University of Cambridge
An Experimental Approach to the Development and Validation of Group Interaction Modelling
Abstract
dc:description.abstractGroup Interaction Modelling (GIM) is a modelling framework for predicting properties of polymers from their chemical structure. It assumes the energy balance of the intermolecular interactions between characteristic groups and external energy fields, and energy dissipation mechanisms are the origin of all bulk properties of polymers. The framework was proposed by David Porter, while he was working in industry and is very much oriented towards thermal and mechanical constitutive properties. It has been tested against available experimental data for many common polymers, mainly thermoplastics such as polyethylene, polystyrene, polycarbonate etc., over a narrow range of conditions and measured properties, related to their intended industrial applications. Hitherto, the predictive power of Group Interaction Modelling has therefore been hindered by its semi-empirical components, reliance on limited data, and gaps in physical detail. Developing this model further, eliminating its shortcomings, and improving the description of fundamental molecular mechanisms, would enable a more complete picture of polymer behaviour. This would enable reliable structure-property relations, even for emerging polymer technology, where experimental data is not available. With this goal in mind, this thesis takes the approach of carefully designed experiments to measure thermomechanical properties of a single-source polymer across a wide range of temperatures and rates. The model framework is thus systematically tested and informed to improve its predictive power, with physically derived additional considerations. The author’s choice of the single-source material of study is Bisphenol A Polycarbonate. After an introduction to Group Interaction Modelling, this thesis presents a series of experiments that have provided a rigorous test of its performance for the single-source polycarbonate. Thermal transport properties were measured using a Transient Plane Source method. Differential Scanning Calorimetry was also performed for comparison. Thermal expansion at several heating and cooling rates was measured using laser interferometry. Dynamic Mechanical Analysis (DMA) was performed to provide a deeper understanding of the timescales of an oscillatory test that are important in dictating the observed glass transition temperature. A Taylor impact-like test was performed to measure the high-rate mechanical impedance. A novel method of measuring energy loss in a split Hopkinson pressure bar configuration was developed and used to construct a loss tangent curve, analogous to that obtained from DMA measurements, but at higher rates of deformation. The speed of sound and attenuation of ultrasonic longitudinal waves were measured with temperature through the glass transition. The same properties of both longitudinal and shear waves were measured varying the frequency, such that the secondary transition was probed at room temperature. Model predictions were compared with the experimental measurements and areas of discrepancy and physical reasoning led to the following improvements to the model. The temperature dependence of the polymer chain stiffness was incorporated in all property predictions. The set of input parameters used to describe the secondary transition of polycarbonate was updated by the author. Comprehensive calculation methods were constructed for predicting enthalpy overshoot peaks in heat capacity measurements performed on heating, and for thermal conductivity predictions. A treatment of ageing effects by adjusting the volumetric input parameter was proposed and the effect on predicted quantities demonstrated. Considerable improvement was achieved for polycarbonate with the author’s contributions, compared to the original expressions by Porter. The research on polycarbonate is brought together by comparing the observed glass and beta transition temperatures, on a common timescale basis defined by the author across all experiments, spanning 13 orders of magnitude of rate. The common timescale or, equivalently, rate was identified as the characteristic timescale of each experiment that dictates the relaxation conditions of the molecules. This characteristic rate was defined for thermal experiments, relating to the rate of thermal expansion, and for mechanical experiments relating to frequency or strain rate, depending on the type of stimulus. The phenomenon of time-temperature superposition was verified across all experiments and the power of the model to predict the expected transition temperature, when given the rate in the correct time basis was demonstrated. To demonstrate the applicability of Group Interaction Modelling for new, emerging materials, the modelling tool was used for a more complex bio-derived material. The glass transitions of this material, with different amounts of plasticiser were successfully predicted by the model. The sizes of loss peaks and the storage modulus, when viewed in the context of the model, provide microscale information about the material. Overall, this research has contributed to a significantly improved tool for predicting polymer properties from their chemical structure, with the improvements backed by physical concepts and verified with experimental measurements. GIM, together with the author’s contributions, has impressive predictive power, both for well-known and novel polymers. Finally, the author discusses suggestions for expansion of this research, for furthering understanding of fundamental polymer science, and in light of industrial needs, the requirement to design environmentally sustainable high-performance materials and processes for the future.
Degree
thesis:*- Name dc:type.qualificationname
- Doctor of Philosophy (PhD)
- Level dc:type.qualificationlevel
- Doctoral
- Grantor dc:publisher.institution
- University of Cambridge
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Constantinou, Malvina
- Advisor dc:contributor.advisor
-
- Williamson, David
Subjects
dc:subject × 4Rights
dc:rightsIdentifiers
dc:identifier.*- DOI dc:identifier.doi
- https://doi.org/10.17863/CAM.115040
- OAI identifier oai:identifier
- oai:www.repository.cam.ac.uk:1810/378761