In-situ Characterisation of the Crystal Growth of Organic Materials through Microscopic Imaging, Image Processing and Machine Learning

Summary

The crystal growth of organic materials is of significant importance within the speciality and fine chemical industries. This reflects its utility in materials purification and its use in preparing a wide range of compounds which have the well-defined crystal size, shape and polymorphic form needed for optimal product performance. The latter is important e.g. in ensuring the reproducible dissolution and stability behaviour needed to maintain the safety and efficacy of ingredients within formulated products.

The inherent molecular-scale complexity of organic materials can directly impact on the physical chemical properties of crystals, notably their crystallisation in low symmetry crystallographic structures which have anisotropic external crystal morphologies and surface properties. Changes to, or variability in, these properties can affect their performance, e.g. their purity, bioavailability, powder flow, stability and manufacturability. Current crystal size measurements can be over-simplistic in terms of shape characterisation being focussed mostly on spherical particles. Such methods do not reflect the facetted (polyhedral) crystal morphologies common found within organic solids where different crystal habit faces can exhibit different surface chemistry which can expose different intermolecular binding interactions within the chemical processing environment. Currently, there is a critical capability gap in terms of being able to relate the molecular structure of a material to its performance in its crystalline particulate form. This knowledge gap has led to increasing interest in fusing in-situ experimental crystallisation studies with a knowledge of its core molecular structure and its simulated surface properties based upon crystallographic data linked to artificial intelligence (AI) and machine learning approaches.

This project addresses the above need by applying digital AI-enabled technology to develop morphologically-based shape descriptors with targeted utility for the precise 3D characterisation of crystalline particulates in-situ. Overall, the project aims to enable the capability to design organic crystalline ingredients and the products resulting from their formulation to a much tighter specification notably higher quality, more consistency and less variability. The proposed project is directly associated with an EPSRC research project (https://eps.leeds.ac.uk/research-project/1/faculty-of-engineering-and-physical-sciences/4417/advanced-crystal-shape-descriptors-for-precision-particulate-design-characterisation-and-processing-shape4ppd).

The aim of this project is to support the production of precision crystalline particulate materials through actual characterisation of 3D crystal shape and size. To do this machine learning will be applied to map the images from in-situ microscopy to a description of 3D crystal shape and functional properties. This will help enable the design and manufacture of crystalline fine chemicals to a much tighter specification for particle size and shape than is currently feasible, resulting in more consistency, less variability in physical and chemical properties and concomitantly higher quality.

This project will suit a self-motivated graduate student with a chemical process engineering or physical sciences background. The student will carry out interdisciplinary research at the University of Leeds which has an international reputation for excellence in teaching and research. The student will seek to integrate crystallisation technology with AI/machine learning approaches, under the supervision of the experts from these two areas, with a focus on the 3D crystal characterisation of crystallisation processes. The student will achieve the project aims through various research tasks including:

1. Implementation of appropriate particle shape descriptors, using combined AI/machine learning and crystal morphological modelling technologies, that allow a seamless interface between laboratory measurement and digital particle shape models;
2. Integration of particle size/shape data from a (Keyence) microscope with the developed AI-enabled crystal morphological characterisation models for digital surface properties including facetted growth rates and mechanisms;
3. Application of the developed models for in-situ 3D crystal growth assessment to generate reliable growth rates and mechanisms;
4. Extending microscopy approach to on-line imaging tools for monitoring the dynamics of the growth of a population of crystals as observed during batch and continuous crystallisation processes, integrating this with process simulations notably through morphological population balance modelling

Overall, the outcomes of this project will help enable the design and control of more efficient and agile manufacturing processes for crystalline organic materials. This can be achieved by flagging potential manufacturing issues downstream, allowing for direct relationships between crystal shape and relevant surface properties. The work will hence have utility in terms of guiding process design to deliver crystals with improved size and shape to tighter specifications than currently is feasible, resulting in materials with more amenable properties for manufacturing and product use.

This project brings together internationally-recognised research groups in crystallisation science and engineering (Prof K J Roberts, Dr C Y Ma) and artificial intelligence (Prof D C Hogg) through integrating synergistic expertise in crystallisation, morphology and surface chemistry, molecular modelling and crystal characterisation (Roberts, Ma), computational fluid dynamics and process modelling (Ma, Roberts), imaging and image analysis (Hogg, Ma), and machine learning (Hogg). The project involves an intimate mixture of experimental (imaging and crystallisation) and computational (molecular modelling, image analysis, morphological population balance modelling, AI/machine learning (deep learning, CNN, GAN, database analysis) sciences.

The proposed project will include research training in crystallisation, crystal shape modelling, laboratory techniques, image processing, AI/machine learning. The student will benefit from access to a wide range of transferable skills training opportunities in the Leeds postgraduate research programme and via interactions with Shape4PPD project and other CDT research researchers (CP3 and M2P etc.). The research outcomes of this project will be disseminated via national/international conferences and journal publications providing excellent training opportunity for emerging industrial scientists in fusing analytics/simulation/machine learning as applied to industrial challenges.

Applicants to research degree programmes should normally have at least a first class or an upper second class British Bachelors Honours degree (or equivalent) in an appropriate discipline. The criteria for entry for some research degrees may be higher, for example, several faculties, also require a Masters degree. Applicants are advised to check with the relevant School prior to making an application. Applicants who are uncertain about the requirements for a particular research degree are advised to contact the School or Graduate School prior to making an application.

The minimum English language entry requirement for research postgraduate research study is an IELTS of 6.0 overall with at least 5.5 in each component (reading, writing, listening and speaking) or equivalent. The test must be dated within two years of the start date of the course in order to be valid. Some schools and faculties have a higher requirement.