Crystal Aggregation and Computer Modelling

Lead Research Organisation: University College London
Department Name: Chemistry


The design of an efficient crystallisation process for the production of pharmaceuticals, catalysts, fine chemicals and other materials of industrial importance is dependent on a number of factors, including internal properties of the product itself such as molecular structure, preferred crystal shape and size and cohesion energy, and external factors, such as fluid flow and drag. Although all of these factors affect the material's aggregation behaviour, the behaviour of the crystallites upon collision is usually described by a single expression, which does not take into account material-specific properties, or indeed separate intrinsic properties of the crystal from topological considerations, such as the relative geometries of the colliding particles. In addition, no consideration is given either to the colliding particles' shapes or size distribution, or the fact that the forces acting upon them may not just be normal to the point of impact, even though shear forces are likely to be very important when two particles collide. In this project, we propose to carry out a comprehensive study of the collision and aggregation behaviour of three different types of material, varying from a purely inorganic material (calcium carbonate) to a purely organic solid (adipic acid). We will use a combination of computational chemistry methods on the one hand and experimental chemical process engineering techniques on the other, to investigate the shapes of the crystallites in solution, the impact geometries of the colliding particles, the chemical bonding between the collided particles, as well as the shear and tensile forces required to separate the particles again after collision. In addition, we will investigate the precipitation of new material at the point of collision, leading to aggregation of the particles, where the collision point may indeed be a 1-D point, a 2-D line or a 3-D planar area. Once new material has grown at the join, we can calculate its resistance against fracture. The outcome of this project will be an in-depth understanding at the atomic level of the chemical and physical processes occurring upon collision of two nano-crystallites in solution. In addition, the results of the project will enable us to formulate general mathematical descriptions of the aggregation behaviour of a number of representative materials, which will include both intrinsic, material-dependent properties and external factors, including solvent effects (shear forces acting on the particles through water drag) and collision geometries. By combining these two approaches we aim to develop a quantitative kinetic description capable of being used with CFD to predict behaviour in stirred crystallisers.


10 25 50

publication icon
Di Tommaso D (2014) Modelling the effects of salt solutions on the hydration of calcium ions. in Physical chemistry chemical physics : PCCP

Description The computational part of this collaborative project with the experimental group in Sheffield has concentrated on two materials: calcite and calcium oxalate monohydrate (COM). Calcite was chosen, because extensive experimental data are available in the literature on its morphology and growth behaviour, and COM, because this would allow direct comparison with the experimental research programme.
We started with a thorough Density Functional Theory study of the initial nucleation processes of calcium carbonate, which information is impossible to obtain experimentally, followed by classical simulations of the calcite material, for which computational models were available, which had been tested and applied successfully in a number of studies. As impurities could have important implications on crystal growth an aggregation, we have first developed and applied a new statistical mechanics approach to investigate the thermodynamics of the uptake of cadmium and manganese in calcite and strontium and magnesium in aragonite (another important calcium carbonate mineral). This work has led to 3 published journal articles, with another paper submitted.
The next part of the project concerned the aggregation of two calcite nano-particles, where we have calculated the interaction energies between the particles, not only taking into account the interaction between different surfaces seen in the crystal morphology (the {10.4} and two {0001} planes), but also investigating surface defects such as surface steps and the incidence of a particle's edge or corner site with another particle's surface. It was found that when the nano-particles approach through solution (water), they have to overcome a series of energy barriers, where the lowest energy pathway requires lateral relative movement of the particles. The calculated interaction energies, together with the experimental results, can be used directly to develop and refine quantitative crystal growth models. A publication on this part of the project is in preparation.
The final part of the computational work programme was the development of molecular-level computational models for calcium oxalate monohydrate. Calcium oxalate exists in a number of different hydration levels: the monohydrate (COM), dihydrate (COD) and anhydrous material. Although COM was investigated by the experimental partners on this project, we have fitted our models to the anhydrous and COD materials as well, which ensures that the model is fully transferable across a range of hydration states. As little experimental data beyond the structural properties were available in the literature, we first carried out a Density Functional Theory study of the three phases to obtain ancillary data, such as elastic properties, which were then used for the derivation of an interatomic potential forcefield. This forcefield for COM reproduced accurately the experimental structural parameters and elastic data of the bulk material from the DFT calculations, and was subsequently used to model the surfaces of COM and calculate the crystal morphology, which is based on the relative stabilities of the different surfaces: only stable surfaces are expressed in the surface morphology and are therefore available to interact in the aggregation process. Using our new computational model, we correctly predicted the relative surface stabilities, leading to a nano-particle shape in close agreement with crystals grown experimentally. This work has recently been accepted for publication in the new journal RSC Advances.
COM is the major component of urinary kidney stones. Although there is an extensive experimental literature on potential growth inhibitors, this is inconclusive, and little computational work has been carried out. We will continue to use our new computational model for COM to investigate and identify promising growth modifiers, thereby ensuring that the impact of this project is extended even further than was originally envisaged.
A comprehensive final report was submitted to the EPSRC upon completion of the grant.
Exploitation Route Peer-reviewed publications in scientific journals

Conference presentations
Sectors Chemicals,Pharmaceuticals and Medical Biotechnology,Other

Description Published in peer-reviewed scientific journals, presented at (inter)national conferences
First Year Of Impact 2011
Sector Chemicals,Other
Impact Types Economic