Interfacial phenomena in (de)hydrogenation reactions

Jinlong Gong*a, Zhihong Nieb and Xinbin Maa
aKey Laboratory for Green Chemical Technology of Ministry of Education, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China. E-mail: jlgong@tju.edu.cn; xbma@tju.edu.cn; Fax: +86-22-87401818
bDepartment of Chemistry and Biochemistry, University of Maryland, College Park, MD 20742, USA. E-mail: znie@umd.edu

Received 12th June 2013, Accepted 12th June 2013
Interfacial phenomena in heterogeneous reactions are ubiquitous. The boundary between the phases has different properties from that of the bulk phase and is vital in various surface processes. Indeed, the interface could play a crucial role in binding, transformation and transport of surface species (e.g., electrons and intermediates) between the phases. However, it is not trivial to observe and understand interfacial phenomena experimentally considering that the two-dimensionality of the interface limits the size of the sample. Therefore, advanced spectroscopies and other surface/interface probes have been developed to investigate interfacial phenomena at the molecular/atomic level. Coupling with experimental observations, theoretical methods based on first principles and/or Monte Carlo approaches are also used to understand the structural and electronic properties of surfaces and interfaces.

Hydrogenation of unsaturated bonds such as C–C, C–O, C–N and dehydrogenation reactions on heterogeneous surfaces are important for a wide range of industrial processes. These reactions typically undergo multiple pathways leading to undesirable byproducts due to the complexity of the surface and interfacial structures, and intrinsic uncontrollable kinetics. Therefore, numerous efforts have been made to understand better the interfacial phenomena of adsorbents and intermediates on well-established solid surfaces with the help of advanced characterization tools and theoretical modeling. This themed issue is targeted to provide the chemical community with recent advances in fundamental understanding of surface and interfacial phenomena occurring during the (de)hydrogenation reactions encompassing both classical and model systems.

Carbon–carbon bonds are among the most easily hydrogenated functional groups, where ambient conditions and low H2 pressures may be sufficient to perform these reactions with the appropriate catalyst. Zaera (DOI: 10.1039/C3CP50402F) has provided a nice Perspective on the state of the art regarding the mechanism of olefin hydrogenation over transition-metal catalysts. He discusses several key fundamental issues involved in the reaction schemes regarding: (i) the specific role of catalyst surfaces covered with strongly adsorbed carbonaceous layers (e.g., effect on the adsorption kinetics of hydrogen); (ii) the configuration of surface pi-bonded intermediates and evolution of surface alkyl intermediates; (iii) the quantitative analysis of competition between the beta-hydride and reductive elimination steps; (iv) the challenges facing the bridging of the pressure and materials gap between ultrahigh vacuum (UHV) model systems and more realistic catalytic conditions. To understand better the role of promoting metal for hydrogenation reactions over bimetallic and/or alloy systems, Corma and co-workers (DOI: 10.1039/C3CP50519G) have synthesized controllably Pt nanoparticles in the channel of an Sn-beta catalyst considering that the presence of Lewis acid sites in the vicinity of the metal nanoparticle can favor the activation of the CO bond. The formed Pt0–Sn4+ sites in the channel facilitate the activation of the carbonyl group enhancing the hydrogenation rate of the C[double bond, length as m-dash]O bond. Flytzani-Stephanopoulos et al. (DOI: 10.1039/C3CP51538A) have successfully employed the model-catalyst concept to fabricate practical Pd–Cu catalysts. The Pd–Cu nanoparticles were designed based on model single atom alloy surfaces, in which individual, isolated Pd atoms act as sites for hydrogen uptake, dissociation, and spillover onto the surrounding Cu surface. This unique Pd–Cu structure leads to an order of magnitude higher activity for phenylacetylene hydrogenation compared to the monometallic Cu counterpart. Chen and his colleagues (DOI: 10.1039/C3CP44688C) have indicated that bimetallic surfaces often show unique activity for reactions involving the consumption and production of hydrogen, such as hydrogenation and reforming reactions, respectively. Yet, these two types of reactions require different bimetallic configurations. For the Pt–Ni bimetallic system, the desirable structure is Pt-terminated for hydrogenation while Ni-terminated for reforming. Based on a number of characterization approaches they show that the nature of the support (e.g., oxygen mobility, metal–support interaction, and acidic–basic properties) determines the reaction path of these bimetallic catalysts. Bhan et al. (DOI: 10.1039/C3CP50855B) have co-fed methane and oxygenates using Mo/H-ZSM-5 catalysts to determine the kinetic and thermodynamic consequences of the reactions.

