Elsevier

Journal of Catalysis

Volume 404, December 2021, Pages 537-550
Journal of Catalysis

Solvent promotion on the metal-support interaction and activity of Pd@ZrO2 Catalyst: Formation of metal hydrides as the new catalytic active phase at the Solid-Liquid interface

https://doi.org/10.1016/j.jcat.2021.10.030Get rights and content

Highlights

  • The Pd-Hx (Hδ-) and ZrO2-(OH)x can be formed by interfacial water dissociation.

  • The Hδ+ undergo the proton coupled electron transfer (PCET) from Pd to form Hδ-.

  • The spatial separation of H and OH in water results in the interface polarization.

  • The metal-support polarization promotes their interaction and flattens Pd cluster.

  • The formed Pd-hydrides could behave as the new active phase in CO oxidation.

Abstract

The supported metal catalysts are usually prepared and widely applied in aqueous phase. Here, by performing density functional theory calculations and ab initio molecular dynamics simulations with explicit solvent waters, the metal-support interaction of Pd@ZrO2 catalyst was found to be enhanced by water dissociation at the metal-support interface. This happens to an appreciable number of interfacial waters, generating ∼ 25% hydrides compared to Pd atoms. After the partially heterolytic water dissociation, the generated Hδ+ would undergo the proton coupled electron transfer (PCET) from Pd to form the hydride (Hδ-). The spatial separation of the generated H on Pd and remaining OH on ZrO2 results in the polarization of metal-support interface, which then flattens the shape of cluster. Moreover, the formed metal-hydride behaves as the new catalytic active phase for CO oxidation, making the catalyst highly active in water vapor.

Introduction

For aqueous reactions catalyzed by the widely-used supported metal catalysts, the solvent waters play a critical role on the rate, yield and selectivity of reactions.[1] That dramatic influence sometimes comes from the following factors: different solubility and mass transport limitations in aqueous phase[2], the unexpected contributions of the soluble species[3], [4] and the structural transformations of the catalyst itself, such as collapse[5], leaching[6], agglomeration[7], [8], fouling[9] and coking[10], [11]. Meanwhile, the more essential factor should relate to the influence of waters on the active sites and those reactive intermediates. For example, the solvent waters could compete for the adsorption with reactants[12], [13], and the solvation could change the stability of the key intermediates and transition states[14], [15], [16], [17]. What is more, except a few techniques (such as melt infiltration and atomic layer deposition), the commonly used preparing methods of supported metal catalysts (such as precipitation, impregnation, and the colloidal synthesis) most undergo in aqueous solutions[18]. Even the “dry impregnation” (or called “incipient wetness impregnation”) needs limited aqueous solution to fill the pore volume of support[19]. That means even for the supported metal catalysts used in gas or vapor phase, solvent waters might still influence their properties at the stage of preparation in aqueous phase.

However, the mechanistic details of how solvent waters influence the metal-support interface, and in turn their catalytic performances, are still unclear. Such limited understanding could partially be attributed to the hard experimental characterization in aqueous phase, such as the background noise in some traditional spectroscopies.[20] Besides the fast-developing in situ and time-resolved characterization tools, the theoretical calculations based on density functional theory (DFT) could be helpful to provide mechanistic insights into the solvent effect, especially when explicit waters are used to model the aqueous phase.[21] Although its computational cost is high, with the fast development of high-performance computing systems, there are more and more DFT calculations using the explicit solvent model during the past two decades.

Such realistic but time-consuming calculations were first applied to simulate the metal-water interfaces[22], [23], [24] and the surface hydration of metallic oxide in water[25], [26], [27]. Then, the explicit solvation model was further applied to catalytic reactions. In electrocatalytic reactions, such as oxygen reduction reaction (ORR)[28], [29] and carbon dioxide reduction reactions (CO2RRs)[30], [31], the solvent waters on metal electrode could not only change the stability of intermediates and transition states, but also, in certain cases, react with intermediates to create new pathways. During the acetic acid ketonization catalyzed by monoclinic ZrO2, Cai et al.[32] showed that unlike the Langmuir-Hinshelwood mechanism in vapor phase, the reaction with explicit solvent waters followed the Eley-Rideal mechanism by forming the β-keto acid intermediate. Wang and co-worker[33] studied the photocatalytic oxygen evolution reaction (OER) on rutile TiO2 with explicit solvent waters, where the experimental unsatisfactory efficiency was suggested to be limited by the TiO2 photoholes rather than the inherent reaction barriers. Recently, Zhao et al.[34] conducted the DFT calculation with explicit solvent waters on the aldol condensation of formaldehyde and acetone catalyzed by anatase TiO2, and the dehydration was suggested to be promoted by solvent waters.

