The Effects of Long Term Sulfur Exposure on Three Way Catalyst Performance in Passive Selective Catalytic Reduction Systems

Calvin  Thomas, University of South Carolina

Introduction

Conventional gasoline-powered engines operate with a stoichiometric air-fuel ratio (AFR), but greater fuel economy is achieved in lean burn engines by increasing the AFR [1]. However, a traditional three-way catalyst (TWC) system is unable to fully reduce nitrogen oxide (NO_X) emissions in the oxidizing exhaust of these engines. Passive selective catalytic reduction (SCR) is a promising approach for the control of NO_X emissions in lean burn systems. By periodically decreasing the AFR and operating fuel-rich, ammonia (NH₃) can be produced over a TWC and stored on a downstream SCR catalyst. Then, during lean operation, stored NH₃ is used to reduce the lean NO_X emissions. Previous work conducted by Prikhodko et al. incorporated both TWC and SCR catalysts into a lean gasoline engine system, showing a projected fuel efficiency benefit of up to 10%, illustrating the viability of Passive SCR [2]. This work is focused on the long term viability of the system, examining the effects of hydrothermal aging and realistic concentrations of sulfur dioxide (SO₂) on the performance of the TWC in a passive SCR system.

Materials and Methods

Two commercially formulated catalysts were studied: A Pd-only TWC, and a NO_X storage TWC (NS-TWC) catalyst, containing Ce for oxygen storage capacity (OSC) and Ba for NO_X storage capacity (NSC). These catalysts were hydrothermally aged for 100 hours at 900°C prior to sulfation. During sulfation and evaluation, the catalysts are exposed to constant cycling between λ values of 0.97 (rich) and 2.0 (lean), where λ is a measure of AFR normalized to stoichiometric such that λ > 1 is fuel lean, and λ < 1 is fuel rich. This cycling is controlled by monitoring the outlet NO_X in the lean phase and the outlet NH₃ in the rich phase, and switching the phase when the integrated effluent reaches a threshold value. During both sulfation and evaluation of the catalyst the outlet concentrations of key gases are measured using a combination of gas-phase FTIR and mass spectrometry. The nitrogen species NO_X, NH₃, and nitrous oxide (N₂O) are monitored, while the reductants hydrogen (H₂), carbon monoxide (CO), and propane (C₃H₈) are also measured.

The catalysts are evaluated in the temperature range 350°C – 650°C in simulated exhaust conditions. After exposure to 2ppm SO₂ for 12.5 hours, another evaluation is conducted.  Catalysts were evaluated a final time after desulfation at 650°C. Individual reactions were investigated to determine more clearly determine the deactivation of specific reaction pathways in the simulated exhaust experiments. Variation of the rich phase AFR was also investigated as a potential route for mitigation of the effect of sulfur. The rich phase λ was varied in the range 0.92 – 0.98 with 2ppm SO₂ in the stream at both low temperature (350°C) and high temperature (650°C).

Results and Discussion

Sulfation caused an increase in outlet C₃H₈, CO, and N₂O, while the outlet H₂ and the production of NH₃ were both decreased. In each case, the initial catalyst performance was recovered through operation at 650°C for 3 hours. The decrease in H₂ could indicate a deactivation of water gas shift (WGS) or steam reformation reactions, which can play a large part in the production of NH₃ [3]. The reaction probe experiments provided evidence that when H₂ is used as a reductant, production of NH₃ is not significantly affected. However, when CO or C₃H₈ is used, the production of NH₃ was deactivated. Furthermore, both WGS and steam reforming were found to be significantly deactivated by sulfation. However, despite low H₂ production from WGS at 350°C, NH₃ production with CO as a reductant remained high, indicating that it may be formed through an alternative reaction pathway. Finally, decreasing the rich phase λ was shown to be effective in increasing the production of NH₃, shortening time spent in the rich phase. However, this decrease would also lead to an increase in fuel consumption during rich operation.

References

1. Parks, J.E., Parks, Prikhodko, V.Y., Partridge, W., Choi, J-S., Norman, K., Huff, S., and Chambon, P. SAE Int. J. Fuels Lubr. 3 962 (2010).
2. Prikhodko, V.Y., Parks, J.E., Pihl, J.A., and Toops, T.J. SAE Int. J. Engines 9 934 (2016).
3. Oh, S. H. and Triplett, T., Today 231 22 (2014).

Download poster presentation: