1. Center for Applied Energy Research, University of Kentucky
NOx Storage-Reduction Characteristics of LNT Catalysts Subjected to Simulated Road Aging
Mark Crocker
Center for Applied Energy Research, University of Kentucky
April 30, 2009

2. Objectives
Gain insights into mechanism(s) of performance degradation in LNT catalysts
Study washcoat component effects for fresh and aged catalysts
Ascertain role of ceria in LNT catalysts, e.g.:
- NOx storage - sulfur storage - promotion of WGS reaction - etc.

3. Approach
Employ well characterized model catalysts which are representative of 2nd generation LNT formulations  use of ceria; also of relevance for lean-burn gasoline LNTs
Examine effect of washcoat components/loadings on catalyst durability:  systematic variation of component concentrations: Pt, Rh, Ba, CeO2(-ZrO2)
Employ realistic aging protocol, with simultaneous measurement of catalyst NOx storage/reduction performance
Perform detailed physico-chemical characterization of aged catalysts (N2 physisorption, H2 chemisorption, ELAN, XRD, SEM/EDS, TEM/EDS)

4. Model Monolith Catalyst Compositions Prepared
Component
Loading
Series 1
Series 2
Series 3
Pt, g/L (g/cuft)
3.53 (100)
3.53 (100), 2.65 (75), 1.77 (50)
Rh, g/L (g/cuft)
0.71 (20)
0.35 (10)
BaO, g/L
15, 30, 45 30
CeO2, g/L
0, 50, 100 - 50
CeO2-ZrO2, g/L
50, 100
Al2O3, g/L
Balance
Target washcoat loading = 260 g/L
Actual average loading = 262 g/L, stand. dev. = 16.6 g/L (6.3%)
Monoliths coated at DCL Int. Inc.

5. Model Catalyst Preparation: Method
Co-impregnation of a commercial, La-stabilized γ-Al2O3 with 1:1 Pt:Rh (Pt(NH3)4(OH)2 and Rh(NO3)3), followed by calcn. at 500 ºC
Impregnation of a commercial, high surface area γ-Al2O3 with Ba(OAc)2, followed by calcn. at 500 ºC
Ball milling the BaO/Al2O3 with CeO2 powder or CeO2-ZrO2, followed by impregn. with Pt (Pt(NH3)4(OH)2) and calcn. at 500 ºC
Mixing the powders from a) and c) with additional (balance) alumina, subsequently adding a small amount of boehmite sol as binder, washcoating onto cordierite monoliths, and calcining at 550 ºC

6. LNT Aging Based on Ford protocol
Initial aging to 50 cycles (ca k miles road equivalent based on fuel sulfur content)
Catalyst performance check every 10 cycles
Optimized desulfation temperature balances desulfation with thermal deactivation ( 700 °C)
At end of run, final desulfation performed with 2% H2 (5% H2O, 5% CO2) at 750 °C for 10 min to remove as much residual sulfur as possible

7. Comparison of Fresh and Aged Catalysts with Respect to OSC, NOx Storage and NOx Reduction

8. Comparison of OSC for Fresh and Aged Catalysts
Key to catalyst codes:
30-0: 0 g CeO2/L
30-50: 50 g CeO2/L
30-100: 100 g CeO2/L
30-100Z: 100 g CeO2-ZrO2/L ● All catalysts contain g BaO/L
T = 350 °C
Sintering of La-stabilized CeO2 reflected in lower OSC values after aging
OSC of CeO2-ZrO2 containing catalyst (30-100Z) hardly affected

9. Comparison of NO2 Profiles for Fresh and Aged Catalysts
Conditions: Lean: 60 s, 300 ppm NO, 10% O2, 5% CO2, 5% H2O, balance N2, GHSV = 30,000 h-1 Rich: 5 s, 2.625% CO, 1.575% H2, 5% CO2, 5% H2O, bal. N2, GHSV = 30,000 h-1
Catalyst 30-0, fresh and aged,
350 °C
Effect of aging:
Considerable increase in lean phase NO2 slip
NOx storage limited by NO2 adsorption, not by NO oxidation