The metal nano-catalysts are usually supported on oxide supports in experiments. However, due to the higher computational cost of involving both the metal and support into the model, in most previous theoretical calculations, especially when explicit solvent waters were involved, pure metal surfaces were applied as a compromise to represent the supported metal catalysts. For instances, by DFT calculations, Zope et al.[35] showed in aqueous phase, both solution-mediated and metal-catalyzed elementary steps were involved in alcohol oxidation at the Au-water interface, although in their experiment the Au particles were supported on TiO2 or C surfaces. According to the ab initio molecular dynamics (AIMD) simulations by Yoon and co-workers[36], the explicit solvent waters on Pt(1 1 1) and Ni(1 1 1) surfaces could strongly influence the energetic preference of keto/enol tautomerization during phenol hydrogenation, resulting in different product selectivity, although the supported Pt and Ni catalysts were used in experiments[37], [38], [39]. Such ignorance of support was suggested to have no significant impact when these metal particles were large in scale of several nanometers. However, there were also lots of reports that the supports, especially those metallic oxide supports, could dramatically change the catalytic activity and selectivity by distinct metal-support interactions[40], [41]. For these reactions, obviously the theoretical study on pure metal surfaces could deviate from the actual situation in experiment, where the support should also be involved in modelling.

The above-mentioned theoretical works largely explored the solid–liquid interface for catalysts in aqueous phase, but all of them focused on either the metal-water interface or the oxide–water interface. There are rare DFT calculations concerning how waters influence the metal-oxide interface. According to the AIMD simulation on Au11@TiO2 by Camellone and Marx[42], the surrounding explicit waters would influence the charges of metal atoms by solvation. Such charge variations on metal atoms were also observed in AIMD simulation of Pt6@CeO2 in explicit waters by Camellone and Fabris[43], where the solvent water dissociation took place. In the recent theoretical study on Au@TiO2 by Chandler et al.[44], with some explicit waters added on both sides of Au nanorods, it was shown the H2 activation could be suppressed by waters. These DFT calculations indeed provide important insights of how solvent waters influence the charges of metal clusters and their catalytic performances. However, in their models[42], [43], [44], the metal clusters were directly supported on the anhydrous support. As stated above, the supported metal catalysts are in most cases prepared in aqueous solution. For many oxide supports, such as the commonly-used Al2O3, TiO2, ZrO2, CeO2, some of their anhydrous surfaces could soon be hydrated by water dissociation,[25], [32] although on specific surfaces such hydration was shown harder.[45] That surface hydration can be determined by many factors, including the chemical composition of oxide, the index of exposed oxide crystal surface and the electric double layer formed at the solid-aqueous interface.[46], [47], [48] The degree of surface hydration can be crucial to the characteristics of supported metal cluster. For example, the feasible shapes of Pt and Pd cluster on the anhydrous and hydroxylated γ-Al2O3 surfaces were dramatically different due to the distinct metal-support interactions.[49] Of course the oxide surfaces could undergo the dehydration during the subsequent drying or calcination in experiment. Nevertheless, it is still of importance to investigate the evolution of supported metal catalysts on the hydrated oxide surface in water, which can provide insights into the properties of metal-support interface in aqueous phase.

