Low-Scaling Algorithm for Nudged Elastic Band Calculations Using a Surrogate Machine Learning Model.

We present the incorporation of a surrogate Gaussian process regression (GPR) atomistic model to greatly accelerate the rate of convergence of classical nudged elastic band (NEB) calculations. In our surrogate model approach, the cost of converging the elastic band no longer scales with the number of moving images on the path. This provides a far more efficient and robust transition state search. In contrast to a conventional NEB calculation, the algorithm presented here eliminates any need for manipulating the number of images to obtain a converged result. This is achieved by inventing a new convergence criteria that exploits the probabilistic nature of the GPR to use uncertainty estimates of all images in combination with the force in the saddle point in the target model potential. Our method is an order of magnitude faster in terms of function evaluations than the conventional NEB method with no accuracy loss for the converged energy barrier values.

Decoupling strain and ligand effects in ternary nanoparticles for improved ORR electrocatalysis.

Ternary Pt-Au-M (M = 3d transition metal) nanoparticles show reduced OH adsorption energies and improved activity for the oxygen reduction reaction (ORR) compared to pure Pt nanoparticles, as obtained by density functional theory. The strain and ligand effects in nanoparticles are decoupled and correlated with the extended Pt(111) surface for benchmarking. The ternary metal in the core allows for tuning the catalytic activity through strain effects. Pt-Au-M for M = Cr, Mn, Co, Cu, Zn nanoparticles are of particular interest as they exhibit an optimal contribution of strain, ligand effects and stability. Good agreement is found with experimental studies showing increased activity of Pt-Au-Fe/Ni nanoparticles, and mid to late 3d transition metals are predicted to exhibit enhance activity and stability with respect to pure Pt nanoparticles.

DFT studies of oxygen dissociation on the 116-atom platinum truncated octahedron particle.

Density functional theory calculations are performed to investigate oxygen dissociation on 116-atom truncated octahedron platinum particles. This work builds on results presented previously [Jennings et al., Nanoscale, 2014, 6, 1153], where it was shown that shell flexibility played an important role in facilitating fast oxygen dissociation. In this study, through investigation of the larger particle size, it is shown that oxygen dissociation on the (111) facet of pure platinum species is still aided by shell flexibility at larger sizes. Only the hollow sites close to the edges of the (111) facet mediate oxygen dissociation; oxygen is bound too weakly at other hollow sites for dissociation to occur. Further studies are performed on the (100) facet, which is larger for the Pt116 particle than for either the Pt38 or Pt79 ones. Much higher dissociation barriers are found on the (100) facet compared to the (111) facet, where the bridge sites are favourable for oxygen dissociation.

Theoretical studies of Pt-Ti nanoparticles for potential use as PEMFC electrocatalysts.

A theoretical investigation is presented of alloying platinum with titanium to form binary Pt-Ti nanoalloys as an alternative to the expensive pure platinum catalysts commonly used for Proton Exchange Membrane Fuel Cell cathode electrocatalysts. Density Functional Theory calculations are performed to investigate compositional effects on structural properties as well as Oxygen Reduction Reaction kinetics and poisoning effects. High symmetry A(32)-B(6) clusters are studied to investigate structural properties. From these structures binding energies of hydroxyl and carbon monoxide are studied on a range of sites on the surface of the clusters. Promising results are obtained suggesting that the bimetallic Pt-Ti nanoalloys may exhibit enhanced properties compared to pure platinum catalysts.

The atomic simulation environment-a Python library for working with atoms.

The atomic simulation environment (ASE) is a software package written in the Python programming language with the aim of setting up, steering, and analyzing atomistic simulations. In ASE, tasks are fully scripted in Python. The powerful syntax of Python combined with the NumPy array library make it possible to perform very complex simulation tasks. For example, a sequence of calculations may be performed with the use of a simple 'for-loop' construction. Calculations of energy, forces, stresses and other quantities are performed through interfaces to many external electronic structure codes or force fields using a uniform interface. On top of this calculator interface, ASE provides modules for performing many standard simulation tasks such as structure optimization, molecular dynamics, handling of constraints and performing nudged elastic band calculations.

A DFT study of oxygen dissociation on platinum based nanoparticles.

Density functional theory calculations are performed on 38 and 79 metal atom truncated octahedron clusters to study oxygen dissociation as a model for the initial stage of the oxygen reduction reaction. Pure platinum and alloyed platinum-titanium core-shell systems are investigated. It is found that barrierless oxygen dissociation occurs on the (111) facet of the pure platinum clusters. A barrier of ~0.3 eV is observed on the (100) facet. For the alloyed cluster, dissociation barriers are found on both facets, typically ~0.6 eV. The differences between the two systems are attributed to the ability of oxygen to distort the (111) surface of the pure platinum clusters. We show that flexibility of the platinum shell is crucial in promotion of fast oxygen dissociation. However, the titanium core stabilises the platinum shell upon alloying, resulting in a less easily distortable surface. Therefore, whilst an alloyed platinum-titanium electrocatalyst has certain advantages over the pure platinum electrocatalyst, we suggest alloying with a more weakly interacting metal will be beneficial for facilitating oxygen dissociation.