Modelling realistic inorganic nanostructures: bridging the gap between theory and experiment
- Martijn A. Zwijnenburg (University College London, United Kingdom)
- Stefan T. Bromley (University of Barcelona , Spain)
The last twenty years have seen a colossal experimental effort to prepare and characterise ever smaller structures of inorganic materials down to as small as one nanometre in size. Such nanostructures have been demonstrated to often have widely different properties to that of the corresponding bulk material. As a result of these unique size-dependent characteristics inorganic nanostructures find application in numerous fields such as catalysis, solar cells, batteries and (bio)medical imaging. The same reduced size that makes these nanostructures interesting, however, also makes it difficult to understand the origins of their unique features fully by experiment alone. This knowledge gap is inherently linked to the difficulty of resolving the detailed atomic structures of nanostructured samples in experiment due to disorder and the associated broadening of diffraction and spectroscopic peaks.
Complementing experimental results with theoretical results is an attractive approach to closing the nanoscale knowledge gap. Here one would use computational chemistry calculations with realistic structural models of the nanostructure to gain atomistic insight into the fundamental processes underlying the size-dependency. Structural models would be obtained through global optimisation, where one aims to find the lowest energy structure for a given nanostructure, together with any information that can be obtained from experiment (e.g. composition, dimensionality, shape). Several groups have been developing such strategies in recent years using different global optimisation strategies (e.g. basin-hoping, genetic algorithms, simulated annealing, data mining) and have published “proof of principle” papers demonstrating that not only is it possible to reproduce experiment but also to successfully obtain microscopic insight unattainable by any other means. Most computational studies, however, at the moment still employ cuts from bulk structures or manually constructed clusters of unknown pedigree.
In the workshop we plan to bring together for the first time not only people working on global optimisation of nanostructures but also computational chemist/physicists interested in modelling their properties, and experimental chemist and physicists who prepare and characterise nanostructures. This mix of experience, as also reflected in objective 1 (see below), is aimed at promoting the developments to people outside the primary global optimisation community who can directly benefit from these approaches in their research. In the proposed workshop we will also for the first time discuss strategies for how to extend the existing global optimisation methods (previously developed for studying nanostructures in vacuo) to handle nanostructures prepared in the presence of solvents and ligands. We finally hope also to shed light on the underexplored field of structure prediction for structures that are extended in one (e.g. nanotubes) or two dimensions (e.g. thin films) and nanosized in the other dimensions.
As discussed above, complementing experimental observations with theoretical predictions is an attractive proposition for closing the knowledge gap that originates from the problems with fully characterising nanostructures by experiment alone. The first computational studies predicting the likely structure of nanostructures from the bottom up focussed primarily on Lennard-Jones clusters; a theoretician’s idealised yet surprisingly useful system. However, since then, the various global optimisation approaches (e.g. genetic algorithms, simulated annealing and basin-hopping) have been applied to a very diverse range of nanostructures with importance relevance to experimental nanoscience: e.g. those made out of (mixed) metals [1,2], metal oxides (SiO2 , ZnO , TiO2 [5,6], In2O3 , CeO2-x ) and metal chalcogenides (ZnS, ZnSe, CdS, CdSe [9-11]). At the same time top-down global optimisation methods were developed that use continuum surface thermodynamics and the surface energies of bulk crystal faces which predict the likely morphologies of crystalline nanostructures as a function of size, temperature and pressure . Results from both bottom up and top down studies have since been applied in a number of “proof of principle studies” to explain the optical [13,14], catalytic [6,8,15,16] and toxicity  properties of nanostructures.
Some very important issues remain, however, that we will address in the workshop. For example, the sheer majority of bottom-up studies focus on pure nanostructures in vacuo while in experiment the surface of such nanostructures is often decorated with organic ligands, hydroxyl groups and/or coordinating solvent molecules. Other issues we intend to address are predicting the likely atomic structures of nanotubes/nanowires  and thin films  and how to tackle ever larger nanostructures with bottom-up methods. Most bottom-up studies currently stop at nanostructures of 20-40 formula units, i.e. at the bottom of the experimental size range, because of technical problems related to the combinatorial explosion of possible low energy structural configurations. We will also discuss how results from top-down methods that do not yield explicit atomic structures, but rather relative expression of crystal faces, can be used to calculate nanostructure properties.
