Diffusivities of bioactive compounds in compressed liquids: molecular dynamics simulation

• Autor(a) principal: Iglésias, João António Pereira
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/29399
• topic: Diffusion coefficient; Molecular dynamics; Transport properties; Quercetin; Ethanol; Ethyl acetate

Quercetin, a flavonoid that can be extracted from various plant sources, exhibit interesting bioactivity due to relevant antioxidant or anti-carcinogenic properties. One way of extracting this flavonoid is by solid-liquid extraction using, for example, green solvents like ethanol (EtOH) or ethyl acetate (EtOAc), which are well accepted in the food industry. Diffusivity, D12, is an important property in solid-liquid extraction, since this separation is frequently limited by mass transfer kinetics, which requires the knowledge of D12 for the accurate design and optimization of that unitary operation. The diffusivities, D12, of quercetin in ethyl acetate and ethanol were measured by the chromatographic peak broadening (CPB) method in the temperature range 30-60 °C and pressure range 1-150 bar. The diffusivities in ethanol were measured in the same laboratory by another researcher. The D12 values ranged from 3.985×10-6 to 7.826×10-6 cm2 s-1, in the case of ethanol, and 1.018×10-5 to 1.628×10-5 cm2 s-1 for ethyl acetate. The obtained D12 data followed the expected trends with temperature and pressure, namely, positive and negative derivatives, being the influence of temperature much more significant. In parallel, classical molecular dynamics (MD) simulations were performed using the GROMACS software package to estimate the diffusion coefficient in order to assess the possibility of using this computational technique to generate diffusivities for distinct pressure and temperature conditions. Different parameters sets were adopted to carry out simulations in NVT ensemble, such as the cut-off radius for short-range interactions, number of solvent and solute molecules, and simulation duration, with the objective to verify their influence on the quality of D12 estimates. The optimization of the parameters used in the MD simulations led to D12 values in good agreement with the experimental data for ethanol at 1 bar, with relative deviations less than 6.54 %. It was also shown that it is possible to obtain reliable results at higher pressures after introducing a multiplicative factor on the atoms charges of ethanol. In the case of ethyl acetate, the error at 30 °C and 1 bar was −22.51 %. Since the MD self-diffusivities of ethyl acetate also differ significantly from the experimental data, it is suggested in this work to optimize the force field parameters used to model this solvent. The agreement found between experimental and MD quercetin diffusivities in ethanol demonstrates that it is possible to obtain reliable D12 values by classical MD simulations. Further studies are suggested on the influence of different functional groups and structure of other flavonoids on D12, with a structural analysis using the radial distribution and spatial distribution function.