The combination of first-principles density functional theory (DFT) calculations of adsorption and reaction energetics with experimental observations could provide significant insights into what properties of the catalyst enable the attainment of high activity and selectivity. Hwang and his group (DOI: 10.1039/C3CP50618E) present mechanistic origins regarding the improvement in the catalytic performance of Pd-based alloys for hydrogenation reactions (e.g., oxygen hydrogenation). For the Pd/Pd3Co model system where one or two Pd overlayers are located on top of the bimetallic substrate, their calculations clearly demonstrate that the subsurface Co atoms assist in facilitating the oxygen reduction reaction by lowering the activation barriers for O/OH hydrogenation with a slight increase in the O2 scission barrier. Furthermore, the analysis of intra- and inter-layer orbital interactions in the near-surface region elucidates the synergetic interplay between the surface electronic structure modification due to the underlying Co atoms (interlayer ligand effect) and the compressive strain caused by the Pd3Co substrate. Sabbe et al. (DOI: 10.1039/C3CP50617G) have carried out a comprehensive DFT analysis regarding the surface/subsurface aggregation of a series of Pt3M/Pt(111) surfaces and Pt3M(111) bulk alloys (M = Fe, Co, Ni, Cu, Pd, Ag, Au) for benzene adsorption/activation. On the surfaces that do not segregate (M = Pd, Ag, Au), the preferred benzene adsorption site is the hollow Pt3-hcp site. On anti-segregated Pt-skin surfaces (M = Fe, Co, Ni, Cu, Pd), which have a top layer composed entirely of Pt, benzene favors bridge sites with a maximized number of solute atoms M in the subsurface layers. As for photocatalytic systems, Wang et al. (DOI: 10.1039/C3CP44651D) have demonstrated that the visible-light absorption and photocatalytic activities of the boron- and/or carbon-doped anatase TiO2 are not only influenced by the energy gaps (Eg) and the distributions of impurity states, but also affected by the locations of Fermi levels (EF) and the energies of the edges of band gaps. The O-rich growth condition is beneficial to the substitutional boron and carbon atoms to Ti atoms, while the Ti-rich growth condition is favorable to the other doped TiO2 including the most stable co-doped TiO2 with the interstitial B atom and the substitutional C atom to O atom. The energies of the edges of band gaps determining the dominant types of oxidants (O2, hole, ˙OH) in the photocatalytic process have also been discussed.

