NH3 generation over commercial Three-Way …...Lean Gasoline Emissions Control: NH 3 generation over...
Transcript of NH3 generation over commercial Three-Way …...Lean Gasoline Emissions Control: NH 3 generation over...
Lean Gasoline Emissions Control:
NH3 generation over commercial
Three-Way Catalysts and Lean-NOx Traps
Todd J. Toops, James E. Parks II and Josh A. Pihl
Oak Ridge National Laboratory
Christopher D. DiGiulio and Michael D. Amiridis
University of South Carolina
2012 DEER Conference
October 18, 2012
2 Managed by UT-Battelle for the U.S. Department of Energy
Small improvements in gasoline fuel economy significantly decreases fuel consumption
Lean gasoline vehicles can
decrease US gasoline
consumption by ~30 million gal/day
• 132,000 million gallons of fuel used by cars and light trucks annually** • New car and light-truck sales dominated by gasoline engines • 10% fuel economy benefit † from base case of 22.6/18.1 mpg** has big impact
– Saves >200,000,000 barrels gasoline annually – 5% of overall petroleum used
• HOWEVER…emissions control challenges exist
*, ** - References: Transportation Energy Data Book, Ed. 29; *2009 data; **2008 data. † - data from chassis dynamometer data at ORNL; see technical backup slides
total petroleum use by sector* cars
light trucks
3 Managed by UT-Battelle for the U.S. Department of Energy
Why NH3-generation on TWCs? • Zeolite-based NH3-SCR has been shown to have very high NOx conversions over
wide temperature window
• Urea injection systems are unlikely solution in lean gasoline systems – Significant additional cost would deter consumers – Higher engine out NOx will require more urea
• introducing urea filling infrastructure issues on this scale – Other NH3 introduction methods being studied
• Utilizing existing TWCs on gasoline vehicles is intriguing since they are already on
the vehicle and will be needed on lean gasoline vehicles – NH3 generation recently explored in “Passive SCR” approach*
• Goal is to investigate potential of using similar levels of PGM on TWC
– while maximizing lean timing and minimizing fuel penalty
* - GM: SAE 2011-01-0306, SAE 2010-01-0366
4 Managed by UT-Battelle for the U.S. Department of Energy
“Passive” SCR for lean gasoline emissions control
TWC SCR stores
NH3
TWC converts: NO to NH3
HCs to CO + H2
slightly rich operation AFR ≈ 14
0
200
400
600
800
1000
1200
1400
Gas C
once
ntra
tion
(ppm
)
TWC output
NH3
NO
N2O
CO
AFR 14.3
NOx feed
5 Managed by UT-Battelle for the U.S. Department of Energy
“Passive” SCR for lean gasoline emissions control
TWC stored NH3
reduces NOx
TWC converts: CO to CO2
HCs to CO2 + H2O
lean operation AFR ≈ 24
0
200
400
600
800
1000
1200
1400
Gas C
once
ntra
tion
(ppm
)
TWC output
NH3
NO
N2O
CO
AFR ~14.8
NOx feed
6 Managed by UT-Battelle for the U.S. Department of Energy
NH3 GENERATION IN BENCH REACTORS
7 Managed by UT-Battelle for the U.S. Department of Energy
TWC and LNT studied in bench-core reactor with varying PGM content • For bench reactor, focusing on modern TWC technology
– 1.3L TWC is a 2 formulation combination (combo) • Total PGM: 0/4.0/0.16 g/L Pt/Pd/Rh (118 g/ft3 total PGM)
