Impact of Rh Oxidation State on NOx Reduction Performance of Multi-Component Lean NOx Trap (LNT) Catalyst.
Compression ignition (diesel) engines are utilized due to their desirable torque characteristics and high fuel economy. Over the last decade, the environmental regulations for NOx emissions from internal combustion engines have become increasingly stringent, which presents major challenges to catalytic reduction of NOx for lean burn engines.
Lean NOx traps (LNT or NOx storage and release catalyst or NSR; NOx absorber catalyst, NAC), represents one of the technologies, capable of meeting increasing emissions challenges as shown in multiple commercialized diesel aftertreatment systems . The NOx reduction on a LNT occurs through a cyclic process achieved transition from fuel-lean (excess oxygen) operation to fuel-rich (excess reductant) operation. During the typical lean conditions, the NO is oxidized to N[O.sub.2] at the catalyst surface and trapped as nitrate or nitrite. During rich operation, the nitrite and nitrate are released and reduced to [N.sub.2] or other products.
In a practical application, an LNT catalyst is typically operated under three different types of conditions, including deNOx (lean/rich cycling in the temperature range of 150-450[degrees]C), desoot (extended lean exposure in the temperature range of 500-700[degrees]C) and desulfation (lean/rich cycle in the temperature range of 500-700[degrees]C). Therefore, the redox state of an LNT catalyst is highly dynamic according to its specific operation history. Significant efforts have been made to explore multiple functions of the LNT catalyst, such as NO oxidation, NOx storage, NOx release and NOx reduction [1, 3]. Nevertheless, limited attention has been payed to the understanding of the effect of the oxidation state of catalyst surface, and specifically of the precious metal components, on the performance of an LNT catalyst although a lot of study had been done for TWC previously [4,5,6,7,8,9,10,11,12, 13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,3 5,36,37,38,].
In this study it had been observed that the oxidation of LNT catalyst can severely decrease its NO reduction activity. In particular, we have demonstrated that the oxidation of Rh plays the key role for this type of reversible deactivation. Detailed kinetics study for the reaction of NO reduction to [N.sub.2] with different aging stages and oxidation states specifically of Rh, has been further performed to elucidate the underlying process and its significance to the overall cyclic operation.
EXPERIMENTAL PROCEDURE AND MATERIALS
Experimental set up used in this work is a bench-flow reactor which allows the catalyst to be exposed to different temperatures and different gas compositions, including rapid and well-defined gas composition transients, in a controlled manner. A micro-core (~6 mm x 25 mm) of a catalyst is placed into a horizontal quartz reactor tube with 1/4" inner diameter. The required gas species are blended from research grade gases using MKS mass flow controllers to form the desired gas composition. Two thermocouples, in 0.3mm-walled glass wells are placed at the inlet and outlet of the catalyst monolith. The inlet flow of the gaseous species is controlled by the flow rates of the mass flow controllers. The micro-reactor is equipped with a chemiluminescent NOx analyzer (California Instrument, 400HCLD, modified for full flow of the sample gas) to measure the outlet NOx concentrations and a Mass Spectrometer (Agilent, 5973N) to measure outlet propylene concentration. More detailed description about the system could be found in our previous work .
The catalysts used for our study were two fully formulated LNTs manufactured by Johnson Matthey. Both catalysts were received in the form of cylindrical, washcoated, honeycomb monoliths, 9 inches in diameter and 5 inches long. Total precious metal loadings on these two commercial LNT catalysts were 60 and 72 g/[ft.sup.3], both with the Pt/Rh ratio of 4:1. The total surface area of the degreened catalyst/monolith combination was measured to be about 30-40 [m.sup.2]/g by [N.sub.2] adsorption using BET method. There was no palladium in these samples.
Additionally, two model catalysts were provided by Johnson Matthey, and used to help interpret the behavior of the fully-formulated LNTs. The first is Rh coated on the ceria/zirconia/alumina (12g Rh/[ft.sup.3]) and the second catalyst is the Pt coated on alumina (48g Pt/[ft.sup.3]) to specifically decouple the effect of oxidation state of Rh and Pt on the NO reduction function over the commercial LNT catalyst.
