Small Au and Pt clusters at the anatase TiO2(101) surface: behavior at terraces, steps, and surface oxygen vacancies

The adsorption properties of Au and Pt metal nanoclusters on TiO2 anatase (101) were calculated using density functional theory. Structures and energetics of adsorbed Au and Pt monomers, dimers, and trimers at clean anatase TiO2(101) terraces and two major step edges, as well as O-vacancies, were systematically determined. The theoretical predictions were tested by vapor-depositing small coverages of Au and Pt on anatase (101) and investigating the resulting clusters with Scanning Tunneling Microscopy. On the clean surface, Au shows a strong tendency to form large clusters that nucleate on step edges. A preference for adsorption at type D-(112) steps is observed, which is probably a result of kinetic effects. For Pt, clusters as small as monomers are observed on the terraces, in agreement with the predicted large binding energy of 2.2 eV. Step edges play a less important role than in the case of Au. Oxygen vacancies, produced by electron irradiation, dramatically influence the growth of Au, while the nucleation behavior of Pt was found to be less affected.     

First principles theoretical study of the hole-assisted conversion of CO to CO2 on the anatase TiO2(101) surface

First principles density functional theory calculations were carried out to investigate the adsorption and oxidation of CO on the positively charged (101) surface of anatase, as well as the desorption of CO(2) from it. We find that the energy gain on adsorption covers the activation energy required for the oxidation, while the energy gain on the latter is sufficient for the desorption of CO(2), leaving an oxygen vacancy behind. Molecular dynamics simulations indicate that the process can be spontaneous at room temperature. The oxidation process described here happens only in the presence of the hole. The possibility of a photocatalytic cycle is discussed assuming electron scavenging by oxygen.     

Adsorption-site-dependent electronic structure of catechol on the anatase TiO2(101) surface

The adsorption of catechol (1,2-benzendiol) on the anatase TiO(2)(101) surface was studied with synchrotron-based ultraviolet photoemission spectroscopy (UPS), X-ray photoemission spectroscopy (XPS), and scanning tunneling microscopy (STM). Catechol adsorbs with a unity sticking coefficient and the phenyl ring intact. STM reveals preferred nucleation at step edges and subsurface point defects, followed by 1D growth and the formation of a 2 × 1 superstructure at full coverage. A gap state of ～1 eV above the valence band maximum is observed for dosages in excess of ～0.4 Langmuir, but such a state is absent for lower coverages. The formation of the band gap states thus correlates with the adsorption at regular lattice sites and the onset of self-assembled superstructures.     

CO2 adsorption on TiO2(101) anatase: a dispersion-corrected density functional theory study

Adsorption, diffusion, and dissociation of CO(2) on the anatase (101) surface were investigated using dispersion-corrected density functional theory. On the oxidized surface several different local minima were identified of which the most stable corresponds to a CO(2) molecule adsorbed at a five-fold coordinated Ti site in a tilted configuration. Surface diffusion is characterized by relatively small activation barriers. Preferential diffusion takes place along Ti rows and involves a cartwheel type of motion. The presence of a bridging oxygen defect or a surface interstitial Ti atom allows creation of several new strong binding configurations the most stable of which have bent CO(2) structures with simultaneous bonding to two surface Ti atoms. Subsurface oxygen vacancy or interstitial Ti defects are found to enhance the bonding of CO(2) molecules to the surface. CO(2) dissociation from these defect sites is calculated to be exothermic with barriers less than 21 kcal/mol. The use of such defects for catalytic activation of CO(2) on anatase (101) surface would require a mechanism for their regeneration.     

Reaction intermediates in the photoreduction of oxygen molecules at the (101) TiO2 (anatase) surface

The structural, electronic, and vibrational properties of intermediates of the O(2) photoreduction at the (101) TiO(2) (anatase) surface have been investigated by performing ab initio density functional calculations. In detail, a recently proposed approach has been used where molecules on the surface are treated like surface defects. Thus, by applying theoretical methods generally used in the physics of semiconductors, we successfully estimate the location and donor/acceptor character of the electronic levels induced by an adsorbed molecule in the TiO(2) energy gap, both crucial for the surface-molecule charge-transfer processes, and investigate the formation and the properties of charged intermediates. The present approach permits a view of the O(2) photoreduction process through several facets, which elucidates the molecule-surface charge-transfer conditions and reveals the key role played by charged intermediates. A comparison of present results with those of a highly sensitive IR (infrared) spectroscopy study of intermediates of the O(2) photoreduction leads to a deeper understanding of this process and to revised vibrational-line assignments and reaction paths.     