Metal-based planar model systems—usually investigated under ultrahigh vacuum (UHV) conditions and frequently complemented by DFT calculations—have yielded insight into the fundamental aspects of (de)hydrogenation reactions at the atomic/molecular level. An integrated approach, which combines modern surface techniques with traditional methods, can significantly enhance our understanding of a broad range of interfacial phenomena that occur on metal/oxide surfaces impacting reactivity and selectivity. Weaver et al. (DOI: 10.1039/C3CP50659B) have used temperature programmed reaction spectroscopy and molecular beam reflectivity measurements to investigate the initial dissociation of n-butane isotopologues on PdO(101) and determine kinetic parameters governing the selectivity of initial C–H(D) bond cleavage. They observe differences in the reactivity of the n-butane isotopologues on PdO(101) due to kinetic isotope effects, and find that the initial dissociation probability decreases with increasing surface temperature for each isotopologue. They also speculate that intermolecular interactions among the n-butane species are responsible for the apparent coverage dependence of the C–H bond selectivity for n-butane dissociation on PdO(101). The interactions of ethylene glycol (EG) with a partially reduced rutile TiO2(110) surface have been examined also using temperature programmed desorption by Dohnalek's group (DOI: 10.1039/C3CP50687H). The saturation coverage of EG on surface Ti rows is determined to be 0.43 of a monolayer. They have observed two major reaction channels via different surface intermediates, dehydration yielding ethylene and water and dehydrogenation yielding acetaldehyde and hydrogen. Specifically, dehydration dominates at lower EG coverages (<0.2 ML) and plateaus as the coverage is increased to saturation. Dehydrogenation primarily occurs at higher EG coverages (>0.2 ML). Huang et al. (DOI: 10.1039/C3CP50292A) have designed a FeO(111)/Pt(111) inverse model catalyst to reveal the reaction mechanism of the water–gas shift reaction (WGS, CO + H2O → H2 + CO2) and preferential oxidation reactions. They have employed XPS and TDS to study the adsorption and surface reactions of H2O, CO and HCOOH on the inverse catalyst. The FeO(111)–Pt(111) interface exposes coordination-unsaturated Fe(II) cations (Fe(II)CUS) that are capable of modifying the reactivity of neighbouring Pt sites. They show that water facilely dissociates on the Fe(II)CUS cations to form hydroxyls that further react to form both water and H2 upon heating. Hydroxyls on the Fe(II)CUS cations can also react with CO(a) on the neighbouring Pt(111) sites to produce CO2 at low temperatures. They suggest that formate is the likely surface intermediate of the CO(a) + OH reaction. Chen and his team (DOI: 10.1039/C3CP50712B) describe the change in chemical state of VOx films on Pt(111) employing classical surface science approaches. On the Pt(111) surface, VOx forms isolated O[double bond, length as m-dash]VOx (x = 0–3) species, surface two-dimensional (2D) (2 × 2)-V2O3 domains, a bi-layer structure with a (3√3 × 6) arrangement, and a complicated tri-layer structure as the coverage increases from submonolayer to multilayer. Under the UHV conditions, the oxidation state of V is mainly +3 and the stability was found to be surface V2O3 > bi-layer V2O3 > tri-layer. These VOx films can be oxidized to higher oxidation states, mainly V2O5, upon oxygen exposure, as evidenced by the shifts of the core-level binding energies and presence of V[double bond, length as m-dash]O.

There has been an unprecedented trend to obtain detailed information on the catalytic solids under technologically relevant working conditions, which can bring about fairly significant changes to the catalytically active sites. For this purpose, advanced in situ spectroscopic cells have been developed that enable researchers to investigate the physicochemical changes taking place on the catalyst surface while working at even the high temperature and pressure of a substrate in a gas or liquid phase. Rodriguez and his team (DOI: 10.1039/C3CP50416F) have employed a series of in situ state-of-the-art techniques including X-ray diffraction, pair-distribution-function analysis, X-ray absorption fine structure, environmental transmission electron microscopy, infrared spectroscopy, and ambient-pressure X-ray photoelectron spectroscopy to understand the structural, electronic and chemical properties of metal oxide catalysts used for the production of hydrogen through the WGS. They examine systematically active phases of a group of catalysts combining Cu, Au or Pt with oxides such as ZnO, CeO2, TiO2, CeOx/TiO2 and Fe2O3. The oxide support undergoes partial reduction, facilitating the dissociation of water and in some cases modifying the chemical properties of the supported metal. Accordingly, a redox mechanism or associative mechanisms that involve either carbonate-like (CO3, HCO3) or carboxyl (HOCO) species has been proposed. Weckhuysen's group (DOI: 10.1039/C3CP50646K) has looked at the deactivation of 0.5 wt% Pt/Al2O3 and 0.5 wt% Pt–1.5 wt% Sn/Al2O3 catalysts by operando Raman spectroscopy during the dehydrogenation of propane and subsequent regeneration in air for 10 successive dehydrogenation–regeneration cycles. It was found that the addition of hydrogen to the feed increases the catalyst performance and decreases the formation of coke deposits. By analyzing the related intensity, band position and bandwidth of the different Raman features, the authors show that smaller graphite crystallites, which have fewer defects, are formed when the partial pressure of hydrogen in the feed is increased. Hutchings et al. (DOI: 0.1039/C3CP50710F) have examined oxidative dehydrogenation and disproportionation of benzyl alcohol on supported gold–palladium nanoclusters employing in situ infrared inelastic neutron scattering spectroscopies. Although several interesting features were found regarding the evolution of intermediates and surface-bound species, they have boldly pointed out that even using a range of in situ vibrational spectroscopies it has proven difficult to unravel some of the subtle details in this reaction because those techniques using shorter timescales for observation of reaction species need to be considered. Walker et al. (DOI: 10.1039/C3CP51330K) have taken in situ powder neutron diffraction measurements to examine the reaction path of dehydrogenation of lithium hydride. Furthermore, they show that the addition of germanium can effectively reduce the dehydrogenation temperature via the formation of lithium germanides.

