Hydrogen spillover1 is the surface migration of activated hydrogen atoms from a metal catalyst particle, on which they are generated, onto the catalyst support2. The phenomenon has been much studied3,4,5,6,7 and its occurrence on reducible supports such as titanium oxide is established, yet questions remain about whether hydrogen spillover can take place on nonreducible supports such as aluminium oxide8,9,10,11,12,13. Here we use the enhanced precision of top-down nanofabrication14,15 to prepare controlled and precisely tunable model systems that allow us to quantify the efficiency and spatial extent of hydrogen spillover on both reducible and nonreducible supports. We place multiple pairs of iron oxide and platinum nanoparticles on titanium oxide and aluminium oxide supports, varying the distance between the pairs from zero to 45 nanometres with a precision of one nanometre. We then observe the extent of the reduction of the iron oxide particles by hydrogen atoms generated on the platinum using single-particle in situ X-ray absorption spectromicroscopy14 applied simultaneously to all particle pairs. The data, in conjunction with density functional theory calculations16,17, reveal fast hydrogen spillover on titanium oxide that reduces remote iron oxide nanoparticles via coupled proton–electron transfer. In contrast, spillover on aluminium oxide is mediated by three-coordinated aluminium centres that also interact with water and that give rise to hydrogen mobility competing with hydrogen desorption; this results in hydrogen spillover about ten orders of magnitude slower than on titanium oxide and restricted to very short distances from the platinum particle. We anticipate that these observations will improve our understanding of hydrogen storage18,19 and catalytic reactions involving hydrogen8,11,12,13, and that our approach to creating and probing model catalyst systems will provide opportunities for studying the origin of synergistic effects in supported catalysts that combine multiple functionalities.
This is a preview of subscription content, access via your institution
Open Access articles citing this article.
Nature Communications Open Access 18 March 2022
Nature Communications Open Access 14 January 2022
Nature Communications Open Access 10 January 2022
Subscribe to Nature+
Get immediate online access to the entire Nature family of 50+ journals
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Khoobiar, S. Particle to particle migration of hydrogen atoms on platinum—alumina catalysts from particle to neighboring particles. J. Phys. Chem. 68, 411–412 (1964)
Conner, W. C. & Falconer, J. L. Spillover in heterogeneous catalysis. Chem. Rev. 95, 759–788 (1995)
Beaumont, S. K., Alayoglu, S., Specht, C., Kruse, N. & Somorjai, G. A. A nanoscale demonstration of hydrogen atom spillover and surface diffusion across silica using the kinetics of CO2 methanation catalyzed on spatially separate Pt and Co nanoparticles. Nano Lett. 14, 4792–4796 (2014)
Merte, L. R. et al. Water-mediated proton hopping on an iron oxide surface. Science 336, 889–893 (2012)
Marcinkowski, M. D. et al. Controlling a spillover pathway with the molecular cork effect. Nat. Mater. 12, 523–528 (2013)
Im, J., Shin, H., Jang, H., Kim, H. & Choi, M. Maximizing the catalytic function of hydrogen spillover in platinum-encapsulated aluminosilicates with controlled nanostructures. Nat. Commun. 5, 3370 (2014)
Collins, S. S. E., Cittadini, M., Pecharromán, C., Martucci, A. & Mulvaney, P. Hydrogen spillover between single gold nanorods and metal oxide supports: a surface plasmon spectroscopy study. ACS Nano 9, 7846–7856 (2015)
Prins, R. Hydrogen spillover. Facts and fiction. Chem. Rev. 112, 2714–2738 (2012)