Density functional study of the interfacial electron transfer pathway for monolayer-adsorbed InN on the TiO(2) anatase (101) surface

Density functional theory (DFT) in connection with ultrasoft pseudopotential (USP) and generalized gradient spin-polarized approximations (GGSA) is applied to calculate the adsorption energies and structures of monolayer-adsorbed InN on the TiO(2) anatase (101) surface and the corresponding electronic properties, that is, partial density of states (PDOS) for surface and bulk layers of the TiO(2) anatase (101) surface and monolayer-adsorbed InN, to shed light on the possible structural modes for initial photoexcitation within the UV/vis adsorption region followed by fast electron injection through the InN/TiO(2) interface for an InN/TiO(2)-based solar cell design. Our calculated adsorption energies found that the two most probable stable structural modes of monolayer-adsorbed InN on the TiO(2) anatase (101) surface are (1) an end-on structure with an adsorption energy of 2.52 eV through N binding to surface 2-fold coordinated O (O(cn2)), that is, InN-O(cn2), and (2) a side-on structure with an adsorption energy of 3.05 eV through both N binding to surface 5-fold coordinated Ti (Ti(cn5)) and In bridging two surface O(cn2), that is, (O(cn2))(2)-InN-Ti(cn5). Our calculated band gaps for both InN-O(cn2) and (O(cn2))2-InN-Ti(cn5) (including a 1.0-eV correction using a scissor operator) of monolayer-adsorbed InN on the TiO(2) anatase (101) surface are red-shifted to 1.7 eV (730 nm) and 2.3 eV (540 nm), respectively, which are within the UV/vis adsorption region similar to Gratzel's black dye solar cell. Our analyses of calculated PDOS for both surface and bulk layers of the TiO(2) anatase (101) surface and monolayer-adsorbed InN on the TiO(2) anatase (101) surface suggest that the (O(cn2))(2)-InN-Ti(n5) configuration of monolayer-adsorbed InN on the TiO(2) anatase (101) surface would provide a more feasible structural mode for the electron injection through the InN/TiO(2) interface. This is due to the presence of both occupied and unoccupied electronic states for monolayer-adsorbed InN within the band gap TiO(2) anatase (101) surface, which will allow the photoexcitation within the UV/vis adsorption region to take place effectively, and subsequently the photoexcited electronic states will overlap with the unoccupied electronic states around the lowest conduction band of the TiO(2) anatase (101) surface, which will ensure the electron injection through the InN/TiO(2) interface. Finally, another thing worth our attention is our preliminary study of double-layer-adsorbed InN on the TiO(2) anatase (101) surface, that is, (O(cn2))(2)-(InN)(2)-Ti(cn5), with a calculated band gap red-shifted to 2.6 eV (477 nm) and a different overlap of electronic states between double-layer-adsorbed InN and the TiO(2) anatase (101) surface qualitatively indicated that there is an effect of the thickness of adsorbed InN on the TiO(2) anatase (101) surface on both photoexcitation and electron injection processes involved in the photoinduced interfacial electron transfer through InN/TiO(2). A more thorough and comprehensive study of different layers of InN adsorbed in all possible different orientations on the TiO(2) anatase (101) surface is presently in progress.     