Beside the above method related points, we also intend to discuss interesting experimental questions that need addressing. One such question is the effect of ligands on order/disorder in nanostructures. For instance, for ZnS nanostructures it has been reported that introducing ligands (and subsequently varying their surface binding strength) appears to dramatically change the nanostructure crystallinity as probed by real-space pair distribution function . Similar effects have been also predicted for small gold clusters in the presence of carbon monoxide  and been observed in a combined experimental/computational study for thin films of Potassium Bromide after adsorption of specific organic molecules . Another interesting experimental question is for above what size the nanostructures can be expected to be crystalline and to have the bulk crystal structure. For several oxide and chalcogenide nanostructures (e.g. SiO2 [3,22], ZnS ) bottom-up approaches show that the lowest energy structures for small size nanostructures are non-crystalline and the transition size is currently not accurately known. Recent work also demonstrated that the experimental presence of broad diffraction peaks is not necessarily a sign of bulk order .
 Z. Li, N.P. Young, M. Di Vece, S. Palomba, R. E. Palmer, A.L. Bleloch, B.C. Curley, R.L. Johnston, J. Jiang, Jun Yuan, Nature, 451, 46, 2008.
 D.T. Tran, R.L. Johnston, Proc. Roy. Soc. A., 2131, 2004, 2011.
 S.T. Bromley, E. Flikkema, Phys. Rev. Lett., 95, 185505, 2005
 A.A. Al-Sunaidi, A.A. Sokol, C.R.A. Catlow, S.M. Woodley, J. Phys. Chem. C, 112, 18860, 2008.
 Z.W. Qu, G.J. Kroes, J. Phys. Chem. C, 111, 16808, 2008.
 M. Calatayud, L. Maldonado, C. Minot, J. Phys. Chem. C, 112, 16087, 2008.
 A.Walsh, S.M. Woodley, Phys. Chem. Chem. Phys., 12, 8446, 2010.
 A. Migani, K.M. Neyman, F. Illas, S.T. Bromley, J. Chem. Phys., 131, 064701, 2009.
 E. Spano, S. Hamad, C.R.A. Catlow, J. Phys. Chem. B, 107, 10337, 2003.
 A. Burnin, E. Sanville, J.J. BelBruno, J. Phys. Chem. A, 109, 5026, 2005.
 E. Sanville, A. Burnin, J.J. BelBruno, J. Phys. Chem. A, 110, 2378, 2006.
 A.S. Barnard, L.A. Curtiss, Nano Lett., 5, 1261, 2005.
 J.M. Matxain, L.A. Eriksson, J.M. Mercero, J.M. Ugalde, E. Spano, S. Hamad, C.R.A. Catlow, Nanotechnology, 17, 4100, 2006.
 M.A. Zwijnenburg, Nanoscale, accepted for publication, June 2011.
 A. Migani, G.N. Vayssilov, S.T. Bromley, F. Illas, K.M. Neyman, Chem. Commun., 46, 5936, 2010.
 A.S. Barnard, Energy & Environmental Sci., 4, 439, 2011.
 W. Sangthong, J. Limtrakul, F. Illas F, S.T. Bromley ST, Nanoscale, 2, 72, 2010.
 J. Goniakowski, C. Noguera, Phys. Rev. B, 79, 155433, 2009.
 B. Gilbert, F. Huang, Z. Lin, C. Goodell, H. Zhang, J.F. Banfield, Nano Lett., 6, 605, 2006.
 K.P. McKenna, A.L. Shluger, J. Phys. Chem. C, 111, 18848, 2007.
 T. Trevethan, B. Such, T. Glatzel, S. Kawai, A.L. Shluger, E. Meyer, P. de Mendoza, A.M. Echavarren, Small, 7, 1264, 2011.
 E. Flikkema, S.T. Bromley, Phys. Rev. B, 80, 035402, 2009.
 S. Hamad, C.R.A. Catlow, J. Cryst. Growth., 294, 2, 2006.