(147c) The Influence of Pt Incorporation on NOx Storage-Reduction Capacity of Nsr Monolith Catalysts

Introduction

Diesel engines and lean burn gasoline engines operates with an excess of oxygen, i.e. lean operation. Controlling the exhaust NO_x emission has been recognized as one of the most challenges aspects for lean-burn engine technology as the conventional three-way catalyts (TWC) are not effective in reducing NO_x in a lean exhaust due to the high oxygen content [1]. A NO_x storage/reduction (NSR) catalyst is among the most promising solutions to control NO_x in lean exhaust.

The NSR concept operates in a cyclic mode with alternating lean and rich periods. NO_x is stored in the catalyst during the long lean periods as nitrites or nitrates and the catalyst is regenerated during short fuel rich periods. During regeneration NO_x is released from the storage sites and reduced to N₂.

There is a lack of agreement in the literature in the order of incorporation of the precious metal and the storage component onto the alumina washcoat, and subsequent mutual interactions among the platinum, the storage component and the support material,  which ultimately can affect the dispersion and particle size of the active metal [2].

In the present work, we have prepared Pt/BaO/Al₂O₃ monolith catalysts by both impregnation and adsorption from solution methods, leading to important differences in the platinum particle size, determined both by H₂ adsorption and transmission electron microscopy (TEM). Also scanning electron microscopy (SEM) may be helpful in determining the distribution of platinum and barium on the alumina surface [3]. The aim of the study is to find some correlation between the platinum particle size and the NO_x storage-reduction capacity of the prepared monoliths.

Experimental

Three cordierite monoliths (A, B and C), 20 mm in length and diameter, were cut from a commercial monolith supplied by Corning, with a cell density of 400 cells per square inch. After a calcination step in air at 1000 ºC to remove possible impurities, the monoliths were washcoated with γ-alumina supplied by Saint-Gobain with a BET area of 190 m² g^-1. The monoliths were immersed for 10 seconds in a slurry consisted of a mixture of 10wt% of alumina in acetic acid (glacial 100%, Merck), afterwards the excess liquid was blown out with compressed air. Then the monoliths were dried at 110 ºC for 30 minutes. This procedure was repeated until the desired amount of alumina was washcoated. Finally the monoliths were calcined in air at 700ºC for 4 h.

The platinum incorporation was carried out by wetness impregnation (monolith A) or adsorption from solution (monoliths B and C). The metallic precursor was tetraamine platinum (II) nitrate supplied by Alpha Aesar. For impregnation the desired amount of salt was dissolved in an amount of water equivalent to that of the catalyst. After filling the channels with the aqueous solution, the samples were dried in air at 110 ºC for 24 h and finally calcined in air at 500 ºC for 4 h (monolith A). When the incorporation of platinum was done by adsorption, the remaining two monoliths were immersed in an aqueous solution with a specific initial concentration of platinum. The pH of solution was set at 11.6 by addition of NH₄OH (ammonia 25% as NH₃, Panreac). The monoliths were maintained in contact with the aqueous solution for 24 h to reach the equilibrium. Then the excess liquid was blown out and the catalysts were dried at room temperature in horizontal position. Finally one of the monoliths was calcined in air at 500 ºC (monolith B) and the other at 550 ºC (monolith C),  for 4 h.

Finally, the incorporation of the NO_x storage compound was carried out by wetness impregnation, by dissolving the desired amount of barium acetate and filling the monolith channels with this aqueous solution. Then the monoliths were dried at 110 ºC for 24 h and finally calcined in air at 500 ºC.

The characterization of the final catalysts was done by scanning electron microscopy (SEM, Sirion FEG) and transmission electron microscopy (TEM, Jeol JEM-2010F). The particle diameter of the metal was measured either by HAADF-STEM and hydrogen chemisorption (Micromeritics ASAP 2010).