Nitrogen/gold codoping of the TiO2(101) anatase surface. A theoretical study based on DFT calculations

The interaction between implanted nitrogen atoms, adsorbed gold atoms, and oxygen vacancies at the anatase TiO(2)(101) surface is investigated by means of periodic density functional theory calculations. Substitutional and interstitial configurations for the N-doping have been considered, as well as several adsorption sites for Au adatoms and different types of vacancies. Our total energy calculations suggest that a synergetic effect takes place between the nitrogen doping on one hand and the adsorption of gold and vacancy formation on the other hand. Thus, while pre-implanted nitrogen increases the adsorption energy for gold and decreases the energy required for the formation of an oxygen vacancy, pre-adsorbed gold or the presence of oxygen vacancies favors the nitrogen doping of anatase. The analysis of the electronic structure and electron densities shows that a charge transfer takes place between implanted-N, adsorbed Au and oxygen vacancies. Moreover, it is predicted that the creation of vacancies on the anatase surface modified with both implanted nitrogen and supported gold atoms produces migration of substitutional N impurities from bulk to surface sites. In any case, the most stable configurations are those where N, Au and vacancies are close to each other.     

A first principles study of Pd deposition on the TiO2(1 1 0) surface

 The adsorption of isolated Pd atoms on the (1 1 0) surface of rutile TiO₂ was investigated through ab initio embedded-cluster calculations performed at the Hartree–Fock, second-order M?ller–Plesset and Becke's three parameter hybrid method with the Lee–Yang–Parr correlation functional levels. The role played by the magnitude of the surrounding charges used in the embedding procedure was carefully analyzed. The most stable site for adsorption consisted of a fourfold hollow site in which the Pd atom was coordinated to a fivefold Ti atom, two basal oxygens, and a protruding oxygen atom. However, the adsorption energies computed after basis set superposition error corrections seemed to favor a bridge site in which the Pd atom binds two protruding oxygen atoms. A periodic slab calculation using gradient-corrected functionals and plane-wave basis sets confirmed that for full coverage, the hollow site was more stable, although Pd displacement along the fivefold Ti channels was almost free. These results agree with the experimental data obtained from scanning tunneling microscopy. Finally, the adsorption energy computed from the periodic calculations was found to be 1.88 eV. Received: 14 September 1999 / Accepted: 3 February 2000 / Published online: 19 April 2000     

Dynamics of water confined on a TiO2 (anatase) surface

Inelastic neutron scattering has been employed to probe the vibrational density of states of water confined by an oxide surface, namely, nanoparticles of the anatase polymorph of TiO2. The heat capacity of confined water has been measured by adiabatic calorimetry and compared with values derived from the vibrational density of states. Both inelastic neutron scattering and calorimetry demonstrate restricted mobility and lower heat capacity and entropy of confined water as compared to the bulk.     

Local work function of Pt clusters vacuum-deposited on a TiO2 surface

Surface topography and work function maps were simultaneously obtained for Pt-evaporated titanium dioxide (TiO(2)) surfaces by using a Kelvin probe force microscope (KPFM). Platinum clusters with diameters of 2-3 nm and heights of 0.2-0.4 nm were obtained on rutile TiO(2)(110)-(1 x 1) surfaces. The work function on the Pt clusters was smaller than that on the surrounding TiO(2) surface. With the assumption that the work function was perturbed by electric dipole moments created at the Pt-TiO(2) interface, the work function decrease indicates that dipole moments were created at the interface and directed toward the vacuum. Such dipole moments can be formed by electron transfer from the originally neutral Pt atoms to the Ti cations exposed on the (1 x 1) surface. A simple model is constructed by assuming a uniform dipole moment per unit interface area. Using this model, the size-dependent perturbation of the work function can be interpreted. The electrostatic potential is more perturbed above the Pt clusters with a larger interface area since the number density of dipole moments is equal to that of the Ti cations and is uniform. A similar correlation between the work function decrease and interface area was observed for the clusters formed on terraces and on step edges. The work function maps showed no peculiar contribution for Ti atoms exposed at the step edges. Vacuum annealing caused a considerable change in the work function on the clusters. The work function was decreased on some clusters relative to the TiO(2) substrate, while it increased on the other clusters. The atomistic structure of the interface may be modified upon annealing, thus perturbing the electron transfer across the interface.     