Prins, R., Palfi, V. K. & Reiher, M. Hydrogen spillover to nonreducible supports. J. Phys. Chem. C 116, 14274–14283 (2012)
Roland, U., Braunschweig, T. & Roessner, F. On the nature of spilt-over hydrogen. J. Mol. Catal. Chem. 127, 61–84 (1997)
Esposito, D. V., Levin, I., Moffat, T. P. & Talin, A. A. H2 evolution at Si-based metal-insulator-semiconductor photoelectrodes enhanced by inversion channel charge collection and H spillover. Nat. Mater. 12, 562–568 (2013)
Nabaho, D., Niemantsverdriet, J. W., Claeys, M. & van Steen, E. Hydrogen spillover in the Fischer–Tropsch synthesis: an analysis of platinum as a promoter for cobalt–alumina catalysts. Catal. Today 261, 17–27 (2016)
Pham, V. H. et al. A catalytic and efficient route for reduction of graphene oxide by hydrogen spillover. J. Mater. Chem. A 1, 1070–1077 (2013)
Karim, W. et al. Size-dependent redox behavior of iron observed by in-situ single nanoparticle spectro-microscopy on well-defined model systems. Sci. Rep. 6, 18818 (2016)
Karim, W. et al. High-resolution and large-area nanoparticle arrays using EUV interference lithography. Nanoscale 7, 7386–7393 (2015)
VandeVondele, J. et al. Quickstep: fast and accurate density functional calculations using a mixed Gaussian and plane waves approach. Comput. Phys. Commun. 167, 103–128 (2005)
Spreafico, C. & VandeVondele, J. The nature of excess electrons in anatase and rutile from hybrid DFT and RPA. Phys. Chem. Chem. Phys. 16, 26144–26152 (2014)
Psofogiannakis, G. M. & Froudakis, G. E. Fundamental studies and perceptions on the spillover mechanism for hydrogen storage. Chem. Commun. 47, 7933–7943 (2011)
Li, Y. & Yang, R. T. Significantly enhanced hydrogen storage in metal-organic frameworks via spillover. J. Am. Chem. Soc. 128, 726–727 (2006)
Lykhach, Y. et al. Hydrogen spillover monitored by resonant photoemission spectroscopy. J. Catal. 285, 6–9 (2012)
Bennett, R. A., Stone, P. & Bowker, M. Pd nanoparticle enhanced re-oxidation of non-stoichiometric TiO2: STM imaging of spillover and a new form of SMSI. Catal. Lett. 59, 99–105 (1999)
Panayotov, D. A. & Yates, J. T. Spectroscopic detection of hydrogen atom spillover from Au nanoparticles supported on TiO2: use of conduction band electrons. J. Phys. Chem. C 111, 2959–2964 (2007)
Alayoglu, S. et al. Pt-mediated reversible reduction and expansion of CeO2 in Pt nanoparticle/mesoporous CeO2 catalyst: in situ X-ray spectroscopy and diffraction studies under redox (H2 and O2) atmospheres. J. Phys. Chem. C 117, 26608–26616 (2013)
van Bokhoven, J. & Lamberti, C. X-ray Absorption and X-ray Emission Spectroscopy: Theory and Applications (John Wiley & Sons, 2016)
Vaz, C. A. F., Balan, A., Nolting, F. & Kleibert, A. In situ magnetic and electronic investigation of the early stage oxidation of Fe nanoparticles using X-ray photo-emission electron microscopy. Phys. Chem. Chem. Phys. 16, 26624–26630 (2014)
Manfrinato, V. R. et al. Resolution limits of electron-beam lithography toward the atomic scale. Nano Lett. 13, 1555–1558 (2013)
van Bokhoven, J. A., van der Eerden, A. M. J. & Koningsberger, D. C. Three-coordinate aluminum in zeolites observed with in situ X-ray absorption near-edge spectroscopy at the Al K-edge: flexibility of aluminum coordinations in zeolites. J. Am. Chem. Soc. 125, 7435–7442 (2003)
Wischert, R., Laurent, P., Copéret, C., Delbecq, F. & Sautet, P. γ-Alumina: the essential and unexpected role of water for the structure, stability, and reactivity of “defect” sites. J. Am. Chem. Soc. 134, 14430–14449 (2012)
Stoyanov, E., Langenhorst, F. & Steinle-Neumann, G. The effect of valence state and site geometry on Ti L3,2 and O K electron energy-loss spectra of TixOy phases. Am. Mineral. 92, 577–586 (2007)