In experiment, the supported Pd nanocluster catalysts are widely applied on both oxidation reaction and reduction reaction, such as CO oxidation[15], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], methanol oxidation[60], [61], [62], phenol hydrogenation[37], [63], [64], [65], [66], [67], [68], and furfural hydrogenation[69], where in many cases the ZrO2 is chosen as the support[51], [52], [53], [56], [61], [62], [67], [68]. Most of these hydrogenation reactions take place in aqueous phase[37], [63], [64], [65], [66], [69]. For those CO and methanol oxidation, the present of water has also been proven benefit for the supported Pd catalysts[15], [60], [70]. In addition, the recent study by Cargnello et. al.[71] shows the steam pretreatment of Pd/Al2O3 catalysts could promote methane oxidation by the grain boundary density through crystal twinning and the associated defective strain, suggesting the waters could not only work during catalytic reactions, but also activate the supported metal catalysts themselves. However, there were seldom theoretical simulations on how solvent waters influence the metal-support interface and in turn its catalytic activity. In this work, the dynamic evolution of Pd@ZrO2 in aqueous phase was systematically investigated by DFT calculations and AIMD simulations. With the Pd cluster, ZrO2 support and abundant explicit solvent waters together involved in the model, it was found the metal-support interaction could be enhanced in aqueous phase by spontaneous water dissociation at the Pd-ZrO2 interface. During dissociation, the generated Hδ+ would undergo the proton coupled electron transfer (PCET) from Pd to form the hydride (Hδ-). It is the spatial separation of the generated H on Pd and the remaining OH on ZrO2 polarizes the metal-support interface, which then flattens the cluster shape. What is more, the generated hydride could promote the activity of CO oxidation as the new catalytically active phase. This work reveals the possibility that in aqueous solution, the metal-support interaction of the supported metal catalyst could be promoted by interfacial water dissociation, resulting in the metal-hydride as the active phase, which might be crucial but lacks of investigation in previous studies.

Section snippets

Computational methods

The stable configurations and transition states are optimized by spin polarized DFT calculation within the framework of periodic boundary condition by using the VASP code[72]. Projected augmented wave (PAW) method[29] is used to describe the ion–core electron interaction, where the exchange and correlation interactions are described by PBE functional[73]. Grimme’s D3 corrections[74] are involved to describe the dispersion interactions. The (1¯11) surface is the stable surface of monoclinic ZrO2

Simulation of Pd13@ZrO2 catalyst in explicit solvent waters.

In previous theoretical studies of supported metal catalysts, the support surfaces are often modelled by either anhydrous surfaces[42], [43], [49] or surface filled artificially with hydroxyl groups[49]. To get the representative hydrated ZrO2 surface in aqueous phase, as shown in Fig. S1, the explicit waters with the density around 1 g/mL were first put above the anhydrous ZrO2 surface, followed by the 12 ps AIMD simulation at the typical supporting temperature of 353 K[80]. Similar to the

Conclusion

With the Pd@ZrO2 catalyst modelled in abundant explicit solvent waters, by DFT calculations and AIMD simulations, it was found the metal-support interaction could be enhanced in aqueous phase by spontaneous water dissociation at the metal-support interface. Such interfacial water dissociation is new and more favorable than the reported dissociations of adsorption waters on metal surfaces and solvent waters adjacent to metal cluster. What is more, this type of dissociation was not limited to one

Funding Information

This work was financially supported by NSFC (No. 22022504; No. 22003022) of China, Natural Science Foundation of Guangdong, China (No. 2021A1515010213), Guangdong “Pearl River” Talent Plan (No. 2019QN01L353), Higher Education Innovation Strong School Project of Guangdong Province of China (2020KTSCX122) and Guangdong Provincial Key Laboratory of Catalysis (No. 2020B121201002).

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was financially supported by NSFC (No. 22022504; No. 22003022) of China, Natural Science Foundation of Guangdong, China (No. 2021A1515010213), Guangdong “Pearl River” Talent Plan (No. 2019QN01L353), Higher Education Innovation Strong School Project of Guangdong Province of China (2020KTSCX122) and Guangdong Provincial Key Laboratory of Catalysis (No. 2020B121201002). The computational resource is supported from the Center for Computational Science and Engineering at SUSTech. The

References (97)

  • L. Liu et al.

    Low-Temperature CO Oxidation over Supported Pt, Pd Catalysts: Particular Role of FeOx Support for Oxygen Supply during Reactions

    J. Catal.

    (2010)
  • E. Bekyarova et al.

    CO Oxidation on Pd/CeO2-ZrO2 Catalysts

    Catal. Today

    (1998)
  • A.S. Ivanova et al.