In Table 1 we listed all the gas mixtures being used during the lean, rich or cyclic operation being reported in the following study. Since engine out gas composition during rich operation always includes hydrocarbons, CO and [H.sub.2], the reductant mixture containing CO, [H.sub.2] and [C.sub.3][H.sub.6] with different ratios were explored in which 4% CO, 1.25% [H.sub.2] and 0.2% propylene (Rich2) was used as the standard rich mixture, unless otherwise stated. The 5% [H.sub.2]O and 5% C[O.sub.2] were also always present in both lean and rich mixtures, to ensure practical relevance of the test results. In the following discussion, the composition of the rich mixture being used is reported as the concentrations of [H.sub.2]/CO/[C.sub.3][H.sub.6]. For example, Rich2 is reported as 1.25/4/0.2, corresponding to the molar composition of 1.25% [H.sub.2], 4.0% CO and 0.2% [C.sub.3][H.sub.6].
Figure 1 (a) and (b) describes the procedure employed to explore the reactivity of an oxidized catalyst and reduced catalyst in our study. As shown in Figure 1 (a), the catalyst was first heated up to 650[degrees]C under lean conditions and the temperature was maintained at 650[degrees]C for 1h to oxidize the catalyst. The oxidation temperature, 650[degrees]C, was chosen as the representative of on-engine desoot conditions. Samples treated in this manner are referred to as oxidized. In Figure 1(a) the temperature was reduced to 300[degrees]C or 350[degrees]C under lean environment then 5 cyclic NOx reduction with 50s Lean/10s Rich (rich2) were performed. This type of treatment was used to probe the cyclic performance of the catalyst, relevant to its practical operation. The catalyst was then pre-reduced by performing 20 Lean (50s)/Rich (10s) cycles using rich composition (Rich2) at 400[degrees]C. Samples treated in this manner are referred to as reduced. The sample was cooled down to 300[degrees]C or 350[degrees]C in Figure 1(a) under the net rich condition to avoid any re-oxidation of the catalyst. Then the cyclic performance was quantified again as shown in Figure 1(a).
In Figure 1(b), catalyst was oxidized or reduced with the same manner being specified in Figure 1(a). After the oxidation, the catalyst temperature was reduced to 140[degrees]C under lean environment then the system was switched to rich (rich2) condition for 30 mins with 200 ppm flow of NO. The temperature was ramped up from 140 to 400[degrees]C to measure the NO light-off performance using TPRx. This type of treatment was used to probe the NO reduction function alone, independently of the other steps in the overall cycle. After the reduction, the sample was again cooled down to 140[degrees]C in Figure 1(b) under the net rich condition to avoid any re-oxidation of the catalyst. NO light-off using TPRx was quantified again as shown in Figure 1(b).
EXPERIMENTAL RESULTS AND DISCUSSION
1. Effect of the Redox History on Cyclic Performance
Figure 2 shows the effect of the redox history of a fully formulated LNT catalyst with 48g/[ft.sup.3] Pt and 12g/[ft.sup.3] Rh on its deNOx performance, using the test procedure described in Figure 1(a). As one can see, cyclic operation of a pre-reduced catalyst at the employed conditions produced no measurable slip of unconverted NOx, resulting in the cycle-average NOx conversion above 99%. In contrast, pre-oxidized catalyst produced large spikes of unconverted NOx at both 300[degrees]C and 350[degrees]C, resulting in cycle-average NOx conversion values of 76% and 87%, respectively. These spikes abated after several cycles, likely due to the catalyst changing its oxidation state. At 350[degrees]C, it only took one cycle to overcome the effect of the oxidative pretreatment, while at 300[degrees]C three cycles were required to accomplish that.
2. Effect of Redox History on NO Reduction under TPRx Conditions
As shown in Figure 2, deNOx performance of the LNT catalyst is sensitive to its redox history. However, these results for a fully-formulated catalyst confound the processes of NOx storage and release, NO oxidation and reduction, and redox state change of the LNT catalyst itself. In the following set of experiments, different approaches have been utilized to de-convolute all the effects in order to better understand the effect of the redox state on the commercial LNT catalyst performance. The first attempt is to probe the effect of redox state of the catalyst on NO reduction only rather than the other function of the LNT catalyst.