Digne, M., Sautet, P., Raybaud, P., Euzen, P. & Toulhoat, H. Use of DFT to achieve a rational understanding of acid–basic properties of γ-alumina surfaces. J. Catal. 226, 54–68 (2004)
Puurunen, R. L. Surface chemistry of atomic layer deposition: a case study for the trimethylaluminum/water process. J. Appl. Phys. 97, 121301 (2005)
Lim, G. T. & Kim, D.-H. Characteristics of TiOx films prepared by chemical vapor deposition using tetrakis-dimethyl-amido-titanium and water. Thin Solid Films 498, 254–258 (2006)
Fraile Rodríguez, A., Nolting, F., Bansmann, J., Kleibert, A. & Heyderman, L. J. X-ray imaging and spectroscopy of individual cobalt nanoparticles using photoemission electron microscopy. J. Magn. Magn. Mater. 316, 426–428 (2007)
van Bokhoven, J. A., van der Eerden, A. M. J. & Koningsberger, D. C. in Studies in Surface Science and Catalysis Vol. 142 (eds Giordano, G ., Aiello, R. & Testa, F. ) 1885–1890 (Elsevier, 2002)
Jozwiak, W. K., Kaczmarek, E., Maniecki, T. P., Ignaczak, W. & Maniukiewicz, W. Reduction behavior of iron oxides in hydrogen and carbon monoxide atmospheres. Appl. Catal. A 326, 17–27 (2007)
Regan, T. et al. Chemical effects at metal/oxide interfaces studied by X-ray-absorption spectroscopy. Phys. Rev. B 64, 214422 (2001)
Sha, X., Chen, L., Cooper, A. C., Pez, G. P. & Cheng, H. Hydrogen absorption and diffusion in bulk α-MoO3 . J. Phys. Chem. C 113, 11399–11407 (2009)
Chen, L., Cooper, A. C., Pez, G. P. & Cheng, H. On the mechanisms of hydrogen spillover in MoO3 . J. Phys. Chem. C 112, 1755–1758 (2008)
Lippert, G., Hutter, J. & Parrinello, M. A hybrid Gaussian and plane wave density functional scheme. Mol. Phys. 92, 477–487 (1997)
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996)
Adamo, C. & Barone, V. Toward reliable density functional methods without adjustable parameters: the PBE0 model. J. Chem. Phys. 110, 6158–6170 (1999)
Guidon, M., Hutter, J. & VandeVondele, J. Robust periodic Hartree-Fock exchange for large-scale simulations using Gaussian basis sets. J. Chem. Theory Comput. 5, 3010–3021 (2009)
Spreafico, C. & VandeVondele, J. Excess electrons and interstitial Li atoms in TiO2 anatase: properties of the (101) interface. J. Phys. Chem. C 119, 15009–15018 (2015)
Guidon, M., Hutter, J. & VandeVondele, J. Auxiliary density matrix methods for Hartree-Fock exchange calculations. J. Chem. Theory Comput. 6, 2348–2364 (2010)
Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132, 154104 (2010)
Hartwigsen, C., Goedecker, S. & Hutter, J. Relativistic separable dual-space Gaussian pseudopotentials from H to Rn. Phys. Rev. B 58, 3641–3662 (1998)
Mills, G., Jónsson, H. & Schenter, G. K. Reversible work transition state theory: application to dissociative adsorption of hydrogen. Surf. Sci. 324, 305–337 (1995)
Henkelman, G., Uberuaga, B. P. & Jónsson, H. A climbing image nudged elastic band method for finding saddle points and minimum energy paths. J. Chem. Phys. 113, 9901–9904 (2000)
Vittadini, A., Casarin, M. & Selloni, A. Chemistry of and on TiO2-anatase surfaces by DFT calculations: a partial review. Theor. Chem. Acc. 117, 663–671 (2007)
Burdett, J. K., Hughbanks, T., Miller, G. J., Richardson, J. W., Jr & Smith, J. V. Structural-electronic relationships in inorganic solids: powder neutron diffraction studies of the rutile and anatase polymorphs of titanium dioxide at 15 and 295 K. J. Am. Chem. Soc. 109, 3639–3646 (1987)
Krokidis, X. et al. Theoretical study of the dehydration process of boehmite to γ-alumina. J. Phys. Chem. B 105, 5121–5130 (2001)
Helali, Z., Jedidi, A., Markovits, A., Minot, C. & Abderrabba, M. Reactivity of transition metal atoms supported or not on TiO2(110) toward CO and H adsorption. Theor. Chem. Acc. 134, http://dx.doi.org/10.1007/s00214-015-1652-4 (2015)
