Computational Prediction of Gel Properties (Reference Berry LRC116)

This is an excellent opportunity for a well-motivated student to participate in a project funded by the Department of Chemistry (https://www.liverpool.ac.uk/chemistry/) and the Leverhulme Research Centre for Functional Material Design (https://www.liverpool.ac.uk/leverhulme-research-centre/) at the University of Liverpool. The student will be based at the University of Liverpool, with a supervisory team from the department of Chemistry (Dr N. Berry) and will work closely with colleagues in the School of Chemistry at the University of Glasgow (Prof. D. Adams). The studentship is fully funded for a period of 42 months starting in October 2018.

Low molecular weight gels are currently being investigated for a range of interesting applications from optoelectronics to tissue engineering. The gels are a result of the self-assembly of small molecules into long fibres, which then entangle to form the gel matrix. Whilst these gels are extremely interesting, a major limitation is that it is enormously difficult to design the materials.
First, there are very limited design rules for molecules that can self-assemble to form gels. This lack of knowledge means that most gelators are found by accident and new analogues prepared by synthetic iteration, not always with great success. To tackle this, we recently published a computational descriptor-based approach to predicting the ability of molecule to form a gel. With the limitations of a restricted phase space, this was a very successful strategy and we are now using these models to guide our synthetic targets.
Second, not all gels have the same properties. The materials have widely differing stiffnesses, ability to recover after damage, stretchability etc. Assuming that one has found a molecule that can form a gel, it is still extremely difficult to predict in advance what properties the gels will have. Hence, to prepare gels for a specific application, one needs to find both a molecule that can form a gel, and form a gel with the desired properties. We have recently shown that the gel properties can be varied by changing how the gels are prepared. It is therefore perhaps more tractable to find a robust, effective gelator and then permutate how the gels are formed. What is really needed now is therefore a predictive tool for what the gel properties will be. This is the aim of this project.

The predictive tool will be realized by developing robust quantitative structure–property relationships (QSPR) models which link measured properties to compound chemical structure. QSPR is based on the principle that experimentally measured endpoints are a function of molecular properties, encoded numerically as descriptors, which capture the chemical information of the molecule for computational processes. Statistical and machine learning methods (e.g. random forests), will be deployed to link these descriptors to the measured endpoint, i.e. gelation, stiffness, etc. QSPR models will shed light on the key molecular characteristics that are linked to the gelation properties of a compound and also, enable rapid computational screening of libraries to identify candidates that are likely to possess the desired gelation properties.

Qualifications: Candidates should have a good (at least 2.1) degree in chemistry, computer science, materials science or related subject.

Please apply by completing the online postgraduate research application form here: https://www.liverpool.ac.uk/study/postgraduate-taught/applying/online/
Please ensure you quote the following reference on your application: Computational Prediction of Gel Properties (Reference Berry LRC116)