Synergistic effects in La/N codoped TiO2 anatase (101) surface correlated with enhanced visible-light photocatalytic activity

The interaction between implanted La, substitutional N, and an oxygen vacancy at TiO(2) anatase (101) surface has been investigated by means of first-principles density function theory calculations to investigate the origin of enhanced visible-light photocatalytic activity of La/N-codoped anatase observed in experiments. Our calculations suggest that both the adsorptive and substitutional La-doped TiO(2) anatase (101) surfaces are probably defective configurations in experiments. The h-Cave-adsorbed La doping decreases the formation energy for the substitutional N implantation and vice versa, while the charge compensation effects do not take effect between the adsorptive La and substitutional N dopants, resulting in some partially occupied states in the band gap acting as traps of the photoexcited electrons. The Ti(5c)-substituted La doping decreases the energy required for the substitutional N implantation, and the substitutional La and N codoping promotes the formation of an oxygen vacancy, which migrates from the O(sb-3c) site at the inner layer toward the surface O(b) site. For the substitutional La/N-codoped (Ti(5c)_O(3c-down)) surface, the charge compensation between the substitutional La and substitutional N leads to the formation of two isolated occupied N(s)-O π* impurity levels in the gap, while the excitation energy from the higher impurity level to the CBM decreases by about 0.89 eV. After further considering an oxygen vacancy on the Ti(5c)_O(3c-down) surface, the two electrons on the double donor levels (O(b) vacancy) passivate the same amount of holes on the acceptor levels (substitutional La and N), forming the acceptor-donor-acceptor compensation pair, which provides a reasonable mechanism for the enhanced visible-light photocatalytic activity of La/N codoped TiO(2) anatase. This knowledge may aid the further design and construction of new effective visible-light photocatalysts.     

Computational study on the reactions of H2O2 on TiO2 anatase (101) and rutile (110) surfaces

This study investigates the adsorption and reactions of H(2)O(2) on TiO(2) anatase (101) and rutile (110) surfaces by first-principles calculations based on the density functional theory in conjunction with the projected augmented wave approach, using PW91, PBE, and revPBE functionals. Adsorption mechanisms of H(2)O(2) and its fragments on both surfaces are analyzed. It is found that H(2)O(2) , H(2)O, and HO preferentially adsorb at the Ti(5c) site, meanwhile HOO, O, and H preferentially adsorb at the (O(2c))(Ti(5c)), (Ti(5c))(2), and O(2c) sites, respectively. Potential energy profiles of the adsorption processes on both surfaces have been constructed using the nudged elastic band method. The two restructured surfaces, the 1/3 ML oxygen covered TiO(2) and the hydroxylated TiO(2), are produced with the H(2)O(2) dehydration and deoxidation, respectively. The formation of main products, H(2)O(g) and the 1/3 ML oxygen covered TiO(2) surface, is exothermic by 2.8 and 5.0 kcal/mol, requiring energy barriers of 0.8 and 1.1 kcal/mol on the rutile (110) and anatase (101) surface, respectively. The rate constants for the H(2)O(2) dehydration processes have been predicted to be 6.65 × 10(-27) T(4.38) exp(-0.14 kcal mol(-1)/RT) and 3.18 × 10(-23) T(5.60) exp(-2.92 kcal mol(-1)/RT) respectively, in units of cm(3) molecule(-1) s(-1).     

Two phase morphology limits lithium diffusion in TiO(2)(anatase): a (7)Li MAS NMR study.

7Li magic angle spinning solid-state nuclear magnetic resonance is applied to investigate the lithium local environment and lithium ion mobility in tetragonal anatase TiO(2) and orthorhombic lithium titanate Li(0.6)TiO(2). Upon lithium insertion, an increasing fraction of the material changes its crystallographic structure from anatase TiO(2) to lithium titanate Li(0.6)TiO(2). Phase separation occurs, and as a result, the Li-rich lithium titanate phase is coexisting with the Li-poor TiO(2) phase containing only small Li amounts approximately equal to 0.01. In both the anatase and the lithium titanate lattice, Li is found to be hopping over the available sites with activation energies of 0.2 and 0.09 eV, respectively. This leads to rapid microscopic diffusion rates at room temperature (D(micr) = 4.7 x 10(-12) cm(2)s(-1) in anatase and D(micr) = 1.3 x 10(-11) cm(2)s(-1) in lithium titanate). However, macroscopic intercalation data show activation energies of approximately 0.5 eV and smaller diffusion coefficients. We suggest that the diffusion through the phase boundary is determining the activation energy of the overall diffusion and the overall diffusion rate itself. The chemical shift of lithium in anatase is independent of temperature up to approximately 250 K but decreases at higher temperatures, reflecting a change in the 3d conduction electron densities. The Li mobility becomes prominent from this same temperature showing that such electronic effects possibly facilitate the mobility.     