Figure 3 reports NO light-off curve with the reduced and oxidized, fully-formulated LNT catalysts treated under the conditions specified in Figure 1 (b). To quantify the NO reduction activity of the catalysts studied in this work, we use the value of [T.sub.50], defined as the temperature at which 50% of NO conversion is achieved. As shown in Figure 3(a), by using a fully formulated LNT catalyst (Pt: 48g/[ft.sup.3], Rh: 12g/[ft.sup.3]), with the exact same sample but different pretreatment conditions, the [T.sub.50] can be reduced from 217[degrees]C for an oxidized catalyst to 173[degrees]C for a reduced one. Figure 3(b) shows very similar behavior on a different commercial LNT catalyst, with a somewhat different formulation and precious metal loadings of 58g/[ft.sup.3] Pt and 14g/[ft.sup.3] Rh, indicating the effect of redox state of precious metal on the reactivity of the LNT catalyst is not specific to a single formulation. In the following test, the commercial LNT catalyst with 48 g Pt/[ft.sup.3] and 12 g Rh/[ft.sup.3] was used as a representative for the commercial catalyst for the kinetics study.
3. Reactivity of Rh and Pt in the NO Reduction Reaction
In the following experiment two model catalysts containing only Rh as a the precious metal (Rh/CeZrOx, 12g Rh/[ft.sup.3]) or only Pt (Pt/Alumina, 48 g Pt/[ft.sup.3]) were used to further de-couple the effect of redox state on the reactivity of commercial LNT catalysts, as reported in the above sections. Rich2 gas composition (1.25/4/0.2) was used during these experiments.
The oxidized or reduced Rh-only and Pt-only catalysts were pretreated with the condition specified in Figure 1(b). As we can see in Figure 4, at the same oxidation state, the Pt-only catalyst showed much lower reactivity toward NO reduction than the Rh-only catalyst. Also, the reduced catalyst always had better light-off activity than oxidized catalyst. The NO light-off temperature shifted 47[degrees]C higher for oxidized Rh compared with reduced Rh. The difference between the reduced and oxidized Pt was much smaller than Rh, in which the oxidized Pt only catalyst had 24[degrees]C higher light-off temperature than the reduced Pt. Overall, Rh component plays the key role in NO reduction, and its reactivity is particularly strongly affected by the oxidation state.
4. NO Reduction Kinetics over LNT Catalyst
In order to further investigate the behavior of the Rh component, we have subjected the results of the TPRx test to kinetic analysis, in the differential range of NO conversion (<40%). The NO reduction under rich condition was assumed to follow the Arrhenius equation, with the first-order dependence of the inlet NO concentration based on previous study [32-33].
Rate = A exp(-[Ea/RT])[[NO].sub.in] Eq-1
In Figure 5 we have replotted the in Arrhenius coordinates the light-off curves for reduced Rh/CeZrOx and Pt/Alumina catalysts, shown in Figure 4. As seen in Figure 5, the apparent activation energy for NO reduction are significantly different over Rh/CeZrOx and Pt/Alumina: the former yield the Ea value of 157 kJ/mol, while the latter 94 kJ/mol. Despite substantially lower apparent activation energy, Pt-based catalyst shows much lower activity because of the substantially lower pre-exponential factor, translating into a lower density of active sites.
In Figure 6 we have replotted in Arrhenius coordinates NO light-off curves of a commercial LNT catalyst with reduced and oxidized pretreatment from Figure 3(a). Interestingly, the apparent activation energy for this commercial catalyst was much closer to the value obtained for the Rhonly catalyst than to the Pt-only one, further supporting the assumption that Rh is the key NO reduction component in the fully formulated catalyst.
As we can see in Figure 6, the Arrhenius plot for the LNT catalyst under reduced or oxidized state looked drastically different. The slope of the Arrhenius equation of the reduced catalyst is a constant across a broad range of temperatures, while for the oxidized catalyst it changed significantly in the course of the reaction. This observation suggested that during the reaction the catalyst surface of the reduced catalyst remains in the same state, while the surface of the oxidized catalyst is changing significantly. This phenomena would further complicate a kinetic study on the oxidized catalyst. Therefore, in the following study, we would focus on reduced catalyst with different aging conditions to obtain the kinetics for NO reduction.