– Front 0.6L of TWC is Pd-only no Ce • High PGM: 0/6.7/0 g/L Pt/Pd/Rh (190 g/ft3 total PGM) • No ceria-based OSC, but oxygen storage measured
– Expected to proceed via Pd-O formation – Rear 0.7L of TWC is Pd/Rh+Ce w/ Ceria
• Low PGM: 0/1.1/0.3 g/L Pt/Pd/Rh (40 g/ft3 total PGM) – Investigating each portion individually and
in combined form • Degreened at 16h at 700C in humidified air (2.7% H2O)
• LNT is commercial formulation from lean gasoline BMW – 2.6L Pt/Pd/Rh = 7/3/1, 3.3 g/L-cat (94 g/ft3); Ba loading: 20 g/L (560 g/ft3); Ce: 56 g/L (1600 g/ft3) – Degreened at 16h at 700°C in humidified air (2.7% H2O)
8 Managed by UT-Battelle for the U.S. Department of Energy
TWC is effective and tunable NH3 generator
• NH3 readily generated; varies with PGM – For Pd-only TWC with high PGM:
• All NO fed converted to NH3 when very rich – For Pd/Rh+Ce (low PGM ) TWC:
• NH3 production is still significant but reduced
• At all conditions, >95% CO conversion – C3H6 not observed in effluent
• N2O formation observed under lean
conditions and varies with PGM content – Up to 56 ppm with high PGM (Pd-only) TWC – Less than 10 ppm with low PGM (Pd+Rh) TWC
Midbed Temperatures: 460-500ºC
0
200
400
600
800
1000
1200
1400
1.1% 1.2% 1.3% 1.4% 1.5% 1.6% 1.7%
Gas C
once
ntra
tion
(ppm
)
O2 Concentration
Pd-only (high PGM)
NH3
NO
N2O
CO
0
200
400
600
800
1000
1200
1400
1.1% 1.2% 1.3% 1.4% 1.5% 1.6% 1.7%
Gas C
once
ntra
tion
(ppm
)
O2 Concentration
Pd/Rh+Ce (low PGM)
NH3
NO
N2O
CO
~AFR O2 NO CO H2 C3H6
14.6 1.59% 0.12% 1.80% 0.60% 0.10% 14.4 1.34% 0.12% 1.80% 0.60% 0.10% 14.2 1.06% 0.12% 1.80% 0.60% 0.10%
NOx feed
NOx feed
• Example feed conditions:
9 Managed by UT-Battelle for the U.S. Department of Energy
PGM content and Pt/Pd/Rh ratios impact NH3 production
TWC
Evaluated multiple upstream catalyst formulations for NH3 generation
High PGM (Pd-only) best for NH3 generation
0%
20%
40%
60%
80%
100%
150 250 350 450 550
NO
to N
H3
conv
ersi
on
TWC inlet T (°C)
Pd/Rh+Ce LNT (Pt/Rh+Ce+Ba)
Combo
Pd-only
10 Managed by UT-Battelle for the U.S. Department of Energy
AFR and temperature dictate NH3 production
Pd TWC
Quantified NH3 generation over Pd only TWC
NH3 generated over wide T window Need richer conditions at higher T
0%
20%
40%
60%
80%
100%
250 300 350 400 450 500 550 600
NO
to N
H3
conv
ersi
on
TWC inlet T (°C)
14.0
14.1
14.2
14.3
14.4
14.59
AFR:
11 Managed by UT-Battelle for the U.S. Department of Energy
NH3 production over LNT and TWC occurs at temperatures relevant to vehicle operation and NH3 storage on SCR
• Histogram of catalyst temperatures during drive cycle with BMW 120i – FTP (hot start) – 200-350ºC for underfloor catalyst – 350-600ºC for close-coupled (cc) TWC
• TWC: tunable NH3 production 250-600ºC
• NH3 production temperatures over cc-TWC mesh well with NH3 storage temperatures on underfloor SCR – More NH3 storage occurs under
rich/stoichiometric conditions – However switching from rich to lean will
result in NH3 release if over-saturated 0.0
0.20.40.60.81.01.21.41.61.8
550500450400350300250200
NH3 s
tora
ge ca
pacit
y (g/
l)
Temperature (°C)
with Oxygen(lean)
w/o Oxygen(rich/stoich.)