Helali, Z., Markovits, A., Minot, C. & Abderrabba, M. Support effect on H adsorption on a metal atom. Chem. Phys. Lett. 565, 45–51 (2013)
Chen, H.-Y. T., Tosoni, S. & Pacchioni, G. Hydrogen adsorption, dissociation, and spillover on Ru10 clusters supported on anatase TiO2 and tetragonal ZrO2 (101) surfaces. ACS Catal. 5, 5486–5495 (2015)
Conradie, J., Gracia, J. & Niemantsverdriet, J. W. Energetic driving force of H spillover between rhodium and titania surfaces: a DFT view. J. Phys. Chem. C 116, 25362–25367 (2012)
Setvin, M., Schmid, M. & Diebold, U. Aggregation and electronically induced migration of oxygen vacancies in TiO2 anatase. Phys. Rev. B 91, 195403 (2015)
Einstein, A. Über die von der molekularkinetischen Theorie der Wärme geforderte Bewegung von in ruhenden Flüssigkeiten suspendierten Teilchen. Ann. Phys. 322, 549–560 (1905)
Oura, K., Katayama, M., Zotov, A., Lifshits, V. & Saranin, A. Surface Science: An Introduction (Springer, 2003)
Part of this work was performed at the Swiss Light Source, Paul Scherrer Institute, Switzerland. We acknowledge the Swiss Light Source for providing synchrotron radiation beamtime at the Surfaces/Interfaces Microscopy beamline. We thank S. A. Tschupp for his assistance during beamtime, V. Guzenko for technical support with the EBL tool, M. Wipf for atomic layer deposition, and M. Ranocchiari and M. Schoenberg for their comments on the manuscript. The Paul Scherrer Institute (PSI) Research Commission provided financial support through the CROSS programme. J.A.v.B. thanks the NCCR MUST and J.V. acknowledges financial support by the European Union FP7 in the form of an ERC Starting Grant (contract number 277910). Calculations were supported by a grant from the Swiss National Supercomputer Center (Project ID ch5).
The authors declare no competing financial interests.
Reviewer Information Nature thanks E. van Steen and the other anonymous reviewer(s) for their contribution to the peer review of this work.
Extended data figures and tables
Extended Data Figure 1 Nanofabrication to achieve model system for hydrogen spillover with particle placement having a precision of 1 nm.
a, Design of the sample with fifteen pairs of iron oxide and platinum particles and a single iron oxide particle over an area of 3 × 3 μm2. The iron oxide particles (red) have a diameter of 60 nm while the platinum particles (green) have a diameter of 30 nm. The distance between the dimeric pairs is 1 μm. b, SEM image of the nanofabricated model system on titanium oxide support (5 nm thick) which is an exact one-to-one rendition of the design (500 nm scale bar). c, Scheme of different stages of overlaid EBL to fabricate the model system (details in Methods).
Fe L3 edge XAS spectra of the iron nanoparticles at room temperature in pair ‘a1’ (overlapping pair), pair ‘c2’ (just touching each other), pair ‘d3’ (45 nm distance) and system ‘d4’, which has no platinum in its vicinity. All the iron nanoparticles have the same initial state, which is the native oxide, formed by exposure of iron particles in air after nanofabrication. Iron oxidation occurs through initial formation of iron(II) oxide followed by progressive transformation to iron(II,III) oxide. A small amount of iron(II) oxide remains owing to its limited oxidation at the interface14,36. The XAS spectrum of the native oxide state shows a resonance peak and a shoulder. With reduction of the iron oxide, the peak resonance starts to decrease and the shoulder peak intensity increases14.
Extended Data Figure 3 Speciation of iron oxide particles on aluminium oxide in the 16 systems at 343 K.