Figure 7(a) shows NO light-off over a reduced commercial LNT catalyst with different histories, resulting either from the aging in the field or in the lab. The majority of the field-aged samples were harvested from the rear of the catalyst brick. Therefore, the sample is relatively free of any chemical contamination while hydrothermal aging is the key aging mechanism. According to the light-off curves shown in Figure 7(a), the hydrothermal aging of the commercial LNT catalyst could be clearly differentiated into three different groups, in which the mildly aged catalysts has NO light-off temperature [T.sub.50] of about 180[degrees]C, the medium aged LNT catalyst - about 186[degrees]C, and the severely aged LNT catalyst has the [T.sub.50] ranging from 200 to 210[degrees]C. Unfortunately, only the severely aged LNT catalyst sample #1 has a well-known thermal history, because it was hydrothermally aged in the lab under lean condition at 750[degrees]C for 24hrs. Other samples were field-returned, with no detailed thermal history. Nevertheless, according to the light-off curves, most of them are mildly or moderately aged and only two of them showed severe thermal aging.
The differential conversion part of the NO light-off curves in Figure 7(a) were used to obtain the apparent activation energy, based on the Arrhenius plot shown in Figure 7(b). All of the curves in Figure 7(b) showed good linearity in the conversion range from 5% to 40%. In addition, linearity of the Arrhenius curves suggests that the oxidation state of all of these reduced samples was not changing during the NO reduction process. As one can see in Figure 7(b), despite different thermal aging, Arrhenius slopes were quite similar among the studied fully formulated LNT samples with different aging histories. Kinetic parameters resulting from this Arrhenius analysis are shown in Table 2. The average apparent activation energy was found to be 179[+ or -]15 kJ/mol, which is very close to the Rh only catalyst. The last column in Table 2 gives the pre-exponential factor (A) number in the Arrhenius equation. As expected, the lower of the [T.sub.50], the higher is the pre-exponential factor. For the least thermally aged catalyst this value was 1.41x[10.sup.22], compared to 7x[10.sup.20] for the most severely aged catalyst. Therefore, at the same temperature, the least aged catalyst is 20 times more active than the most severely aged catalyst. It is well accepted that the thermal aging results in the sintering of the precious metal, which leads to larger particle size and poor metal dispersion . In agreement with these findings, the pre-exponential number determined here could be used to quantify changes in Rh dispersion on the LNT catalysts with different thermal aging history.
The effect of different reductant type on the NO reduction over a lab-aged catalyst (Pt: 48 g/[ft.sup.3], Rh: 12 g/[ft.sup.3]) was explored using the same approach. Figure 8(a) reports NO light-off curves for the reduced lab aged catalyst, by using different rich mixtures during the TPRx. Figure 8(b) shows Arrhenius plots for the same experiments. The resulting kinetic parameters are summarized in Table 3. The NO light-off temperature with the same lab aged catalyst by using red1 (0/0/0.8), red2 (1.25/4/0.2), red3 (1.25/1.25/0.2) and red4 (0/1.25/0.2) are 166, 172, 191, 183 kJ/mol, respectively. The light off temperature by using red4 (0/1.25/0.2) is very similar to pure [C.sub.3][H.sub.6] (0/0/0.8), which both showed relatively high [T.sub.50] in Figure 8(a). The previous study had indicated that under rich condition, majority of [C.sub.3][H.sub.6] would be converted to CO, which would explain why [C.sub.3][H.sub.6] and CO as the reductant showed very similar light-off behavior for NO reduction. The addition of [H.sub.2] (red3, 1.25/1.25/0.2) significantly improved the reaction rate which was indicated by much lower [T.sub.50] and much larger A number in Table 3. The addition of extra 2.5% of CO (red2, 1.25/4/0.2) did not further improve the light-off temperature or A number compared with red3, suggesting that the total richness (0.86 vs. 0.91) is not the limiting factor, but the addition of [H.sub.2] in the feed-stream. One of the factors for this performance improvement by adding [H.sub.2] is that the reactivity of [H.sub.2] on NO reduction is much higher than CO. Therefore, once [H.sub.2] is added, it became the dominant reductant.
