Elsevier

Journal of CO2 Utilization

Volume 20, July 2017, Pages 59-65
Journal of CO2 Utilization

Reaction mechanisms of methanol synthesis from CO/CO2 hydrogenation on Cu2O(111): Comparison with Cu(111)

https://doi.org/10.1016/j.jcou.2017.05.005Get rights and content

Highlights

  • Methanol can be synthesized from CO/CO2 hydrogenation which is dependent on Cu valence.

  • CO is the main carbon source when the surface is predominantly covered by Cu+ species.

  • CO2 is the primary carbon source when metallic Cu covers the surface.

Abstract

A systematic theoretical study was performed to investigate methanol synthesis from CO/CO2 hydrogenation and the water-gas-shift (WGS) reaction on Cu(111) and Cu2O(111) surfaces using density functional theory (DFT) and kinetic Monte Carlo (KMC) simulations. Specifically, DFT was used to investigate methanol synthesis from CO/CO2 hydrogenation on these surfaces at P = 80 atm, T = 553 K and (CO+ CO2)/H2 = 20/80. The results show that methanol can be synthesized from CO or CO2 hydrogenation and is dependent on the catalyst’s preparation as well as the active site type. Further, CO is the main carbon source when the surface is predominantly covered by Cu+ species. However, CO2 is the primary carbon source when metallic Cu covers the surface. Under the reaction conditions investigated, H2 and CO easily reduce Cu2O to metallic Cu, and the Cu+ species are stabilized by the presence of H2O, CO2, carrier (such as MgO) or alkali metals. For this reason, the scale of methanol produced from CO or CO2 hydrogenation depends on the ratio of Cu+/Cu0.

Introduction

Methanol synthesis from CO or CO2 hydrogenation on Cu/ZnO-based catalysts has been industrialized. It is generally accepted that the role of ZnO is to increase Cu dispersion on the catalyst’s surface, while eliciting a synergistic effect with Cu species [1], [2], [3], [4], [5], and the role of the Cu species is to provide active sites for catalysis [3], [6], [7]. Cu2O is easily reduced to metallic Cu with H2 or CO [8], [9]. The presence of H2O or CO2 can maintain copper in its Cu+ state [10], [11], and ZnO will stabilize Cu+ species [12]. Although these details have been reasonably well established, the active center of the copper species has been the subject of debate for the past 20 years[13], [14], [15]. In general, Cu/ZnO-based catalysts are the best catalysts for the water-gas-shift (WGS) reaction (i.e., CO + H2O  CO2+ H2) or its reverse reaction (RWGS) [16]. The interconversion of CO and CO2 through the WGS or RWGS reactions has led to debate of the carbon source in methanol produced from CO/CO2 hydrogenation [16], [17], [18], [19], [20].

Given the importance of methanol, methanol synthesis from CO/CO2 hydrogenation on metallic Cu or Cu+ has been studied extensively using density functional theory (DFT) [16], [19], [21], [22], [23], [24], [25], [26], [27], [28], [29]. On metallic Cu, it is generally accepted that the reaction pathway via CO2 synthesis proceeds through HCOO, H2COO(HCOOH), H2COOH, H2CO, H3CO and H3COH [16], [23], [24], [27], [28]. From CO hydrogenation, the reaction pathway occurs as follows: CO  HCO  H2CO  H3CO  H3COH [16], [19], [22], [25]. Some have proposed that a redox mechanism is operative in the WGS reaction [30], [31]. Further, it has been proposed that methanol is difficult to synthesize from CO hydrogenation since the energy barrier for HCO formation from CO hydrogenation is larger than the desorption energy of CO. Consequently, CO preferentially desorbs rather than hydrogenates to form HCO [32]. Moreover, HCO dehydrogenation is almost barrierless (0.03 eV), which is in stark contrast to the energy barrier for HCO hydrogenation (0.45 eV) [28]. Recently, Grabow and Mavrikakis systematically studied methanol synthesis using kinetic Monte Carlo (KMC) simulations in which all of the parameters were initially derived from DFT calculations on the Cu(111) surface [16]. The authors found that CO2 hydrogenation is responsible for ∼2/3 of the methanol produced, whereas only ∼1/3 of the methanol originates from CO (T = 499.3 K, P = 29.9 atm, yCO = 0.053, yCO2 = 0.047, yH2 = 0.90). Notably, the results are consistent with many experimental observations on different Cu-based catalysts and provide a framework for methanol synthesis on Cu-based catalysts. However, as noted by the authors, the results indicate a discrepancy for CO-rich feeds or catalysts with a basic type of support.

Relatively few studies have focused on methanol synthesis on the Cu2O surface. Indeed, most has focused instead on the adsorption behavior of species such as H2O, CO and H2 [33], [34], [35], [36], [37], [38], [39]. Recently, Uzunova et al. [21] have studied the CO2 reduction mechanism to methanol on Cu32O16 and Cu14O7 and found that the pathway involves COOH, HCOOH, H2CO, H2COH and H3COH as intermediates. However, this group did not investigate methanol synthesis from CO hydrogenation and the WGS reaction. In this regard, we have previously studied H3COH synthesis from CO hydrogenation on a Cu3 cluster on a ZnO surface [40] and observed electron-charge transfer between the metallic Cu and the ZnO carrier; further, the Cu valence was found to be greater than zero and less than one. As a result, the energy barrier for HCO dehydrogenation is similar to that for HCO hydrogenation. Overall, these results give rise to the following questions: How does methanol synthesis from CO/CO2 hydrogenation differ on metal Cu versus Cu+? What is the actual carbon source for methanol synthesis?

