Steam Methane Reforming Over Ni and Ni/Ag Catalysts − Gaining Mechanistic Insight through DFT and Experiment

Reforming of fossil fuels, in particular steam methane reforming (SMR), is responsible for most of the hydrogen produced worldwide today [[1]].  Nickel is the preferred SMR catalyst because of its cost and availability; however, it is susceptible to deactivation via carbon formation [[2]].  Thus, the design of new SMR catalysts that are inexpensive but resistant to deactivation is of particular interest.  To aid in this search for improved catalysts, we seek an improved understanding of the processes occurring on the Ni catalyst surface through a combination of Density Functional Theory (DFT) and experiment.  We then extend the resulting mechanistic insight to guide the investigation of Ni-bimetallic alloy catalysts, such as Ni/Ag.

Planewave DFT calculations are performed with the software package Dacapo [[3]], using the RPBE functional with spin polarization [[4]].  Planewave and density cutoffs of 340 eV are used, and the Brillouin zone is sampled by a (4,4,1) k-point Monkhorst-Pack grid.  Statistical thermodynamics are applied to the DFT data to obtain thermochemical and kinetic parameters at industrially relevant temperatures and pressures.  The binding energies of the possible surface intermediates that can be formed from the combination of one molecule of CH₄(g) and one of H₂O(g) are calculated.  In addition, nudged elastic band studies and first-order saddle point searches are employed to find the transition state energies.  For experimental investigation, hydrotalcite-derived 12 wt% Ni catalyst is prepared by a co-precipitation method. A method using surface redox reaction between Ni and AgNO₃ has been developed to precisely control the surface alloy, making it possible to compare experimental kinetic data with DFT predictions. The kinetic studies of SMR over Ni and Ni/Ag catalysts are performed in a fixed bed reactor at 1 bar and a temperature range of 500-600 °C.

The thermochemical and kinetic parameters from the DFT studies are combined in a microkinetic model, which is used to perform flux and sensitivity analyses to investigate key pathways on the catalyst surface.  The dominant reforming pathways on the Ni terrace are found to be through the CH* species' combination with either O* or OH* to form CHO* and CHOH*, respectively.  In addition to CH₄(g) adsorption, the formation of the CH_xO_y* complex is found to be a sensitive step in the SMR mechanism.  The CH_xO_y* complex readily dissociates to form the adsorbed products CO* and H* [[5]].  This analysis is also extended to multi-faceted kinetic modeling where the role of Ni(211) step sites are included in pathway and sensitivity analyses.

The insights into active reforming pathways and sensitive mechanistic steps gained from studying the Ni catalyst are applied to guide the computational investigation of the Ni catalyst doped with 0.25 monolayer Ag.  Analysis of the Ni/Ag surface at this Ag coverage has predicted a destabilization of key intermediates such as CH₃*, CHO*, and CHOH*.  As a result, we predict increases in the barriers of the key reactions forming these intermediates.  When compared to pure Ni catalyst data in a fixed bed reactor, the Ni/Ag catalyst is found to inhibit carbon formation but is also found to be less active to SMR, with an increase of approximately 30 kJ/mol in the apparent activation energy, similar to the alloying effect predicted through quantum chemistry.

---

[1].      Logan, B.E.  Environmental Science and Technology  38, 9 (2004).

[2].      Sehested, J.  Catal. Today  111, 1 (2006).

[3].      Dacapo, v.2.7.7.  Available as open source software at http://wiki.fysik.dtu.dk/dacapo.

[4].      Hammer, B.; Hansen, L.B.; Norskov, J.K.  Phys. Rev. B  59, 11 (1999).

[5].      Blaylock, D.W.; Ogura, T.; Green, W.H.; Beran, G.J.O.  J. Phys. Chem. C  113, 12 (2009).