In addition to the promotion of NO reduction using [H.sub.2] as a reductant, the surface coverage of the catalyst will be improved in favor of the NO reduction. Hauptmann et al. had observed the promotion effect of [H.sub.2] on CO oxidation over precious metal surface. It was proposed that for NO reduction with CO, the addition of [H.sub.2] would help to promote the surface oxygen consumption by adsorbed H* as well as hydroxyl group while the reaction of CO with the surface oxygen is slow . This would allow more available sites for NO dissociation. It is known that CO would poison the precious metal surface at low temperature . In our study, we had also tested the NO reduction with 1.25% [H.sub.2] only as the reductant specified as rich 5. The test had indicated that in the absence of CO, the NO reduction performance could achieve 100% conversion even at 100[degrees]C over reduced LNT catalyst, which is much lower than any of the NO light-off temperature with the other reductant mixture in the presence of CO which again confirmed that CO poisoning is present during NO reduction. The previous study by Novakova had shown that actually in the absence of CO, Pt is more active than Rh on NO reduction with [H.sub.2] as the reductant . Therefore, under the NO light-off with no CO, Pt acts as the active component for the reaction over the commercial LNT catalyst.
The total scope of all the experiments in Table 2 and Table 3 gave 180[+ or -]14 kJ/mol as the apparent activation energy. Granger et al. reported the activation energy of NO reduction by CO over Rh/[Al.sub.2][O.sub.3] to be 181kJ/mol [31-32]. The kinetics study and DFT calculation of NO dissociation by using single crystal of Rh gave 200 kJ/mol as the activation energy . Such reasonable agreement of the apparent activation energy with the literature once again indicates that Rh plays the key role for NO reduction. Interestingly, the activation energy of the NO reduction by using different rich mixture (red1, red2, red3 and ref4) as the reductant or using catalysts with different thermal histories (Table 2) are very similar, suggesting that the rate limiting step is the same. Based on the literature, NO dissociation is considered to be the rate limiting step for NO reduction for reduced Rh. The above results are consistent with the earlier findings, that NO reduction with CO on Rh-based proceeds through competitive adsorption on a single type of active site, with NO dissociation being the rate-limiting step [31-32].
NOx reduction performance of commercial LNT catalysts was shown to be strongly affected by their oxidation history. During both cyclic operation and temperature-programmed rich NOx reduction, all the studied LNT samples consistently showed substantially better performance in the reduced state, than in the oxidized state. The catalysts can change their state in the process of operation, however that transition is not instantaneous and some amount of NOx may slip unreduced, while the catalyst surface is being transformed.
Further study using model catalysts, including Rh-only or Pt-only ones, and using various reductant mixtures including [H.sub.2], CO and [C.sub.3][H.sub.6], indicated that Rh is the key catalyst for NO reduction under net reducing conditions. Rh was also found to be responsible for the large hysteresis between oxidized and reduced states in the commercial LNTs. Kinetics analysis indicated that NO reduction over reduced LNT catalyst followed the Arrhenius equation with the first order dependence on NO inlet concentration and the apparent activation energy of 180[+ or -]14 kJ/mol, regardless of the thermal aging or the type of reductant. Across the broad range of catalyst ages, the respective changes in their performance could be represented by the change in the pre-exponential factor, consistent with Rh sintering being the leading mechanism of aging.
These findings provide insights into optimization of the catalyst composition, as well as modeling and optimization of the operation strategy, in order to minimize the effect of redox history on the NOx reduction efficiency of LNT catalysts. The derived knowledge should be also applicable to three-way catalysts, which operate under similar conditions.