To better understand these questions and to provide a framework for the rational design of Cu-based catalysts, H3COH synthesis from CO/CO2 hydrogenation and the WGS reaction has been studied is the present work using DFT. To the best of our knowledge, the exact type of Cu+ species of the Cu-based catalysts for methanol synthesis remains uncertain. In this paper, Cu2O is used to indicate the Cu+ species, which is meant to reflect the difference of methanol synthesis on Cu+ and metallic Cu. We study these reactions on Cu (111) and Cu2O (111) surfaces since X-ray diffraction studies have shown that these are the most common surfaces of Cu and Cu2O [41], [42].

Section snippets

DFT calculations

All periodic DFT calculations were performed using the PBE density functional [43] in the Vienna ab initio simulation package (VASP) [44], [45]. A plane-wave energy cutoff of 400 eV was used for all geometry optimizations, and k-points were sampled using a 3 × 3 × 1 Monkhorst-Pack mesh [46]. A convergence of 0.01 eV of the total electronic energy was used for the optimizations. The slab was separated from its periodic images normal to the surface by a 15 Å vacuum gap, which was sufficient to eliminate

Methanol synthesis from syngas

Table 1, Table 2 show the binding energies of the possible intermediates on the Cu2O(111) and Cu(111) surfaces at their favorable sites during methanol synthesis from syngas. Their corresponding adsorption configurations are shown in Figs. S3 and S4 .

On the Cu2O(111) surface, CO* prefers to adsorb on the CuCUS site through the C atom seeing Fig. S3. The corresponding binding energy is −1.46 eV. Additionally, HCO*, COH*, HCOH* and H2COH* preferentially occupy the CuCUS site through their C atoms,

Conclusion

DFT and KMC simulations were used in the present study to compare methanol synthesis and the WGS on Cu(111) and Cu2O(111). DFT was used to examine methanol synthesis via CO/CO2 hydrogenation on Cu(111) and Cu2O(111) surfaces at P = 80 atm, T = 553 K and (CO + CO2)/H2 = 20/80. On the Cu(111) surface, the KMC results indicate that CO hydrogenation yields methanol through H2CO* and H3CO*, and CO2 is hydrogenated via HCOO*, H2COOH* and H2CO*. Additionally, the RWGS reaction pathway proceeds via COOH*.

Acknowledgments

The authors gratefully acknowledge the financial support of this study by the key project of the National Natural Science Foundation of China (21336006), the National Natural Science Foundation of China (21306125), the key project of Basic Industrial Research of Shanxi (201603D121014), and the Program for the Outstanding Innovative Teams of Higher Learning Institutions of Shanxi

References (72)

  • O.B. Ayodele et al.

    J. CO2 Util.

    (2017)
  • P. Gao et al.

    J. CO2 Util.

    (2016)
  • R.G. Herman et al.

    J. Catal.

    (1979)
  • K. Klier et al.

    J. Catal.

    (1982)
  • K. Klier

    Advances in Catalysis

  • Y.W. Suh et al.

    Catal. Today

    (2000)
  • T. Fujitani et al.

    Applied Catalysis A

    (2000)
  • G. Liu et al.

    J. Catal.

    (1985)
  • Y. Yang et al.

    J. Catal.

    (2013)
  • E.L. Kunkes et al.

    J. Catal.

    (2015)
  • F. Studt et al.

    J. Catal.

    (2012)
  • P. Liu et al.

    Surf. Sci. Rep.

    (2013)
  • Y.F. Zhao et al.

    J. Catal.

    (2011)
  • J. Greeley et al.

    J. Catal.

    (2002)
  • H. Wu et al.

    Chem. Phys. Lett.

    (2013)
  • R. Zhang et al.

    Appl. Surf. Sci.

    (2011)
  • Z.J. Zuo et al.

    Appl. Surf. Sci.

    (2010)
  • B.Z. Sun et al.

    Appl. Surf. Sci.

    (2008)
  • A.K. Mishra et al.

    J. CO2 Util.

    (2016)
  • Z.J. Zuo et al.

    Appl. Surf. Sci.

    (2012)
  • J. Anton et al.

    J. Catal.

    (2016)
  • G. Kresse et al.

    Comp. Mater. Sci.

    (1996)
  • S. Kattel et al.

    J. Catal.

    (2016)
  • Q.L. Tang et al.

    J. Catal.

    (2009)
  • Z.J. Zuo et al.

    Appl. Catal. A

    (2010)
  • Z.J. Zuo et al.

    Inter. J. Hydrogen Energy

    (2014)
  • Z.J. Zuo et al.

    Appl. Surf. Sci.

    (2014)
  • G.C. Chinchen et al.

    Appl. Catal.

    (1987)
  • R.A. Koeppel et al.

    Appl. Catal. A

    (1992)
  • J.A. Brown Bourzutschky et al.

    J. Catal.

    (1990)
  • D.B. Clarke et al.

    J. Catal.

    (1995)
  • E.M. Calverley et al.

    J. Catal.

    (1991)
  • P.E.H.N.J.B. Hansen
  • J. Nakamura et al.

    Top. Catal.

    (2003)
  • K.C. Waugh

    Catal. Lett.

    (2012)
  • O. Martin et al.

    ACS Catal.

    (2015)
  • Cited by (0)

    1

    The authors contributed equally.

    View full text