Unique adsorption behaviors of NO and O2 at hydrogenated anatase TiO2(101)

The extra electron on the hydrogenated anatase TiO₂(101) is localized at the nearest Ti_5c only, and the chargetransfer promoted NO and O₂ adsorptions are also site-selective. These results are totally different from those at hydrogenated rutile TiO₂(110).     

Reactivity of anatase TiO(2) nanoparticles: the role of the minority (001) surface

Methanol adsorption on clean and hydrated anatase TiO(2)(001)-1 x 1 is studied using density functional theory calculations and first principles molecular dynamics simulations. It is found that (i) dissociative adsorption is favored on clean TiO(2)(001) at both low and high methanol coverages; (ii) on the partially hydrated surface, methanol dissociation is not affected by the coadsorbed water and can still occur very easily; (iii) the dissociative adsorption energy of methanol is always larger than that of water under similar conditions. This implies that water replacement by methanol is energetically favored, in agreement with recent experimental observations on colloidal anatase nanoparticles.     

O2-coverage-dependent CO oxidation on reduced TiO2(110): A first principles study

First principles periodic slab calculations based on gradient-corrected density functional theory have been performed to investigate CO oxidation on rutile TiO2(110) at varying O2 coverages (theta = 1, 2, and 3, where theta is defined as the number of O2 per oxygen vacancy). For each coverage we only present the reaction of CO with oxygen species in the most stable configuration. Our results show a significant variation in the oxidation activation energy with O2 coverage.     

First principles study of O2 dissociation on Pt(111) surface: Stepwise mechanism

We have performed first‐principle density functional theory calculations to investigate O₂ dissociation on Pt(111) surface. A stepwise mechanism has been proposed. First, the adsorbed O₂ dissociate into two oxygen atoms to get adsorbed on the nearby adsorption sites. Then, oxygen atoms further migrate to other more stable adsorption sites. The influence of solvent water on oxygen dissociation was also examined. The results show that the co‐adsorption of water has little impact on O₂ dissociation. However, when water participates in the reaction, the energy barriers were reduced greatly. These results have very important significance to understand the mechanism of oxygen reduction. © 2016 Wiley Periodicals, Inc.     

Adsorption of thiols on the Pd(111) surface: a first principles study

Using density functional theory formalism, we have investigated the adsorption behavior of thiols on the Pd(111) surface. Two different thiol molecules, viz. (a) methane thiol and (b) thiophene 2-thiol (TSH), were used as model adsorbates for this purpose. The results revealed that whereas the methane thiol molecule undergoes spontaneous dissociative chemisorption onto the palladium surface, the adsorption of the thiophene 2-thiol molecule does not involve cleavage of the S-H bond, leading to weak interaction energy or physisorption. The variation in the adsorption behavior has been explained based on the difference in the electronic environment of the terminal sulfur atom. The nature of binding at the interface has been analyzed through calculation of the partial density of states of the sulfur atom at the interface.     

Vertical and lateral order in adsorbed water layers on anatase TiO2(101)

The structure and energetics of thin water overlayers on the (101) surface of TiO(2)-anatase have been studied through first-principles molecular dynamics simulations at T = 160 K. At one monolayer coverage, H(2)O molecules are adsorbed at the 5-fold Ti sites (Ti(5c)), forming an ordered crystal-like 2D layer with no significant water-water interactions. For an adsorbed bilayer, H(2)O molecules at both Ti(5c) and bridging oxygen (O(2c)) sites form a partially ordered structure, where the water oxygens occupy regular sites but the orientation of the molecules is disordered; in addition, stress-relieving defects are usually present. When a third layer is adsorbed, very limited parallel and perpendicular order is observed above the first bilayer. The calculated energetics of multilayer adsorption is in good agreement with recent temperature-programmed desorption data.