[1.] Epling, W.S., Campbell, L. E., Yezerets, A., Currier, N. W., and Parks J. E., II, "Overview of the Fundamental Reactions and Degradation Mechanisms of NOx Storage/Reduction Catalysts," Catal. Reviews 46: 163-245, 2004
[2.] Novakova, J., "Reduction of NO by Hydrogen versus Reduction by CO over Pt, Pd and Rh Clusters in NaX Zeolite," Appl. Catal. B: Environ. 30: 445-457, 2001
[3.] Roy, S., Hedge, M.S., and Madras, G., "Catalysis for NOx abatement" Appl. Energy 86: 2283-2297, 2000
[4.] Nakatsuji, T., and Komppa, V., "A catalytic NOx reduction system using periodic two steps: an operation in oxidizing conditions and a relatively short operation in reducing conditions," Appl. Cat. B: Environ 30: 209-223, 2001
[5.] Oh, S. E., and Carpenter, J. E., "The Oxidation State and Catalytic Activity of Supported Rhodium," J. Catal. 80: 472-478, 1983
[6.] Root, T. W., Schmidt, T. W., and Fisher, G., "Adsorption and reaction of nitric oxide and oxygen on Rh(111)," Surface Science 134: 30-45, 1983
[7.] Fisher, G. B., Oh, S. H., Carpenter, J. E., Dimaggio, C. L., Schmieg, S. J., Goodman, D. W., Root, T. W., Schwartz, S. B., Schmidt, L. D. "Mechanisms of the Carbon Monoxide Oxidation and Nitric Oxide Reduction Reactions over Single Crystal and Supported Rhodium Catalysts: High Pressure Rates Explained Using Ultrahigh Vacuum Surface Science," Studies in Surface Science and Catalysis 30:215-220, 1987
[8.] Root, T.W., Schmidt, L.D. and Fisher, G. B. "Nitric oxide reduction by CO ON Rh(111): Temperature programmed reaction," Surface Science 150: 173-192, 1985
[9.] Grunwaldt, J., Basini, L., and Clausen, B. "In-situ EXAFS Study of Rh/[Al.sub.2][O.sub.3] Catalysts for Catalytic Partial Oxidation of Methane," J. Catal. 200: 321-329, 2001
[10.] Xu, J., and Overbury, S. H., "[H.sub.2] reduction of Ce[O.sub.2] (111) surface via boundary Rh-O mediation," J. Catal. 222:167-173, 2004
[11.] Tomczak, D.C., Lei, G. D., Schunemann, Trevino, V., H., and Sachtler, W. M. H., "Chemistry of rhodium in zeolite Y," Micro. Mater. 5: 263-278, 1996
[12.] Newton, M. A., Dent, A.J., Fiddy, S. G., Jyoti, B., and Evans, J., "Combining diffuse reflectance infrared spectroscopy (DRIFTS) dispersive EXAFS and mass spectrometry with high time resolution Potential limitations and application to the study of NO interaction with supported Rh catalysts," Catal. Today 126: 64-72, 2007
[13.] Holles, J. H., Switzer, M.A., and Davis, R. J. "Influence of Ceria and Lanthana Promoters on the Kinetics of NO and [N.sub.2]O Reduction by CO over Alumina-Supported Palladium and Rhodium," J. Catal. 190: 247-260, 2000
[14.] Penner, S., Jenewein, B., Wang, D., Schlogl, R., and Hayek, K., "Structure-activity correlation in thin film model catalysts: CO hydrogenation on Rh/VOx Part I: The morphology, composition and structure of vanadia-supported and -promoted Rh particles upon oxidation and reduction," Appl. Catal. A: Gen 308: 31-42, 2006
[15.] Gao, F., Wang, Y., and Goodman, D. W., "CO/NO and CO/NO/[O.sub.2] reactions over a Au-Pd single crystal catalyst," J. Catal. 268: 115-121, 2009
[16.] Brown, W. A., and King, D. A., "NO Chemisorption and Reactions on Metal Surfaces: A New Perspective," J. Phys. Chem. B 104: 2578-2595, 2000
[17.] Cao, C., Bourane, A., Schlup, J. R., and Hohn, K. L., "In situ IR investigation of activation and catalytic ignition of methane over Rh/[Al.sub.2][O.sub.3] catalysts," Appl. Catal. A: Gen 344: 78-87, 2008
[18.] Dent, A. J., Evans, J., Fiddy, S. G., Jyoti, B., Newton, M. A., and Tromp, M., "Rhodium dispersion during NO/CO conversions," Angew. Chem. Int. Ed. 46: 5356-5358, 2007
[19.] Zhong, F., Xiao, Y., Weng, X., Wei, K., Cai, G., Zheng, Y., and Zheng, Q., "Thermally stable Ce[O.sub.2]-Zr[O.sub.2]-[La.sub.2][O.sub.3] Ternary Oxides Prepared by Deposition-Precipitation as Support of Rh Catalyst for Catalytic Reduction of NO by CO," Catal. Lett. 133: 125-133, 2009
[20.] Grunwaldt, J., Basini, L., and Clausen, B. S., "In Situ EXAFS Study of Rh/[Al.sub.2][O.sub.3] Catalysts for Catalytic Partial Oxidation of Methane," J. Catal. 200: 321-329, 2001
