Elsevier

Journal of Catalysis

Volume 309, January 2014, Pages 397-407
Journal of Catalysis

Thermochemistry and micro-kinetic analysis of methanol synthesis on ZnO (0 0 0 1)

https://doi.org/10.1016/j.jcat.2013.10.015Get rights and content

Abstract

In this work, we examine the thermochemistry of methanol synthesis intermediates using density functional theory (DFT) and analyze the methanol synthesis reaction network using a steady-state micro-kinetic model. The energetics for methanol synthesis over Zn-terminated ZnO (0 0 0 1) are obtained from DFT calculations using the RPBE and BEEF-vdW functionals. The energies obtained from the two functionals are compared and it is determined that the BEEF-vdW functional is more appropriate for the reaction. The BEEF-vdW energetics are used to construct surface phase diagrams as a function of CO, H2O, and H2 chemical potentials. The computed binding energies along with activation barriers from literature are used as inputs for a mean-field micro-kinetic model for methanol synthesis including the CO and CO2 hydrogenation routes and the water–gas shift reaction. The kinetic model is used to investigate the methanol synthesis rate as a function of temperature and pressure. The results show qualitative agreement with experiment and yield information on the optimal working conditions of ZnO catalysts.

Graphical abstract

Highlights

  • Most stable surface phase depends on carbon chemical potential.

  • Formaldehyde or methoxy hydrogenation is rate-limiting step.

  • Carbon dioxide poisoning of ZnO catalysts is due to formate coverage.

  • Micro-kinetic model predicts high temperature and pressure activity.

Introduction

The catalytic synthesis of methanol from synthesis gas is one of the most important industrial chemical processes with a worldwide production capacity of around 40 million tons per year [1]. Currently, methanol is used primarily as a starting material or solvent for chemical synthesis, although increasingly, it is employed as a fuel [1], [2], [3]. The diversity of compounds that can be synthesized from methanol along with its promise as an energy carrier makes the development of more efficient and sustainable production schemes a priority for the chemical industry of the future [3].

Methanol was originally produced from synthesis gas using a zinc oxide (ZnO) catalyst discovered in the 1920s. The catalyst required high temperatures (600–700 K) and pressures (200–300 bar) to operate industrially [1], [2], [4]. The high pressure methanol process was rendered obsolete in the 1960s with the development of a lower pressure process based on a Cu–ZnO catalyst [2], [4], [5]. The active phase of this Cu–ZnO system has been the subject of much debate [1], [5], [6], [7], [8], [9], [10], though several investigations have shown strong evidence that the active site is on Cu and that the reaction is structure sensitive [10], [11], [12], [13].

A detailed understanding of the methanol synthesis reaction is hindered by several fundamental obstacles. The industrial process is carried out at very high pressures which are required to drive the reaction thermodynamics. This presents a challenge in extrapolation of the results of ultra-high vacuum (UHV) studies to relevant conditions. In addition, the elementary steps in methanol synthesis form a complex network due to the possibility of hydrogenating carbon monoxide (CO) or carbon dioxide (CO2), along with the water–gas shift reaction, which couples the two pathways together. This makes it challenging to elucidate the reaction mechanism, although convincing evidence has been shown for CO hydrogenation on ZnO [14] and for CO2 hydrogenation on Cu–ZnO [1], [5], [15]. Furthermore, the industrial Cu–ZnO catalyst exhibits a highly heterogeneous morphology and the activity is dependent on preparation procedures [5], [10]. For this reason, it can be difficult to glean understanding from the simplified model systems commonly used in experiment and theory.

In this work, we seek to make progress toward understanding methanol synthesis by investigating the reaction on the ZnO catalyst. The thermochemistry of oxygen and CO adsorption on several ZnO faces is examined, confirming the polar surfaces as the most likely active facets. The interaction of methanol synthesis intermediates with the Zn-terminated ZnO (0 0 0 1) surface is examined in detail using density functional theory (DFT). These results are used to construct a surface phase diagram as a function of CO, H2O and H2 chemical potentials. The phase diagram indicates that surface terminations including formate and methoxy are highly stable even at low CO pressures, revealing the importance of considering a carbon reservoir. The energetics are also used in a mean-field micro-kinetic model which includes CO and CO2 hydrogenation pathways as well as the water–gas shift reaction. This model is employed in order to examine the temperature and pressure dependence of the methanol synthesis rate on ZnO. The results indicate that this model is capable of describing the experimental trends in the operating conditions of the ZnO catalyst and suggest site poisoning by formate (HCOO) as the reason that high temperatures, high total pressures, and low CO2 concentrations are required.