    Metal-Support Interactions in Pt/Al2O3 and Pd/Al2O3 Catalysts for CO Oxidation

    Appl. Catal. B Environ.

    (2010)
  • H. Zhu et al.

    Pd/CeO2-TiO2 Catalyst for CO Oxidation at Low Temperature: A TPR Study with H2 and CO as Reducing Agents

    J. Catal.

    (2004)
  • A. Toso et al.

    High Stability and Activity of Solution Combustion Synthesized Pd-Based Catalysts for Methane Combustion in Presence of Water

    Appl. Catal. B Environ.

    (2018)
  • W.S. Epling et al.

    Catalytic Oxidation of Methane over ZrO2-Supported Pd Catalysts

    J. Catal.

    (1999)
  • S. Ding et al.

    Pd Nanoparticles Supported on N-Doped Porous Carbons Derived from ZIF-67: Enhanced Catalytic Performance in Phenol Hydrogenation

    J. Ind. Eng. Chem.

    (2017)
  • C. Newman et al.

    Effects of Support Identity and Metal Dispersion in Supported Ruthenium Hydrodeoxygenation Catalysts

    Appl. Catal. A Gen.

    (2014)
  • Y. Zhao et al.

    Hydrogen Adsorption and Dissociation on Pd 19 Cluster Using Density Functional Calculations

    Comput. Theor. Chem.

    (2012)
  • G.C. Wang et al.

    A Systematic Theoretical Study of Water Dissociation on Clean and Oxygen-Preadsorbed Transition Metals

    J. Catal.

    (2006)
  • Z. Vít et al.

    Effect of Catalyst Precursor and Its Pretreatment on the Amount of β-Pd Hydride Phase and HDS Activity of Pd-Pt/Silica-Alumina

    Appl. Catal. B Environ.

    (2014)
  • A.L. Bugaev et al.

    In Situ Formation of Hydrides and Carbides in Palladium Catalyst: When XANES Is Better than EXAFS and XRD

    Catal. Today

    (2017)
  • O. Pozdnyakova et al.

    Preferential CO Oxidation in Hydrogen (PROX) on Ceria-Supported Catalysts, Part II: Oxidation States and Surface Species on Pd/CeO2 under Reaction Conditions Suggested Reaction Mechanism

    J. Catal.

    (2006)
  • T. Kitanosono et al.

    Catalytic Organic Reactions in Water toward Sustainable Society

    Chem. Rev.

    (2018)
  • R.A. Sheldon et al.

    Heterogeneous Catalysts for Liquid-Phase Oxidations: Philosophers’ Stones or Trojan Horses?

    Acc. Chem. Res.

    (1998)
  • D. Roy et al.

    Cu-Based Catalysts Show Low Temperature Activity for Glycerol Conversion to Lactic Acid

    ACS Catal.

    (2011)
  • B. Qiao et al.

    Highly Efficient Catalysis of Preferential Oxidation of CO in H2-Rich Stream by Gold Single-Atom Catalysts

    ACS Catal.

    (2015)
  • I. Sádaba et al.

    Deactivation of Solid Catalysts in Liquid Media: The Case of Leaching of Active Sites in Biomass Conversion Reactions

    Green Chem.

    (2015)
  • R.M. Ravenelle et al.

    Structural Changes of γ-Al2O3-Supported Catalysts in Hot Liquid Water

    ACS Catal.

    (2011)
  • H.N. Pham et al.

    Carbon Overcoating of Supported Metal Catalysts for Improved Hydrothermal Stability

    ACS Catal.

    (2015)
  • C.H. Kuo et al.

    Heterogeneous Acidic TiO2 Nanoparticles for Efficient Conversion of Biomass Derived Carbohydrates

    Green Chem.

    (2014)
  • R. Xiong et al.

    Adsorption of HMF from Water/DMSO Solutions onto Hydrophobic Zeolites: Experiment and Simulation

    ChemSusChem

    (2014)
  • J.R. Copeland et al.

    Surface Interactions of C2 and C3 Polyols with γ-Al2o3 and the Role of Coadsorbed Water

    Langmuir

    (2013)
  • S.D. Ebbesen et al.