[21.] Legage, M., Visser, T., Soulimani, F., Beale, A.M., Iglesias-Juez, A., Eerden, A., and Weckhuysen, B. M., "Promotion Effects of the Oxidation of CO over Zeolite-supported Rh Nanoparticles," J. Phys. Chem. C 112: 9394-9404, 2008
[22.] Manuel, I., Chaubet, J., Thomas, C., Colas, H., Matthess, N., and Djega-Mariadassou, G., "Simulation of the transient CO oxidation over [Rh.sup.0]/Si[O.sub.2] and [Rh.sup.x+]/[Ce.sub.0.68]Z[r.sub.0.32][O.sub.2] catalysts," J. Catal. 224: 269-277, 2004
[23.] Xiao, L., and Schneider, W. F., "Influence of alumina supports on oxygen binding to Pd, Ag, Pt and Au," Chem. Phys. Lett. 484: 231-236, 2010
[24.] Djega-Mariadassou, G., Fajardie, F., Tempere, J., Manoli, J., Touret, O., and Blanchard, G., "A general model for both three-way and deNOx catalysis: dissociative or associative nitric oxide adsorption, and its assisted decomposition in the presence of a reductant Part I. Nitric oxide decomposition assisted by CO over reduced or oxidized rhodium species supported on ceria," J. Molecular Catal. A: Chem. 161: 179-189, 2000
[25.] Adamowska, M., Krzton, A., Najbar, M., Camra, J., Djega-Mariadassou, G., and Costa, P. D., "Ceriazirconia-supported rhodium catalyst for NOx reduction from coal combustion flue gases," Appl. Catal. B: Environ. 90: 535-544, 2009
[26.] Hu, Y., Grififiths, K., and Norton, P. R., "Surface science studies of selective catalytic reduction of NO: progress in the last ten years," Surf. Sci. 603: 1740-1750, 209
[27.] Kaspar, J., Fornasiero, P., and Hickey, N., "Surface science studies of selective catalytic reduction of NO: progress in the last ten years," Catal. Today 77: 419-449, 2003
[28.] Neyertz, C., Volpe, M., Perez, D., Costilla, I., Sanchez, M., and Gigola, C., "NO reduction with CO in the presence and absence of [H.sub.2]O over Pd/[gamma]-[Al.sub.2][O.sub.3] and Pd-VOx/[gamma]-[Al.sub.2][O.sub.3] catalysts: The formation of HNCO, N[H.sub.3] and stable surface species," Appl. Catal. A: Gen 368: 146-157, 2009
[29.] Lindholm, A., Currier, N. W., Fridell, E., Yezerets, A., and Olsson, L., "NOx storage and reduction over Pt based catalysts with hydrogen as the reducing agent: Influence of [H.sub.2]O and C[O.sub.2]," Appl. Catal. B: Environ. 75: 78-87, 2007
[30.] Li, J., Currier, N., Yezerets, A., Chen, H. et al., "Evaluation of Spatially Resolved Performance of NOx Adsorber Catalysts," SAE Technical Paper 2009-01-0275, 2009, doi:10.4271/2009-01-0275.
[31.] Granger, P., Lecomte, J. J., Dathy, C., Leclercq, L., and Leclercq, G., "Kinetics of the CO+NO Reaction over Rhodium and Platinum-Rhodium on Alumina," J. Catal. 175: 194-203, 1998
[32.] Granger, P., Delannoy, L., Lecomte, J. J., Dathy, C., Praliaud, H., Leclercq, L., Leclercq, G., "Kinetics of the CO+NO Reaction over Bimetallic Platinum-Rhodium on Alumina: Effect of Ceria Incorporation into Noble Metals," J. Catal. 207: 202-212, 2002
[33.] Huuhtanen, M., Kolli, T., Maunula, T., Keiski, R.L. "In situ FTIR study on NO reduction by [C.sub.3][H.sub.6] over Pdbased catalysts," Catal. Today 75: 379-384, 2002
[34.] Berlowitz, P.J., Peden, C. H. F., and Goodman, D. W., "Kinetics of CO Oxidation on Single-Crystal Pd, Pt, and Ir," J. Phys. Chem. 92: 5213-5221, 1988
[35.] Salomons, S., Hayes, R.E., and Votsmeier, M., "The promotion of carbon monoxide oxidation by hydrogen on supported platinum catalyst," Appl. Catal. A: Gen 352: 27-34, 2009
[36.] Bhatia, D., Harold, M. P., Balakotaiah, V., "Kinetic and bifurcationanalysis of the cooxidation of CO and [H.sub.2] in catalytic monolith reactors," Chemical Engineering Science 64: 1544-1558, 2009
[37.] Hauptmann, W., Votsmeier, M., Vogel, H., and Vlachos, D.G., "Modeling the simultaneous oxidation of CO and [H.sub.2] on Pt - Promoting effect of [H.sub.2] on the CO-light-off," Applied Catalysis A: General 397: 174-182, 2011
[38.] Fernandes, D. M., Scofield, C. F., Neto, A. A., Baldini Cardoso, M. J., and Zanon Zotina, F. M., "Thermal deactivation of Pt/Rh commercial automotive catalysts," Chemical Engineering Journal 160: 85-92, 2010
Junhui Li, Neal Currier, and Aleksey Yezerets
Hai-Ying Chen, Howard Hess, and Shadab Mulla
Johnson Matthey Inc.