The fits are extracted using linear combination fitting of the experimental XAS spectra with the reference of metallic iron(0), iron(II) oxide, iron(II,III) oxide, and iron(III) oxide14,36. The Fe L3 edge XAS spectra in Fig. 2a shows higher reduction for the iron oxide particles at distances less than 15 nm from the platinum particles, that is, in pairs ‘a1’ to ‘c4’. The initial state was iron(II,III) oxide and iron (II) oxide (Extended Data Fig. 2). During reduction, the fraction of iron(II,III) oxide decreases and that of iron (II) oxide increases. A decrease in the fraction of iron(II) oxide on further reduction results in the evolution of metallic iron(0). The pair with overlapping iron oxide and platinum particles (pair ‘a1’) is most reduced, producing 16.8% of iron(0). The extent of reduction decreases as the distance between the particles increases in each following pair. The iron oxide particle in pair ‘c4’ (with an interparticle distance of 15 nm) has 1.2% of iron(0) and the iron oxide particles in pairs ‘d1’, ‘d2’ and ‘d3’ and system ‘d4’, at distances more than 15 nm from their corresponding platinum particles, have no iron(0). This shows the gradient of reduction depending on the distance from the platinum catalyst up to 15 nm.
a, Al K edge XAS spectra of the aluminium oxide support at around 300 K and 10−10 mbar without hydrogen. The absorption edge observed in the measured spectra at 1,566 eV and the characteristic broad peak at about 20 eV indicate a dominant tetrahedral aluminium coordination. An increase in intensity in the range 1,567–1,575 eV is visible, along with the lower white line intensity and the lower intensity of the broad peak at 1,583 eV compared to purely tetrahedral aluminium coordination: this is indicative of a small amount of octahedrally coordinated aluminium at the cost of tetrahedrally coordinated aluminium34. The visible pre-edge feature is characteristic of three-coordinated aluminium sites (Al3c)27, which confirms their existence under the experimental conditions. At higher water coverage, these Al3c sites are known to be unavailable28, which will lead to the disappearance of this pre-edge feature. b, Calculated thermodynamic diagram showing that the availability of Al3c sites depends on the partial pressure of water in the system. Coloured areas indicate the percentage of free Al sites on the surface for a given value of temperature and .
Fe L3 edge XAS spectra of the single iron oxide particle on the titanium oxide support, without any platinum (particle ‘d4’ in Extended Fig. 1a) compared with the overlapping pair (pair ‘a1’) and the pair with interparticle distance 45 nm (pair ‘d3’) at different stages of reduction: a, Initial state at room temperature; b, at 343 K; and c, at 353 K. All the iron oxide particles are equally reduced irrespective of the distance from platinum.
Extended Data Figure 6 Speciation of iron oxide particles on titanium oxide at different temperatures.
Fits are extracted using linear combination fitting of the experimental XAS spectra with the reference of metallic iron(0), iron (II) oxide, iron(II,III) oxide, and iron(III) oxide14,36. The XAS spectra of all 16 iron oxide nanoparticles on titanium oxide coincide at all temperatures (Extended Data Fig. 5). The iron–iron oxide core–shell particles have multilayered scales of iron(0), iron (II) oxide, iron(II,III) oxide and iron(III) oxide at the native oxide stage at 298 K. With hydrogen dosage of 1 × 10−5 mbar, the particles start to reduce as temperature is increased leading to interconversion of oxides and evolution of a metallic iron core. Initial reduction of the outermost shells of iron(III) oxide and iron(II,III) oxide leads to an increase in concentration of iron(II) oxide, some of which reduces to metallic iron. The concentration of metallic iron, signifying the extent of reduction at each stage, increases further as iron(II) oxide reduces and the iron(III) oxide is negligible. 17% of iron(0) is seen at 343 K for all iron particles, which increases to 33.6% at 353 K as the particles reduce further.
