(617hl) The Mechanism of CO2 Hydrogenation over Pd/Al2O3: A Steady State Isotopic Transient Kinetic Analysis (SSITKA)/Operando-FTIR Spectroscopy Investigation 

Xiang Wang, János Szanyi*

Institute for Integrated Catalysis, Pacific Northwest National Laboratory, Richland, WA 99352, USA

*Corresponding author: janos.szanyi@pnnl.gov

1. Introduction

Understanding the critical steps involved in the heterogeneous catalytic CO₂ reduction has attracted a lot of attention recently. In order to fully understand the mechanism of this reaction the determination of both the rate-determining steps and reaction intermediates are vital. Steady-State Isotopic Transient Kinetic Analysis (SSITKA) is one of the most powerful techniques used to investigate the elementary steps under steady-state reaction conditions. This technique provides valuable information on mean resident lifetime of surface intermediates, surface concentrations of adsorbed reactant species and an upper bound of the turnover frequency.[1-3] Coupling SSITKA with operando-FTIR spectroscopy allows us to discriminate between active and spectator species present on the catalytic surface under steady state reaction conditions. Most of the reaction cells applied in these types of studies possess a large volume leading to long gas hold-up times, which makes it impossible to do kinetic study on the surface species. In this work, we used a transmission IR cell which has an internal reactor volume of 0.48 cm³, providing a short effective response time. The fast time grants us the ability to obtain kinetic parameters of surface species and relate a particular surface intermediate to a given reaction product by comparison with kinetic parameters from SSITKA.

On Pd/Al₂O₃ catalyst, our previous results showed that adsorbed CO was hydrogenated to methane while H-assisted formate decomposition can yield gaseous CO.[4] Those results were obtained from transient switches between hydrogen and reaction gases rather than under steady state reaction conditions. Therefore, it is difficult to understand the roles of adsorbed CO and formate on the surface of Pd/Al₂O₃ under steady state reaction conditions. In the present work, for the first time, operando SSITKA experiments coupled with transmission FTIR, mass spectrometry (MS) and gas chromatography (GC) were performed to probe both the chemical nature and kinetics of reactive intermediates over a 5% Pd-Al₂O₃ catalyst and provide a clear mechanistic picture of the CO₂ hydrogenation reaction by revealing the rate-determining steps for CH₄and CO production under steady state conditions.

2. Experimental

The 5 wt% Pd/Al₂O₃ catalyst was prepared on a commercial γ-Al₂O₃ powder (Condea, BET surface area=200 m²/g) by the incipient wetness method using Pd(NH₃)₄(NO₃)₂ as the precursor. The reactor was loaded with 40 mg catalyst, of which 13 mg were pressed onto a tungsten mesh centered in the optical path of a transmission FTIR spectrometer. Steady-state CO₂ hydrogenation was carried out in the temperature range of 533-573 K. The total gas flow was kept at 10 ml/min, and the gas compositions were switched between H₂/CO₂/Ar and H₂/¹³CO₂ (4 ml/min H₂, 1 ml/min CO₂ or ¹³CO₂and 1 ml/min Ar with dilution of He). The effluent of the cell was analyzed by a MS that was calibrated by a GC.

3. Results and discussion

Figure 1 shows normalized real-time signals for the decay and increase of methane (a) and carbon-monoxide (b) in the effluent at 533 K reaction temperature after the feed gas was switched at 0 s from CO₂/H₂/Ar mixture to ¹³CO₂/H₂ mixture. The fast decay of the Ar gas concentration and the feed gas indicates that the gas phase hold-up of the system is negligible. With increasing temperature, the decay of CH₄ and CO get faster.(not shown) By integration under the decay curves , the mean surface-residence times _CH4 and _CO), the abundance of adsorbed surface intermediates leading to CH₄ and CO products _CH4 and _CO) at 533-573 K were obtained and summarized in Table 1. At low temperature, CO₂ methanation is slower than the reverse water-gas shift reaction, but became faster as the temperature was increased over 563 K. The total amount of intermediates converted to CH₄ and CO is 3.50.2 μmol at 533-573 K. The similar apparent activation energies obtained for the hydrogenation of adsorbed CO and for the formation of CH₄ indicates that the hydrogenation of CO is the rate-determining step during the CO₂methanation reaction. Moreover, the similar apparent activation energies estimated for the consumption of adsorbed formates (FTIR) and for the formation of CO (MS), indicates that the H-assisted decomposition of formates is the rate determining step in the reverse water/gas shift reaction.

Figure 1. Normalized response of (a) CH₄ and ¹³CH₄ products and (b) CO and ¹³CO products as functions of time.

Table 1. Mean surface-residence time _CH4 and _CO, abundance _CH4 and _CO of adsorbed surface intermediates leading to CH₄and CO products.

4. Conclusions

The amount of active sites for both CH₄ and CO products is around 3.50.2 μmol. The CO₂ methanation and reverse water shift reactions share the same formate intermediate. The H-assisted decomposition of adsorbed formate is the rate-determining step for the CO production, whereas the hydrogenation of adsorbed CO is the rate-determining step for the formation of CH₄. Once the adsorbed CO formed on Pd, it will convert to CH₄. At low temperatures ( 563 K), formate decomposition to CO (P-CO) is faster than the hydrogenation of adsorbed CO to CH₄ (P-CH₄). However, P-CH₄ becomes faster than P-CO as the temperature is increased above 563 K. The faster P-CH₄ leads to the conversion of more adsorbed CO (produced in the decomposition of formates) to form CH₄ rather than desorb as CO. In other words, more of the total amount of intermediates (3.50.2 μmol) will form CH₄rather than CO with increasing reaction temperature.

References

1. B.C. McClaine, R.J. Davis, J. Catal., 211 (2002) 379-386.

2. J. Gao, X.H. Mo, J.G. Goodwin, J. Catal., 275 (2010) 211-217.

3. C. Ledesma, J. Yang, D. Chen, A. Holmen, ACS Catal., 4 (2014) 4527-4547.

4. X. Wang, H. Shi, JH Kwak, J. Szanyi, ACS Catal., 5 (2015), 6337â??6349