Urea Decomposition and SCR Performance at Low Temperature

Scott Sluder, John Storey, Sam Lewis, Linda Lewis,

Oak Ridge National Laboratory

10th Annual DEER Workshop Coronado, CA September 2, 2004

Sponsors: DOE-OFCVT, Steve Goguen and Kevin Stork

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 1 Why study urea decomposition?

• Under light-duty conditions (< 250 °C), there is negligible urea decomposition in the gas phase. Current models assume NH3 as input to catalyst. • It appears that urea decomposition kinetics on catalyst surfaces has to be understood. This will affect the length of the catalyst that is used for urea decomposition.

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 2 Urea/SCR Chemistry

Ideally, injected urea decomposes to NH3

O Heat

NH3 + HCNO H2O, O2 isocyanic acid H 2N NH2 Heat

HCNO NH3 + CO2

isocyanic acid H2O

then reacts with NOX to form N2.

2 NH3 + NO + NO2 ⇔ 2 N2 + 3 H2O

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 3 Unusual things can happen, such as ammonium nitrate formation and decomposition at T < 200 °C.

2NH3 + 2NO2 →NH4NO3 + N2 + 3 H2O

NH4NO3 ⇔ NH3 + HNO3

2HNO3 + NO → 3 NO2 + H2O

M. Koebel et al. in Ind.Eng.Chem.Res.: 39,p. 4120; 40, p. 52; 41, p.4008; 42, p.2093 Also in Catal.Today, 73, p.239

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 4 Experimental Approach • Use engine controls to: − Vary temperature

− Keep constant: NOX and exhaust flow. • Investigate urea decomposition (if any) upstream of SCR catalyst. • Analyze exhaust products exiting undersized monoliths to elucidate urea decomposition − 3” and 6” monoliths = space velocities of 25K and 50K − const. urea residence time in the exhaust stream. (~0.1 sec) − From clean condition, begin injecting urea: Wait 20 minutes (3” monolith) or, 40 minutes (6” monolith), OAK RIDGEThenNATIONALcollectLABORATORY data and samples. U. S. DEPARTMENT OF ENERGY 5 Experiments conducted at NTRC cell 4 using Mercedes engine and BP15 fuel.

Engine

Oxidation catalyst (beneath To exhaust engine) stack

Urea SCR catalyst Injector OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 6 Experiments conducted at NTRC using Mercedes engine and BP15 fuel.

Engine

Oxidation catalyst (beneath To exhaust engine) stack

3” monolith Urea SCR catalyst Injector OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 7 Experiments conducted at NTRC cell 4 using Mercedes engine and BP15 fuel.

Engine

Oxidation catalyst (beneath To exhaust engine) stack

6” monolith Urea SCR catalyst Injector OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 8 Measurement of actual NH3 in the presence of urea and isocyanic acid is difficult.

• Problematic if making measurements of urea decomposition products. • Short experiments showed that urea decomposed (to ammonia) on instrument filter at T as low as 80 °C. • Filters are required for sampling PM-laden exhaust gas. • Instrument filter T is typically > 100 °C to prevent condensation and loss of water soluble NH3 • Using 60 °C filter allows measurement of NH3 without decomposition of precursors, but precludes same stream from being used for HC measurements.

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 9 Higher T decrease in NOX conversion (for 3” brick) implies competing processes…. 100 90 80 70 60 50 40 30 20

Apparent Conversion (%) Apparent Conversion 3 inch SCR 10 6 inch SCR 0 150 170 190 210 230 250 270 OAK RIDGE NATIONAL LABORATORY SCR inlet T (C) U. S. DEPARTMENT OF ENERGY 10 The lower conversion of NOX is due at high temperature is due to lower NO2 conversion.

100 90 80 70 60 50 40 30 20 Apparent Conversion (%) Apparent Conversion 3 inch SCR 10 6 inch SCR 0 150 170 190 210 230 250 270 SCR inlet T (C) OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 11 Undersized SCRs slipped some NH3 at all conditions, indicating ample reductant available for NO reduction. 60 X 3 inch SCR 50 6 inch SCR

40

30

20 Concentration (PPM) Concentration 10

0 150 170 190 210 230 250 270 OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY SCR inlet T (C) 12 N2O slip indicates incomplete reduction in the shorter catalyst

50 3 inch SCR 45 6 inch SCR 40 35 30 25 20 15

Concentration (PPM) 10 5 0 150 170 190 210 230 250 270 OAK RIDGE NATIONAL LABORATORY SCR inlet T (C) U. S. DEPARTMENT OF ENERGY 13 HCN was also formed as an intermediate at low concentrations at the temperatures studied.

9 3 inch SCR 8 6 inch SCR 7 6 5 4 3

Concentration (PPM) 2 1 0 150 170 190 210 230 250 270 SCR inlet T (C) OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 14 Steady-state data suggest urea decomposition competes with NOX reduction processes.

• NO2 not converted as efficiently at high temperatures with the 3” SCR. • High concentration of intermediate N2O in the 3” SCR could account for most NOX disappearance. • Concentrations of intermediates drop with longer catalyst. • NH3 slip indicates available reductant. • Temperature gradients associated with urea decomposition and NOX reduction were too small to measure.

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 15 Results from onset of injection

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 16 NO, NO2 disappearance at 250 °C using urea and ammonia as reductants suggests similarity of reduction processes. 140 NO w/ Urea NO2 w/ Urea 120 NO with NH3 NO2 w/ NH3 100 Injection Starts 80

60

40

Concentration (ppm) Concentration 20

0 0 200 400 600 800 1000 1200 Time (s) OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 17 Onset of N2O slip and NH3 slip at 250 °C is due to competition of urea decomposition process

14 N2O w/Urea NH3 w/Urea N2O w/NH3 NH3 w/NH3 12

10 Injection Starts 8 6 4 Concentration (ppm) Concentration 2 0 0 200 400 600 800 1000 1200 OAK RIDGE NATIONAL LABORATORY Time (s) U. S. DEPARTMENT OF ENERGY 18 NO and NO2 disappearance similar at 180 °C when using urea and ammonia as reductants.

160 NO w/ Urea NO2 w/ Urea NO with NH3 NO2 w/ NH3 120

80 Injection Starts

40 Concentration (ppm) Concentration

0 0 200 400 600 800 1000 1200

OAK RIDGE NATIONAL LABORATORY Time (s) U. S. DEPARTMENT OF ENERGY 19 Large NH3 slip again occurs much sooner with urea injection than with ammonia injection. 15 N2O w/Urea NH3 w/Urea N2O w/NH3 NH3 w/NH3

10

5 Concentration (ppm) Concentration Injection Starts 0 0 200 400 600 800 1000 1200 OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY Time (s) 20 Urea decomposition hampers NOX reduction in at least the first 3” of SCR catalyst.

• Steady state data showed presence of NH3, intermediates downstream of 3” and 6” SCR. • Transient data showed that NO, NO2 disappearance were similar using both NH3 and urea as reductants. • Ammonia injection data showed intermediates (HCN, N2O) at ~1/3 of concentration vs. urea • Ammonia injection data showed longer NH3 storage times before slip (~2X) than when urea was used.

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 21 Further investigation of the dynamics of urea decomposition is planned.

• Confirmatory, direct measurements of urea, isocyanic acid downstream of SCR are forthcoming. • Study 3”+6” SCRs together (as a 9”) to determine whether same trends are observed in rear 3” of 6” monolith. • Comparisons of urea and ammonia injection at other conditions to confirm

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 22