10. Comparison of NOx Conversion for Fresh and Aged Catalysts
All the catalysts showed decreased NOx conversion after aging, especially at 450 °C
Catalyst 30-0 shows a more pronounced decrease in NOx conversion compared to the ceria-containing catalysts

11. Comparison of Selectivity to N2
Fresh
Aged
N2 selectivity (%)
All the catalysts show decreased N2 selectivity at T ≥ 250 °C after aging
Ceria-containing catalysts retain high selectivity to N2 at high temperature relative to 30-0

12. Comparison of Selectivity to NH3
Fresh
Aged
NH3 selectivity (%)
NH3 selectivity (%)
All catalysts show increased selectivity to NH3 at T ≥ 250 °C
Increased selectivity to NH3 correlates with decreased OSC

13. Comparison of Cycle Averaged and First Cycle NOx Storage Efficiencies Before and After Aging
T = 350 °C, 60/5 s lean/rich cycles
1st cycle NOx storage efficiencies (“clean catalyst”)
Cycle averaged NOx storage efficiencies
Clear decrease in intrinsic NSE for 30-0 after aging
Deterioration in cycle av. NSE reflects combination of reduced intrinsic NSE and lowered extent of rich phase LNT regeneration: → both effects important

14. Effect of Aging on Rich Phase NOx Release
T = 350 °C, 60/5 s lean/rich cycles
% Rich phase NOx release = rich phase NOx slip x 100
NOx stored in lean phase
Decreased NOx reduction efficiency evident after aging, especially for 30-0

15. Physico-chemical Characterization of Aged Catalysts

16. Surface Sulfur Concentrations by XPS
B.E. of ~170 eV confirms that sulfur is present exclusively as sulfate
Increase of residual S with increase in BaO loading
For aged 30-0, ~25% Ba sites observed by XPS are occupied by S

17. Sulfur Content of Aged Catalysts (LECO Analyzer)
Residual sulfur present that cannot be removed at 750 °C
Tendency for lower residual sulfur loading as catalyst ceria content increases

18. Effect of Aging on Washcoat Surface Area
Fresh catalysts:
● Surface area decreases with increasing Ba loading Aged catalysts:
● Sintering of La-stabilized CeO2 apparent
● CeO2-ZrO2 appears fairly stable to sintering under aging conditions used

19. PGM Sintering After Aging
H2 chemisorption at -78 °C: Sintering of Pt particles evident for aged catalysts
Use of CeO2(-ZrO2) as co- support appears to retard Pt sintering
Ability of ceria to stabilize high PGM dispersions: A.F. Diwell, R.R. Rajaram, H.A. Shaw, T.J. Truex, Stud. Surf. Sci. Catal. 1991, 71, 139.

20. Mechanisms of LNT Aging
Hypothesis:
1) Accumulation of sulfur in washcoat leads to lower intrinsic NOx storage efficiency
2) Pt sintering and concomitant Pt/Ba phase segregation results in: less efficient NO2 capture less efficient LNT regeneration (reductant spillover) less efficient NOx reduction however, NO oxidation not significantly affected (literature: rate of NO oxidation increases with Pt particle size*)
Results from SEM/TEM/EDS and steady state NOx reduction measurements should help to confirm these ideas
* S. Bernard, L. Retailleau, F. Gaillard, P. Vernoux and A. Giroir-Fendler, Appl. Catal. B: Environ. 2005, 55, 11

21. Conclusions
For aged catalysts, degradation of performance is manifested by: - lower lean phase storage efficiency; this is attributed to the inability of the catalyst to be regenerated sufficiently during rich purging, as well as some loss of intrinsic NSE - lower rich phase NOx reduction efficiency
Spectacular improvement in catalyst durability via the incorporation of CeO2-ZrO2 and La-stabilized CeO2 has been demonstrated
The benefits arising from CeO2-ZrO2 and La-stabilized CeO2 incorporation are attributed to: - facile desulfation of CeO2(-ZrO2), hence high intrinsic NSE retained - stabilization of Pt w.r.t. sintering

22. Acknowledgements
Yaying Ji, Vencon Easterling, Eduardo Santillan-Jimenez, Shelly Hopps (UK CAER)
Jae-Soon Choi (ORNL)
Mojghan Naseri (DCL International Inc.)
U.S. Department of Energy, Office of Vehicle Technology