Prior to reaction, the samples were reduced with a flow of 4%H₂/N₂ at 250 ºC for 20 minutes. The activity of the catalyst for storage and reduction of NO_x was evaluated in a flow reactor. All the gases were fed via mass flow controllers and the total flow rate in all experiments was 3365 ml min^-1, which corresponds to a space velocity of about 32100 h^-1. The inlet gas composition of the 150 s long lean period was 370 ppm NOand 6% O₂ using N₂ as the balance gas. For the 20 s rich period the composition was 370 ppm NO and 2.3% H₂. The outlet gases were continuously measured by using chemiluminiscence (NO_x) and paramagnetic (O₂) detectors (Rosemount Analytical).

Results and discussion

Notable differences in the dispersion and the homogeneity of the metallic compound over the monolithic catalyst were found. When the platinum was incorporated by wetness impregnation (A) the final distribution resulted in an egg-shell type, with a non-homogeneous distribution of platinum with a particle size up to 100 nm. When the incorporation method was adsorption from solution (B and C) a homogeneous distribution and finely dispersed platinum were achieved, with a mean particle diameter of 1.3 nm (Figure 1).

Figure 1. Particle size distribution and HAADF-STEM image, for the Pt/Al₂O₃ sample prepared by adsorption from solution, calcined at 500ºC.

Another crucial step in the performance of a Pt/BaO/Al₂O₃ monolithic catalyst is the calcination temperature. From chemisorption experiments, an important decrease in the platinum dispersion between 500 and 550 ºC temperature calcinations was observed, from near 100% to 48%. On the other hand, as revealed by TEM analysis, no significant differences in Pt dispersion were observed between the samples calcined at 450 and 500ºC. It could be concluded that a calcination step at 500ºC is suitable for achieving both a high Pt dispersion and an appropriate thermal stabilization of the monolith, thereby preventing structural changes during reactions.

The catalytic activity of the prepared monoliths was evaluated in lean and rich periods. Two consecutive storage-reduction cycles are shown in Figure 2. During the first seconds the NO_x signal was negligible (efficient storage) but, the concentration of NO_x increased as the lean cycle extended because the storage capacity of the catalysts was limited. The objective in practice is to reach the larger storage capacity with the minimun length rich period to reduce or regenerate the storage sites.

Figure 2. Outlet NO_x concentration for two lean/rich cycles over the Pt/BaO/Al₂O₃ monolith B, at 300 ºC.

The NO_x storage-reduction behavior of the prepared monoliths was evaluated through the storage ratio defined as:

where F_NO⁰ is the feed molar flow rate of NO, and F_NOx is the outlet molar flow rate of NO_x.

Figure 3 shows the NO_x conversion (Equation 1) obtained with Pt/BaO/Al₂O₃ monoliths A, B and C under the operating conditions above mentioned at various temperatures from 250 to 400ºC in step of 50ºC. Several conclusions can be deduced from Figure 3: i) For all three samples, there is an optimal reaction temperature at which NO_x storage is maximum, ii) The activity is ranked as follows: B > C > A, iii) The conversion maximum is shifted to lower temperatures, iv) The best performance corresponds to a conversion of 70% attained with the monolith B, i.e. prepared by adsorption from solution and calcined at 500ºC, when the reaction was carried out around 300ºC. This optimal behavior was correlated to the higher dispersion and a more homogeneous distribution of platinum.

Figure 3. Comparison of catalytic performance of the Pt/BaO/Al₂O₃ monoliths A, B and C.

References

[1] J.R. González-Velasco, M.A. Gutiérrez-Ortiz, J.L. Marc, M.P. González-Marcos, G. Blanchard (2001), Appl. Catal. B: Environmental, 33, 333-314.

[2] U. Elizundia, R. López-Fonseca, I. Landa, M.A. Gutiérrez-Ortiz, J.R. González-Velasco (2007), Topics in Catalysis, 42-43 (1-4) 37-41.

[3] J. Dawody, L. Eurenius, H. Abdulhamid, M. Skoglundh, E. Olsson, E. Fridell (2005), Appl. Catal. A: General, 296, 157-168.

[4] M. Fuchs, B. Jenewein, S. Penner, K. Hayek, G. Rupprechter, D. Wang, R. Schlögl, J.J. Calvino, S. Bernal (2005), Appl. Catal. A: General, 294, 279-289.