Close-coupled TWC
Underfloor Catalyst
12 Managed by UT-Battelle for the U.S. Department of Energy
PASSIVE SCR APPROACH IN BENCH REACTOR
13 Managed by UT-Battelle for the U.S. Department of Energy
To FTIR
Furnace 1: TWC
300-600°C
Furnace 2: NH3–SCR
300°C
Gas Pre-heater
Gas Pre-heater
Feed
L R Vent Vent
Lean 600 ppm NOX 8% O2 5% H2O, 5% CO2, Bal. N2
Rich 1200 ppm NOX 1.8% CO 0.6% H2 0.1% C3H6 0.79%, 0.98%, 1.06%, 1.20% :O2 14.0 14.1 14.2 14.3 :AFR 5% H2O, 5% CO2, Bal. N2
Pd Only HPGM TWC
Cu/Zeolite
75,000 hr-1
30,000 hr-1
Fully-automated two furnace bench reactor employed with TWC and SCR
• Switches from lean to rich when FTIR reads 20 ppm NOx
TWC + SCR NH3
• Rich to lean when predicted NH3 storage is half filled
14 Managed by UT-Battelle for the U.S. Department of Energy
Manual optimization of cycling illustrates the potential of this approach
Flow reactor proof of concept: TWC+SCR under cyclic operation TWC inlet 450 °C
SCR inlet 300 °C rich time 60 s (AFR≈14) lean time 110 s (AFR≈24) NOx conv. 99.5% CO conv. 93.9%
avg (ppm)
max (ppm)
NO 4 31 N2O 2 32 NH3 3 4 CO 1200 3900
Pd TWC
0
200
400
600
800
1000
1200
0 3 6 9 12 15
conc
entr
atio
n (p
pm)
time (min)
NO in
NO
N2O
NH3
Estimated fuel economy gain: 5-6%
15 Managed by UT-Battelle for the U.S. Department of Energy
Fuel economy impacts of NOx emissions compliance with passive SCR
• Rich excursions to generate NH3 increase fuel consumption due to: – temporary loss of lean operation
fuel economy boost (assumed 10%) – injection of excess fuel
• Overall impact on fuel economy depends on: – how long the engine can run lean – how rich it must go to generate NH3
16 Managed by UT-Battelle for the U.S. Department of Energy
Fuel economy impacts of NOx emissions compliance with passive SCR
• Fraction of time lean depends on: – NH3 yield during rich operation
• decreases with TWC T and AFR – relative NOx flux (concentration
and flow rate) during rich vs. lean • rich NOx = 2 x lean NOx here
• Exploiting flow changes during transient driving could increase lean operation time
• Higher than expected lean time at 300°C due to NOx storage over TWC during lean operation – possible formulation strategy
• For all conditions shown NOx
conversion is > 99%
17 Managed by UT-Battelle for the U.S. Department of Energy
Fuel economy impacts of NOx emissions compliance with passive SCR
• Current implementation of passive SCR generates a net fuel economy benefit of 5-6% over a comparable stoich. engine
• Optimal AFR depends on TWC T
• Possibilities for reducing fuel penalty: – higher NOx flux ratios during
transient operation on vehicle – addition of a NOx storage material
18 Managed by UT-Battelle for the U.S. Department of Energy
For a fixed cycle, addition of NOx storage significantly increases NH3 formation • Cycle Averaged Alpha is 3+ times higher than best TWC formulation
– Alpha: NH3 produced / NOx in effluent
0.0
1.0
2.0
3.0
4.0
5.0
200 250 300 350 400 450 500 550
Cylc
e Av
erag
ed α
Temperature ( °C)
HPGM TWC Dual TWC LPGM TWC LNTPd/Rh+Ce LNT (Pt/Rh+Ce+Ba) Combo Pd-only
19 Managed by UT-Battelle for the U.S. Department of Energy
Emissions other than NOx still present potential challenges • Current operating strategies optimized for
NOx reduction • Minimal HCs observed with 0.1% C3H6 in feed
– Consumed by reactions with NOx, O2, or steam reforming