Section snippets

Density functional theory

All electronic structure calculations are carried out using the grid-based projector augmented wave code GPAW [16]. Exchange–correlation energies are treated using both the RPBE [17] and BEEF-vdW [18] functionals. All calculations are preformed on a finite-difference grid, and the Brillouin zone is sampled by a Monkhorst–Pack k-point mesh [19]. Several surface facets and system sizes are investigated, and the values of the grid spacing and k-point density required for convergence varied; all

ZnO bulk and surface facets

Bulk ZnO is most stable in a hexagonal wurtzite structure. The lattice constants are determined via minimization of the energy using a downhill simplex algorithm [36]. The values are found to be a = 3.33 Å, c = 5.32 Å (a = 3.35 Å, c = 5.32 Å) for the BEEF-vdW (RPBE) functionals. These are within 3% of the experimental values (a = 3.25 Å, c = 5.21 Å) [37]. The enthalpy of formation (converted to the standard NIST references via experimental values) of ZnO is computed to be −3.42 eV (−sss3.06 eV) with BEEF-vdW

Conclusions

The binding energies of CO and oxygen are used to probe the reactivity of various ZnO surface facets. It is shown that the non-polar (101¯0) and (112¯0) surfaces interact very weakly with adsorbed oxygen, while the polar zinc-terminated (0 0 0 1) surface is considerably more reactive. Interestingly the CO binding energy is comparable for both surfaces despite the large difference in oxygen reactivity. Since methoxy and most intermediates in the CO2 hydrogenation pathway bind through oxygen, the

Acknowledgements

A.J.M. was supported by the National Science Foundation in cooperation with the Danish National Research Foundation through the Nordic Research Opportunity and Graduate Research Fellowship Program (Grant No. DGE-1147470). Computational resources were provided by Haldor Topsøe A/S (A.J.M., J.S., P.G.M.), the Danish Center for Scientific Computing (A.J.M., J.R., I.C.) and the U.S. Department of Energy, Office of Basic Energy Sciences (A.J.M., F.S., J.K.N.).

References (63)

  • W. Cheng

    Structure sensitivity in methanol decomposition on ZnO single-crystal surfaces

    Journal of Catalysis

    (1983)
  • J. Strunk et al.

    The surface chemistry of ZnO nanoparticles applied as heterogeneous catalysts in methanol synthesis

    Surface Science

    (2009)
  • P. Mars et al.

    Oxidations carried out by means of vanadium oxide catalysts

    Chemical Engineering Science

    (1954)
  • K.C. Waugh

    Methanol synthesis

    Catalysis Letters

    (2012)
  • E. Fiedler et al.

    Methanol

  • X.-M. Liu et al.

    Recent advances in catalysts for methanol synthesis via hydrogenation of CO and CO2

    Industrial & Engineering Chemistry Research

    (2003)
  • G. Natta

    Synthesis of methanol

  • E. Solomon et al.

    Electronic structures of active sites on metal oxide surfaces: definition of the copper–zinc oxide methanol synthesis catalyst by photoelectron spectroscopy

    Chemical Reviews

    (1993)
  • T. Fujitani et al.

    Creation of the active site for methanol synthesis on a Cu/SiO2 catalyst

    Catalysis Letters

    (1997)
  • J. Nakamura et al.

    On the issue of the active site and the role of ZnO in Cu/ZnO methanol synthesis catalysts

    Topics in Catalysis

    (2003)
  • M. Behrens et al.

    The active site of methanol synthesis over Cu/ZnO/Al203 industrial catalysts

    Science

    (2012)
  • H. Nakano et al.

    Structure-dependent kinetics for synthesis and decomposition of formate species over Cu(1 1 1) and Cu(1 1 0) model catalysts

    The Journal of Physical Chemistry B

    (2001)
  • M. Kurtz et al.

    Active sites on oxide surfaces: ZnO-catalyzed synthesis of methanol from CO and H2

    Angewandte Chemie

    (2005)
  • L.C. Grabow et al.

    Mechanism of methanol synthesis on Cu through CO2 and CO hydrogenation

    ACS Catalysis

    (2011)
  • J. Enkovaara et al.

    Electronic structure calculations with GPAW: a real-space implementation of the projector augmented-wave method

    Journal of Physics: Condensed Matter

    (2010)
  • B. Hammer et al.

    Improved adsorption energetics within density-functional theory using revised Perdew–Burke–Ernzerhof functionals

    Physical Review B

    (1999)
  • J. Wellendorff et al.

    Density functionals for surface science: exchange-correlation model development with Bayesian error estimation

    Physical Review B

    (2012)
  • H.J. Monkhorst et al.

    Special points for Brillouin-zone integrations

    Physical Review B

    (1976)
  • L. Bengtsson

    Dipole correction for surface supercell calculations

    Physical Review B

    (1999)
  • B. Meyer et al.

    Density-functional study of the structure and stability of ZnO surfaces

    Physical Review B

    (2003)
  • J. Kiss et al.

    Methanol synthesis on ZnO(0 0 0 1). III. Free energy landscapes, reaction pathways, and mechanistic insights

    The Journal of Chemical Physics

    (2011)
  • Cited by (0)

    View full text