The rational design and synthesis of heterogeneous catalysts for (de)hydrogenation has long been considered more of an art than a science. Thanks to rapid developments in modern synthetic methodologies, it is now realistic to fabricate complex materials with well defined structures and desirable chemical compositions to understand better the surface phenomena during the reactions. Bowker and co-workers (DOI: 10.1039/C3CP50399B) have successfully synthesized sub-monolayer Mo oxides on Fe2O3 surfaces for selective oxidative dehydrogenation of methanol. Although hematite itself is a very poor catalyst with high selectivity to combustion products, when only 0.25 monolayers of Mo are deposited on the hematite surface, formaldehyde and CO selectivities are greatly enhanced and CO2 production is greatly diminished. Wang, Liu, and co-workers (DOI: 10.1039/C3CP50694K) have cleverly designed a sulfur-resistant Ni/SiO2 catalyst for methanation via the plasma decomposition of a nickel precursor. The plasma decomposed catalyst shows less defect sites on Ni particles, which can act to adsorb of sulfur species. Production of hydrogen from photocatalytic water splitting has become an attractive research area. However, many photocatalysts have suffered from high combination rate of charges and holes and low transfer speed of excitons. To improve the quantum efficiency, Yu's group has designed a NiS nanoparticles–CdS nanorod photocatalyst (DOI: 10.1039/C3CP50734C). The assembly of p-type NiS NPs on the surface of n-type CdS NRs forms a large number of p–n junctions, which could reduce effectively the recombination rates of electrons and holes, thus enhancing the photocatalytic activity. The CdS quantum dots have been further used to modify branched TiO2 nano-arrays leading to efficient electron transfer from the CdS phase to TiO2 (DOI: 10.1039/C3CP51291F).

In summary, this themed issue exhibits that the combination of advanced experimental approaches and electronic structure theory brings great precision to understand surface/interfacial phenomena in (de)hydrogenation reactions. It covers a broad range of reactions from hydrogenation of unsaturated C–C and C–O bonds to hydrogen production via reforming, WGS, decomposition, and photocatalytic water splitting. Twenty-three papers will appear in this special issue of Physical Chemistry Chemical Physics.

The Guest Editors would like to thank all the authors for their excellent contributions and the referees for their dedication and responsibility. We also would be happy to acknowledge Philip Earis (Editor) and Dr Heather Montgomery (Deputy Editor) for their valuable advice and cooperation as well as other editorial staff—Dr Michael Spencelayh, Dr Rowan Frame, Dr Rachel Wood, Dr Jessica Brand, Dr Tanya Smekal, and Dr William Bergius, to name just a few—for their assistance in preparing this issue throughout all the stages.


This journal is © the Owner Societies 2013