The authors would like to thank Jim Lucas for his help with collecting the experimental data.
Table 1. Gas Composition during Catalyst Performance Test Gas Species Rich Red1 Red2 Red3 NO (ppm) 200 200 200 [H.sub.2](%) 0 1.25 1.25 CO (%) 0 4.00 1.25 [C.sub.3][H.sub.6] (%) 0.8 0.2 0.2 [.sub.2] (%) 0 0 0 [H.sub.2]O (%) 5 5 5 C[O.sub.2] (%) 5 5 5 [H.sub.2]/CO/[C.sub.3] [H.sub.6] 0/0/0.8 1.25/4/0.2 1.25/1.25/0.2 Lambda 0.86 0.86 0.91 Gas Species Rich Red4 Red5 Lean NO (ppm) 200 200 0/200 [H.sub.2](%) 0 1.25 0 CO (%) 1.25 0 0 [C.sub.3][H.sub.6] (%) 0.2 0 0 [.sub.2] (%) 0 0 5 [H.sub.2]O (%) 5 5 5 C[O.sub.2] (%) 5 5 5 [H.sub.2]/CO/[C.sub.3][H.sub.6] 0/1.25/0.2 1.25/0/0.2 - Lambda 0.93 0.93 1.31 Table 2. NO light-off temperature [T.sub.50] and kinetic parameters for reduced LNT catalyst (Pt: 48 g/[ft.sup.3], Rh: 12 g/[ft.sup.3]) with different thermal aging histories Intercept, A Aging Sample [T.sub.50] Ea based on (x10 (21)) History ([degrees]C) (kJ/mol) average Ea Mildly aged 1 178 204 51.0 14.09 on a vehicle 2a 179 189 51.0 14.09 2b 180 171 51.0 14.09 2c 180 183 50.9 12.75 Medium aged 1 185 179 50.4 7.73 on a vehicle 2 186 156 50.5 8.55 3 187 161 50.3 7.00 Severely aged 1 201 172 48.8 1.56 on a vehicle 2 210 176 48.0 0.70 or lab 3 206 195 48.0 0.70 Average Ea 179 [+ or -]15kJ/mol Table 3. NO light-off temperature [T.sub.50] and kinetic parameters for reduced, lab-aged LNT catalyst (Pt: 48 g/[ft.sup.3], Rh: 12 g/[ft.sup.3]) with different reductants used during TPRx Reductant compositio [T.sub.50] Ea(kJ Sample n in TPRx ([degrees]C) /mol) [[H.sub.2]/ CO / [C.sub.3][H.sub.6]] Average of all samples 1.25/4/0.2 179+15 with Red2 1 1.25/4/0.2 201 172 1 0/0/0.8 232 167 1 1.25/1.25/0.2 196 195 1 0/1.25/0.2 227 184 1 1.25/0/0 <100 Intercept, based on A Sample (x10 (21)) average Ea Average of all samples with Red2 1 48.8 1.56 1 45.9 0.09 1 49.2 2.33 1 46.4 0.14 1
|Printer friendly Cite/link Email Feedback|
|Author:||Li, Junhui; Currier, Neal; Yezerets, Aleksey; Chen, Hai-Ying; Hess, Howard; Mulla, Shadab|
|Publication:||SAE International Journal of Engines|
|Date:||Sep 1, 2016|
|Previous Article:||Lubricant-Derived Ash Impact on Gasoline Particulate Filter Performance.|
|Next Article:||An Approach to Controlling N2O Emission on HDD On-Road Applications.|