Spectra are taken at each of the pairs, at the lone iron oxide particle and in an isolated area on the surface at least 1 μm from any platinum or iron oxide particles at different temperatures: a, 300 K; b, 323 K; c, 343 K; and d, 353 K. Reduction of titanium oxide occurs uniformly all over the surface, irrespective of the distance from platinum. The rapid diffusion of H atoms results in a range much larger than 1 μm for the reduction of the surface.
a, The most stable hydrogen coverage of a Pt13 cluster supported on anatase, as a function of temperature and partial hydrogen pressure . Hydrogen adsorption on the metal becomes less favourable with increasing temperature and decreasing . The average number of hydrogen atoms per Pt cluster is indicated by coloured lines. They are indicative only, because of the multitude of possible hydrogen and platinum configurations. The experimental conditions correspond to a coverage between 28 and 18 H atoms on a Pt13 cluster, yielding an estimated H:Pt ratio between 1.15 and 2.38 under the experimental conditions. b, The most stable coverage of the anatase (101) surface at 350 K, as a function of and . Hydrogen and water adsorb on two different sites on anatase, so that a mixed H2/H2O coverage is possible for a relatively wide range of and (green–yellow–orange areas in the graph). The percentage coverage indicated in the legend refer to the percentage of occupied two-coordinated oxygen (O2c) sites (H adsorption) and the occupied five-coordinated titanium (Ti5c) sites (H2O adsorption), respectively.
a, Eact profiles computed with the NEB method for migration of hydrogen atoms on the anatase surface in presence (blue) or absence (red) of a neighbouring water molecule, acting as a bridge between sites. b, Comparison of activation energies for proton migration (green) and hydrogen migration (blue) via a bridging water molecule. Proton migration in the absence of water is also reported (black). c, Ball-and-stick model of the oxide surface with co-adsorbates, describing the molecular parameters used to compute the reaction coordinate. Colour coding: red, O; yellow, H; silver, Ti. The blue density surface represents a localized electron in a filled Ti 3d orbital. The reaction coordinate without water mediation is computed as the combination of interatomic distances, . The reaction coordinate with water mediation is computed as . When water molecules are present on the surface, the distances between surface hydrogen atoms and the neighbouring water molecules are also reported: 2.14 Å and 2.34 Å.
Extended Data Figure 10 Illustration of the most relevant mechanisms of hydrogen adsorption and migration on titanium oxide (101) and γ-aluminium oxide (100), computed by the NEB method.
The results corroborate experimental observations. All energies are normalized to the energy of the isolated hydrogen molecule in the gas phase. Colour coding in the ball-and-stick models: red, O; yellow, H; silver, Ti; blue, Pt; and cyan, Al. Filled Ti 3d orbitals are indicated by a blue density surface. In each of the two surface models the main stages of hydrogen adsorption are illustrated. A is the gas-phase hydrogen molecule; B is a hydrogen molecule weakly interacting with the oxide surface (physisorption); and C is a hydrogen molecule dissociatively adsorbed on the oxide surface. a, Hydrogen dissociation on clean anatase titanium oxide: Eact = 2.05 eV. b, Hydrogen dissociation on γ-aluminium oxide: Eact = 0.63 eV on the clean surface (red) and Eact = 0.99 eV on the hydrated surface (blue).
Visualization of the nanofabricated model system in the X-ray photoemission electron microscope (X-PEEM)
X-PEEM images for the range of photon energies around the Fe L3 edges generating a XAS spectra video, showing the sixteen iron oxide nanoparticles ofb the model system in Fig. 1c. Platinum has no absorption edge in this energy, and is therefore invisible but the position of platinum particles is known from SEM images. Bright intensity corresponds to the high absorption intensity at the Fe L3 absorption edge. Individual nanoparticles are then selected to extract XAS spectra from individual particles enabling simultaneous spectroscopy of all sixteen particles. (AVI 3608 kb)
About this article
Cite this article
Karim, W., Spreafico, C., Kleibert, A. et al. Catalyst support effects on hydrogen spillover. Nature 541, 68–71 (2017). https://doi.org/10.1038/nature20782
Nature Reviews Chemistry (2022)
Nature Communications (2022)
Nature Communications (2022)
Nature Communications (2022)
Catalysis Letters (2022)