Fat bloom in chocolate is a major cause of concern for any chocolate manufacturer,
however, the mechanism for this phenomenon is often poorly understood, or there
could be various mechanisms acting simultaneously. It is known that fat bloom occurs
when fat crystals crystallise in an uncontrolled fashion, which can occur when
triacylglycerides (TAGs) which are present in the chocolate fat phase interact with
other, different, TAGs which migrate from the chocolate filling or inclusions, such as
nuts, to the chocolate surface. Understanding how these TAGs interact would thus
help a chocolate manufacturer design better fat bloom-stable products. One approach
to this would be to simulate such interactions, which is the subject of this thesis.
The main results in this thesis centre around the development and application of the
COarse-Grained Interchangeable Triacylglyceride-Optimised (COGITO) molecular
dynamics (MD) force field (FF) for the simulation of TAGs. COGITO was developed
using both a bottom-up approach (building on atomistic simulations described in the
first part of this work) and a top-down approach (parameterising the FF against
empirical measurements), including the use of machine learning techniques to speed
up the parameter optimisation process. The FF has been shown to be suitable for both
saturated and unsaturated TAGs, including TAGs which were not used during the
parameterisation process. This included the simulation of macroscopic properties
such as crystalline and liquid densities, and thermodynamic properties such as the
melting point of a range of TAG crystals, as pure or binary systems. New,
computationally rapid, metrics have also been developed to determine the
crystallinity, or not, of such molecules, as well as to distinguish between polymorphs,
all of which can be used to monitor the evolution of MD simulations easily.
COGITO has been shown to be a suitable FF for the efficient simulations of TAGs. This
can now be used to simulate the interactions of different TAGs, originating from the
chocolate fat phase and any fillings, to determine the stability, or otherwise, of such fat
systems.
| Date of Award | 25 Aug 2023 |
|---|
| Original language | English |
|---|
| Awarding Institution | - University Of Strathclyde
|
|---|
| Sponsors | University of Strathclyde |
|---|
| Supervisor | Tell Tuttle (Supervisor) & Nico Bruns (Supervisor) |
|---|