• Significant CO slips during rich operation – higher AFRs reduce but do not eliminate – CO slip a complicated function of TWC
temperature due to water-gas shift kinetics and thermodynamics
– downstream cleanup catalyst and secondary air may be required
• N2O may be problematic at low TWC temperatures – potential for mitigation by changes in
formulation or operating strategy
20 Managed by UT-Battelle for the U.S. Department of Energy
Summary • TWCs shown to be able to produce NH3 over a
broad temperature window – Key variables are PGM content, temperature
and AFR
• Greater than 99% NOx conversion observed in passive approach – Lean-only conversion is >98% – CO slip is a concern and will need to be
accounted for
• Significant fuel economy gain realized which could be improved with NOx storage on TWC – Future efforts exploring addition of NOx
storage component on Pt-free TWC
0
200
400
600
800
1000
1200
0 3 6 9 12 15
conc
entr
atio
n (p
pm)
time (min)
NO in
NO
N2O
NH3
0%
20%
40%
60%
80%
100%
250 300 350 400 450 500 550 600
NO
to N
H3
conv
ersi
on
TWC inlet T (°C)
14.0
14.1
14.2
14.3
14.4
14.59
AFR:
21 Managed by UT-Battelle for the U.S. Department of Energy
Acknowledgements • Funding provided by US-DOE Vehicle Technologies Program
– Gurpreet Singh and Ken Howden
• University of South Carolina – Graduate Student support of Chris DiGiulio
• General Motors – Monthly teleconferences – Wei Li, Chang Kim, Kushal Narayanaswamy, Paul Nagt
• Umicore – Catalysts and Monthly teleconferences – Chris Owens, Ken Price, Doug Ball, Tom Pauly, Corey Negohosian
• University of Wisconsin
– Modeling (not discussed here) and Monthly teleconferences – Chris Rutland and Jian Gong
22 Managed by UT-Battelle for the U.S. Department of Energy
ADDITIONAL SLIDES
23 Managed by UT-Battelle for the U.S. Department of Energy
Reaction Conditions
Lambda AFR O2 NO CO H2 C3H6
1.000 14.59 1.59% 0.12% 1.80% 0.60% 0.10%
0.996 14.54 1.525% 0.12% 1.80% 0.60% 0.10%
0.995 14.52 1.51% 0.12% 1.80% 0.60% 0.10%
0.987 14.40 1.34% 0.12% 1.80% 0.60% 0.10%
0.973 14.20 1.06% 0.12% 1.80% 0.60% 0.10%
0.960 14.00 0.79% 0.12% 1.80% 0.60% 0.10%
24 Managed by UT-Battelle for the U.S. Department of Energy
This image cannot currently be displayed.
BMW lean GDI LNT benchmarking against CLEERS reference
CLEERS reference Lean GDI, 2004, provided by Umicore
This image cannot currently be displayed.
New LNT Lean GDI, 2009, from BMW 120i vehicle
Ce & Zr-rich
Mg & Al-rich
CLEERS reference
New LNT
Cell density (cpsi) 625 413 Ba loading (g/ft3) 442 565 PGM loading (g/ft3) 103 94 Pt/Pd/Rh ratio 8/3/1 7/3/1
• Responsive to FY10 review comments • Initial characterization indicates similar basic
components in both LNTs – Reactor evaluation planned – Catalyst used in an ORNL lean gasoline engine-
based project (see ACE033 talk)
Cordierite
25 Managed by UT-Battelle for the U.S. Department of Energy
Detailed elemental analysis of different LNT technologies on hand
cf. CLEERS Umicore LNT Element Loading (g/ft3) Loading (g/in3) Loading (g/L) Ba 442 0.256 15.6 Ce 1978 1.145 69.9 PGM ratio Pt 72 0.042 2.5 8 Pd 23 0.013 0.8 3 Rh 9 0.005 0.3 1
PGM loading 103 0.06 3.6
BMW LNT Element Loading (g/ft3) Loading (g/in3) Loading (g/L) Ba 565 0.33 19.9 Ce 1572 0.91 55.5 Zr 122 0.07 4.3 La 69 0.04 2.5 PGM ratio Pt 62 0.04 2.2 7 Pd 24 0.01 0.8 3 Rh 8 0.00 0.3 1
PGM loading 94 0.05 3.3
New LNT Lean GDI, 2009, from BMW 120i vehicle