    The Influence of Water and PH on Adsorption and Oxidation of CO on Pd/Al2O3-an Investigation by Attenuated Total Reflection Infrared Spectroscopy

    Phys. Chem. Chem. Phys.

    (2009)
  • S. Behtash et al.

    Solvent Effects on the Hydrodeoxygenation of Propanoic Acid over Pd(111) Model Surfaces

    Green Chem.

    (2014)
  • P. Munnik et al.

    Recent Developments in the Synthesis of Supported Catalysts

    Chem. Rev.

    (2015)
  • Mehrabadi, B. A. T.; Eskandari, S.; Khan, U.; White, R. D.; Regalbuto, J. R. A Review of Preparation Methods for...
  • J.M. Andanson et al.

    Exploring Catalytic Solid/Liquid Interfaces by in Situ Attenuated Total Reflection Infrared Spectroscopy

    Chem. Soc. Rev.

    (2010)
  • M. Saleheen et al.

    Liquid-Phase Modeling in Heterogeneous Catalysis

    ACS Catal.

    (2018)
  • L.S. Pedroza et al.

    Local Order of Liquid Water at Metallic Electrode Surfaces

    J. Chem. Phys.

    (2015)
  • S. Izvekov et al.

    Ab Initio Molecular Dynamics Simulation of the Ag(111)-Water Interface

    J. Chem. Phys.

    (2001)
  • M. Sumita et al.

    Interface Water on TiO2 Anatase (101) and (001) Surfaces: First-Principles Study with TiO2 Slabs Dipped in Bulk Water

    J. Phys. Chem. C

    (2010)
  • L.M. Liu et al.

    Structure and Dynamics of Liquid Water on Rutile TiO2(110)

    Phys. Rev. B - Condens. Matter Mater. Phys.

    (2010)
  • G. Tocci et al.

    Solvent-Induced Proton Hopping at a Water-Oxide Interface

    J. Phys. Chem. Lett.

    (2014)
  • T. Cheng et al.

    Mechanism and Kinetics of the Electrocatalytic Reaction Responsible for the High Cost of Hydrogen Fuel Cells

    Phys. Chem. Chem. Phys.

    (2017)
  • G. Kresse et al.

    From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method

    Phys. Rev. B - Condens. Matter Mater. Phys.

    (1999)
  • T. Sheng et al.

    Electrochemical Reduction of CO2 into CO on Cu(100): A New Insight into the C-O Bond Breaking Mechanism

    Chem. Commun.

    (2017)
  • T. Cheng et al.

    Reaction Intermediates during Operando Electrocatalysis Identified from Full Solvent Quantum Mechanics Molecular Dynamics

    Proc. Natl. Acad. Sci. U. S. A.

    (2019)
  • Cited by (12)

    • Crystal plane induced metal-support interaction in Pd/Pr-CeO<inf>2</inf> catalyst boosts H<inf>2</inf>O-assisted CO oxidation

      2023, Journal of Catalysis
      Citation Excerpt :

      Using XPS, H2-TPR and Raman spectroscopy characterization, it was found that the Pd/Pr-CeO2-NRs catalyst contains more active oxygen species and lower oxygen reduction temperature which indicate that oxygen vacancies play an important role in the CO oxidation process. Nowadays, the mechanism of CO oxidation on metal oxides catalysts remains controversial, especially in the presence of H2O [68,69]. Previous literature has reported that the mechanism of CO oxidation reaction over CeO2 catalyst follow the MvK model.

    • Water accelerated activity of Ru NPs in sequential hydrogenation of nitrobenzene to cyclohexylamine

      2022, Journal of Catalysis
      Citation Excerpt :

      The participation of water can follow a conventional L-H route or a proton coupled electron transfer (PCET) pathway [34,35]. Though it is unclear which route is dominant at present stage, it is reasonable to propose that the PCET may be involved in the hydrogenation process because the large amounts of hydrated hydronium ions and adsorbed H on the Ru surfaces are beneficial to facilitate the PCET pathway [35–37]. In detail, the activity of all the catalysts is promoted to 4–5 folds for NB hydrogenation to AN in IPA/H2O (95:5 v/v%) compared with that in IPA (Fig. 6c).

    View all citing articles on Scopus
    View full text