Scanning Probe Microscopy We are very active in the simulation of Scanning Probe Microscopy (SPM), particulary Atomic Force Microscopy (AFM), Scanning Tunneling Microscopy (STM) and Kelvin Probe Force Microscopy (KPFM). In general, experimental studies of surfaces with atomic resolution are often difficult to interpret and simulations provide a link between measured images and surface topography and electronic structure. SPM techniques can be applied to the study of a wide variety of systems, including ideal and defective surfaces, molecular adsorbates, organic films, DNA, proteins, and the corresponding theoretical toolbox is equally wide.

On the TiO₂ (110) surface, combined SFM experiments and simulations show that adsorbed water on the surface marks the surface sublattices, allowing us to identify the tip polarity. This provides unprecedented atomic scale information on the structure and behaviour of dissociated water on the surface. Specifically, we find that two very different types of resolution with SFM are predominant on the hydroxylated TiO₂ (110) surface. In one type of image contrast, the five-fold coordinated Ti rows are imaged as bright rows and the OH groups are visible as surprisingly bright protrusions located in between Ti rows. Conversely, in the other predominant type of images the hydroxyls are imaged as deep depressions in line with the rows of bridging oxygen atoms. A more detailed study of the time dependence of the image contrast, reveals the dynamic nature of the water reaction. If the surface is imaged rapidly after preparation we observe oxygen vacancies, and single and double OH groups, but as time passes more water reacts at vacancies until images are dominated by only OH species. Our current studies focus on using combined STM and SFM to study subsurface defects in TiO₂.

Recently, it was proposed to dope NaCl crystals with divalent impurity cations as part of a combined SFM and Kelvin probe microscopy study. Above a certain divalent impurity content, the doped NaCl system creates precipitates in their well-known Suzuki phase on the surface. The precipitates are embedded in the NaCl(001) matrix, so that two different types of surfaces regions, which are well separated, can be found. A quantitative comparison between experiment and our simulations shows that all ions of the Suzuki structure on (001) surfaces of Mg²⁺ or Cd²⁺ doped NaCl crystals can be unambiguously identified despite the tip-surface distance, differences in impurity chemistry and surface termination. The unambiguous identification is valid for any possible tip termination and can be used to calibrate the potential of the tip's last atom. It is proposed to use these surfaces in particular for better characterization of deposited nano-objects, also taking advantage of the associated nano-templating of the surface with areas of different electronic and reactive properties.

Exact exchange Standard density functional theory (DFT) ab initio approaches for materials simulations perform exceptionally for studying the ground state properties of most systems. However, for certain properties, such as band gaps, organic molecular structures and magnetic coupling, improvements beyond vanilla DFT are required. We are implementing 'hybrid' exchange-correlation functionals into the SIESTA code. These hybrid functionals use either the full Hartree-Fock exchange, or a screened version for computational efficiency, combined with standard correlation to provide an improved functional.

Nanomanipulation Scanning Probe Microscopy (SPM) has become a dominant tool for the design and activation of nanodevices. SPM direct mechanical manipulation can be used to build nanosystems piece by piece. It relies on locally changing the migration barriers by close-approach with the SPM tip – “pushing”, “pulling” the molecule. This offers local control of molecular properties, and is in principle viable for any system. More importantly, SPM provides in situ high-resolution characterization, allowing direct observation of the dynamics even at the atomic scale. As well as being a tool for nanomanipulation, SPM provides a direct measure of the energy dissipated at the nanocontact between tip and cluster, and offers insight into the energy loss due to friction during manipulation. In a pioneering study, we recently showed that it is possible to identify the chemical stages of the reaction of water with a salt surface by calculating the adsorption and diffusion barriers of various complexes and comparing them to manipulation experiments. We aim to extend this to other molecules and materials, and show that nanomanipulation can be used to characterize the reaction environment and even control the reaction itself. In parallel to our studies of the catalytic properties of metal nanoclusters on insulators, we study the lateral mobility and dissipation of nanoclusters adsorbed on surfaces. The nanoclusters also represent prototypical systems for the study of electron transport phenomena in nanostructures. Manipulation can be used to form structures of choice, such as ordered arrays of metallic nanoclusters acting as metallic nanowires, with their transport properties controlled by the cluster size and the cluster density. The current techniques available for controlling size, charge and adsorption site of deposited metallic nanoclusters provides a large investigative space for systematically studying the influence of these factors on cluster mobility and dissipated energy.

Nanotribology The study of friction remains an important element in the development of many industrial and technological processes, which are wear-dependent. As the components of technologies are reduced in size, then the resolution of understanding must also increase. We intend to understand friction and wear properties in the extreme case of atomic scale friction, were only a few atoms constitute the tip-sample contact. Despite recent successes in the understanding atomic friction processes, where the velocity dependence, load dependence and new effects like superlubricity (structural and externally induced) have been targeted, many properties are still heavily under dispute. At this point, a multitude of experimental and theoretical work exists, however, only a few papers report on the direct overlap of experiments and theory. Atomic scale friction is particularly well suited for direct comparison, since the contact size is as small as possible, and thus much better defined then in conventional tribology experiments. This invites direct comparison of atomic friction experiments with molecular dynamics simulations (MD) based on discrete atom geometries. We are particularly interested in recent developments in using different oscillation modes of an AFM to probe lateral forces on surfaces - torsional resonance mode. We are also part of a large European Network studying friction at the nanoscale, FANAS.

Nanocatalysis Much of the research into heterogeneous catalysis remains focused on the properties of small metal nanoclusters adsorbed onto surfaces of single oxide materials. Although these model catalytic systems are far from the real industrial systems, they can provide great insight into the nature of the fundamental reactions at the heart of the catalytic process. Due to their particular relevance as replacements for Pt- or Pd-based catalytic converters in car exhausts, many recent studies have focused on studying the reactivity of small gold clusters . These gold nanoclusters are catalytically active only when the size of the cluster is less than about 4 nm. The reason for this increased reactivity remains controversial The most recent results suggest that primarily the low-coordinated atoms are bonding sites for CO and O₂ and that the interaction with the substrate or charge transfer could also play a role in the catalysis by gold clusters . Scanning Probe Microscopy offers an excellent possibility for in situ studies of the properties of nanoclusters, and Scanning Tunnelling Microscopy (STM) has been applied to several model systems, but the requirement of a conducting sample has restricted the STM's access to the important class of insulating surfaces. In principle non-contact Atomic Force Microscopy (AFM) offers the capability of imaging both the adsorbed metal clusters and the insulating surface in atomic resolution, hence providing unprecedented information about the cluster structural properties. In this work, we use a combination of first principles theory and AFM experiments.

Magnetism of carbon nanostructures The origin of the magnetic signal reported for various carbon-based materials such as fullerenes and graphite has not yet been fully understood, but in principle would offer the tantalizing prospect of a zero gap, high temperature, semiconducting ferromagnet. In order to study this phenomenon we have applied extensive first principles calculations to study the role of intrinsic defects and non-magnetic impurities on the surface of graphite and carbon nanotubes. Initially we considered the most common defects in these systems, carbon adatoms and vacancies. We found that these do exist in magnetic configurations on both graphite and most nanotubes, but that the mobility of adatoms on the surface is likely to reduce their influence. We also found that hydrogen forms very stable magnetic configurations within graphite, providing both an intrinsic magnetic moment and also preventing the saturation of magnetic vacancies by diffusing adatoms. This agrees with recent experimental studies which showed a clear magnetic signal after graphite samples were irradiated with protons. Future studies concentrate on the role of doping in the magnetism of graphene and carbon nanostructures.