
ParabolicMirror
Sample
GasIn
GasOut
HeatElement
ParabolicMirror
Sample
GasIn
GasOut
HeatElement
ParabolicMirror
Sample
GasIn
GasOut
HeatElement
ParabolicMirror
Sample
GasIn
GasOut
IR Beam
in situ DRIFTS reactor
14
NH3 Consumption by NOx
Exposure conditions: 350 ppm NO or NO2, 350 ppm NH3, 14% O2, 5% CO2, 4.5% H2O
• Surface NH3 continually consumed by SCR reactions
• NO more effective than NO2at decreasing adsorbed NH3
3800 3300 2800 2300 1800 13004300
wavenumber (cm-1)
abso
rban
ce
200°C
300°C
-0.15-0.1
-0.050
0.050.1
0.150.2
0.250.3
0.35
-0.1-0.08-0.06-0.04-0.02
00.020.040.060.08
0.1
NH3 NH3+NO NH3+NO2
15
Toluene Inhibition at 200°C
• Introduction of toluene generates new adsorbates‒ 1300-1800 cm-1: most likely
toluene and derivatives
• Increased NH3 on surface due to poisoned NO+NH3reaction
• No changes upon removal of toluene‒ Poisoning irreversible at 200 C
• Little effect on NO2-SCR
3800 3300 2800 2300 1800 13004300
wavenumber (cm-1)
abso
rban
ceab
sorb
ance
diff
eren
ce
-0.15-0.1
-0.050
0.050.1
0.150.2
0.250.3
0.35
[NH3][NH3+NO]
-0.1
-0.05
0
0.05
0.1
0.15
[NH3+NO+C6H5CH3]
[NH3+NO+C6H5CH3] - [NH3+NO]
[NH3+NO(-C6H5CH3)]
[NH3+NO(-C6H5CH3)] - [NH3+NO]
16
Reduced Inhibition at 300°C
• Toluene addition shifts spectrum considerably‒ Increased NH3 on the surface‒ Increased toluene-derived
species
• Some recovery upon removal of toluene‒ Increased NH3 consumption‒ Reduced toluene-derived species
3800 3300 2800 2300 1800 13004300
wavenumber (cm-1)
-0.1-0.08-0.06-0.04-0.02
00.020.040.060.08
0.1
[NH3][NH3+NO]
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
[NH3+NO+C6H5CH3]
[NH3+NO+C6H5CH3] - [NH3+NO]
[NH3+NO(-C6H5CH3)]
[NH3+NO(-C6H5CH3)] - [NH3+NO]
abso
rban
ceab
sorb
ance
diff
eren
ce
17
Outline
• Lab Reactor Experiments
• DRIFT Experiments
• Hydrocarbon Storage & Inhibition Modeling
• Conclusions & Future Work
18
Overview of PNNL 1-D SCR Model• Gas phase, surface phase concentrations and NH3 storage as states
• Coded as ‘C’ S-functions and developed in Matlab/Simulink
• Optimized and validated using steady state and thermal transient reactor data
19
Hydrocarbon Storage Model: Toluene
totalHCstotaleqst NcTKNn87,, )(
111+=
dtnnneqt
outHCinHCeqst )(0
,,, 8787∫ −=
Typical Adsorption Test
Langmuir Isotherms
Test Matrix
Storage rate parameters are obtained through approximation of Langmuir isotherms.
20
Single Site Kinetics Model for HC Storage
( )
( )
)(1
)1( ,,,,
,,,,
desadstol
jjiisiggi
is
isiggiigig
rrdt
d
rccAt
c
ccAx
cu
tc
−Ω
=
+−=∂
∂−
−−∂
∂−=
∂
∂
∑θ
βε
βεε
• Simulated using a variable step solver ode23tb, a TR-BDF2 algorithm
• Spatial derivative term approximated by a first order Euler integration scheme
• A total of 10 tanks (cells or axial increments) considered in series, each represented by a ‘C’ s-function and implemented in Matlab/Simulink
Validation at 150ppm, 100C
21
Modeling HC Inhibition of NO OxidationFeed: 350 ppm NO, 14% O2, 25 ppm toluene (175 C1)
Step Temperature Test
Arrhenius Plot for rate parameter (Ktol)
Model Validation on Temperature Ramps
Rate parameter (Ktol) for the inhibition model, follows Arrhenius dependency. It is calculated from steady state data and validated on temperature ramp tests.
tol
toltol
eq
NOoNOoxif
K
Kc
cckr
θθ−
+
−
=
11
)( 2
2
5.0,
22
Conclusions• Effects of hydrocarbons on NOx reduction pathways over
Fe-zeolite catalyst has been investigated.No effect of ethylene and propane on NOx reduction
Detrimental effects of toluene and n-dodecane on Standard SCR through suppressed NO oxidation
Increased surface NH3 due to suppressed SCR
HC poisoning irreversible at low temps
• 1-D model has been developed to describe the effects of hydrocarbons on SCR reaction kinetics.HC storage model developed using Langmuir isotherms
A single site kinetic model developed and validated to predict the effects of toluene & dodecane on NO oxidation and SCR reaction
23
Next Steps
• Investigate the competitive adsorption kinetics (NH3, H2O, HC) on Cu-zeolite SCR catalysts through experiments and modeling.
• Investigate the effects of catalyst aging on kinetic parameters and physicochemical properties of Cu-zeolite catalysts, and develop a catalyst aging model.
24
Acknowledgments
• Ken Howden and Gurpreet Singh, DOE-OVT
• CLEERS Team
• Kalyan Chakravarthy (ORNL)
• Umicore