CLEERS reference Lean GDI, 2004, provided by Umicore
26 Managed by UT-Battelle for the U.S. Department of Energy
0%
5%
10%
15%
20%
25%
150 250 350 450 550 650 750
N2O
Yie
ld. (
%)
Mid-Bed Temp. ( °C)
0.79 % O2 1.06 % O2 1.34 % O2
1.525 % O2 1.59 % O2
Pd only TWC (high PGM)
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
150 250 350 450 550 650 750
N2O
Yie
ld. (
%)
Mid-Bed Temp. ( °C)
0.79 % O2 1.06 % O2 1.34 % O2
1.525 % O2 1.59 % O2Pd-Rh/CeO2 TWC (low PGM)
0%
2%
4%
6%
8%
10%
150 250 350 450 550 650 750
N2O
Yie
ld (%
)
Mid-Bed Temp. ( °C)
0.79 % O2 1.06 % O2 1.34 % O2
1.51 % O2 1.525 % O2 1.59 % O2
Pd + Pd-Rh/CeO2 TWC (high + low PGM)
0%
2%
4%
6%
8%
10%
150 250 350 450 550 650 750
N2O
Yie
ld. (
%)
Mid-Bed Temp. ( °C)
0.79 % O2 1.06 % O2 1.34 % O2
1.510 % O2 1.525 % O2 1.59 % O2
LNT (High PGM, Pt/Pd/Rh)
N2O formation can be significant on all catalysts and under lean or rich operation
27 Managed by UT-Battelle for the U.S. Department of Energy
TWC (high PGM Pd-only) transients • NH3 profiles during lean to rich transitions:
• CO profiles during lean to rich transitions:
0
0.2
0.4
0.6
0.8
1
1.2
1.5 1.6 1.7 1.8 1.9 2 2.1
CO c
once
ntra
tion
(%)
time (min)
14 14.1 14.2 14.3
0
0.2
0.4
0.6
0.8
1
1.2
1.5 1.6 1.7 1.8 1.9 2 2.1
CO c
once
ntra
tion
(%)
time (min)
14 14.1 14.2
0
0.2
0.4
0.6
0.8
1
1.2
1.5 1.6 1.7 1.8 1.9 2 2.1
CO c
once
ntra
tion
(%)
time (min)
14.0 14.1
0
200
400
600
800
1000
1200
1400
1600
1800
2000
1.5 1.6 1.7 1.8 1.9 2 2.1
NH3
con
cent
ratio
n (p
pm)
time (min)
14 14.1 14.2 14.3
0
200
400
600
800
1000
1200
1400
1600
1800
2000
1.5 1.6 1.7 1.8 1.9 2 2.1
NH3
con
cent
ratio
n (p
pm)
time (min)
14 14.1 14.2
0
200
400
600
800
1000
1200
1400
1600
1800
2000
1.5 1.6 1.7 1.8 1.9 2 2.1
NH3
con
cent
ratio
n (p
pm)
time (min)
14.0 14.1
300°C 450°C 600°C
300°C 450°C 600°C
28 Managed by UT-Battelle for the U.S. Department of Energy
TWC (high PGM Pd-only) transients • NO profiles during lean to rich transition illustrate some storage at low temperature:
• N2O profiles during lean to rich transitions:
300°C 450°C 600°C
0
100
200
300
400
500
600
700
800
2 2.2 2.4 2.6 2.8 3 3.2
NO
con
cent
ratio
n (p
pm)
time (min)
14 14.1 14.2 14.3
0
100
200
300
400
500
600
700
800
2 2.2 2.4 2.6 2.8 3 3.2N
O c
once
ntra
tion
(ppm
)
time (min)
14 14.1 14.2
0
100
200
300
400
500
600
700
800
2 2.2 2.4 2.6 2.8 3 3.2
NO
con
cent
ratio
n (p
pm)
time (min)
14.0 14.1
0
20
40
60
80
100
120
140
160
180
1.5 1.6 1.7 1.8 1.9 2 2.1
N2O
con
cent
ratio
n (p
pm)
time (min)
14 14.1 14.2 14.3
0
20
40
60
80
100
120
140
160
180
1.5 1.6 1.7 1.8 1.9 2 2.1
N2O
con
cent
ratio
n (p
pm)
time (min)
14 14.1 14.2
0
20
40
60
80
100
120
140
160
180
1.5 1.6 1.7 1.8 1.9 2 2.1N
2O c
once
ntra
tion
(ppm
)
time (min)
14.0 14.1
300°C 450°C 600°C
29 Managed by UT-Battelle for the U.S. Department of Energy
TWC + SCR NH3
SCR Screening:
NH3–SCR Storage/Reaction: SV = 30,000 hr–1
Storage: 500 ppm NH3
5% H2O, bal. N2
Reaction: 500 ppm NO
10% O2 5%H2O, bal. N2
0.00
0.10
0.20
0.30
0.40
0.50
0.60
200 250 300 350 400 450 500 550 600
Temperature (°C)
Rich
Sel. Reaction
0.0
0.5
1.0
1.5
2.0
2.5
200 250 300 350 400 450 500 550 600
Stor
ed N
H 3
(gN
H3/L
mon
olith
)
Temperature (°C)
Rich
Sel. Reaction
Cu/Zeolite Fe/Zeolite
Steady-State NH3–SCR: SV = 30,000 hr–1
500 ppm NO 500 ppm NH3
10% O2 5% H2O
0
20
40
60
80
100
150 200 250 300 350 400 450 500 550 600
NO
XCo
nver
sion
(%)
Temperature ( °C)
Cu/Zeolite Fe/Zeolite