Multiscale simulations of liquid-liquid interfaces: From lipid membrane to graphene
by Alireza Soleimani
Date of Examination:2023-09-19
Date of issue:2023-11-06
Advisor:Dr. Herre Jelger Risselada
Referee:Dr. Herre Jelger Risselada
Referee:Prof. Dr. Marcus Müller
Referee:Prof. Dr. Matthias Krüger
Referee:Prof. Dr. Peter Sollich
Referee:Prof. Dr. Steffen Schumann
Referee:Dr. Richard L. C. Vink
Files in this item
This file will be freely accessible after 2024-09-18.
EnglishThere is an increasing demand to enhance the development of coarse-grained models in order to facilitate highly efficient simulations of realistic biological systems. In our endeavor to tackle this challenge, in chapter 2, taking advantage of evolutionary algorithms, we constructed a highly coarse-grained membrane model that not only exhibits the realistic behavior of lipid membranes such as self-assembly into bilayers, vesicle formation, membrane fusion, and the formation of the gel phase, but also accurately reproduces experimentally observed structural and thermodynamic properties of membranes. Hence, the developed model allows for a reduction in degrees of freedom by coarse-graining the positions and orientations of individual lipid molecules into effective particles, all while preserving the essential behavior of the system. Additionally, we also highlight that the here-developed coarse-grained model is compatible with the Martini coarse-grained model for biomolecular simulations, thereby providing a pathway for simulating mixed-resolution complex systems. However, it is essential to not assume that coarse-grained models accurately represent collective phenomena. Therefore, it is highly important to carefully compare the results obtained from coarse-grained models with the behaviors observed in various experimental scenarios and fine-grained simulations. Critical questions that must be addressed when utilizing (ultra) coarse-grained models are: what factors can restrict the predictive power of coarse-grained models; which degrees of freedom and interactions should be preserved at the coarse-grained scale in order to capture the fundamental physics of the system; what is the trade-off between accelerating computational speed and gaining access to larger length and timescales without compromising the level of chemical detail. Consequently, in order to shed light on the aforementioned questions, in chapter 3, we carried out an investigation to explore the surfactant behavior of a pristine graphene sheet at the liquid-liquid, i.e., octanol-water interface using both atomistic and coarse-grained molecular dynamics simulations. In fact, the primary purpose of such an investigation was to assess the universality of the observed phenomena and results across different levels of resolution. Significantly, our simulation results revealed that not all the coarse-grained models can successfully restore the unexpected surfactant behavior of graphene observed in atomistic simulations due to lack of the degrees of freedom required to capture the underlying physics of the problem. Specifically, the atomistically observed surfactant behavior did not appear when using a coarse-grained two-site octanol model, whereas the effect was recovered when employing coarse-grained models of longer alcohols, such as dodecan-1-ol (3-site model) and hexadecan-1-ol (4-site model). This once again highlights the importance of identifying the key features and properties of the system that are relevant to the phenomena under investigation in order to achieve a successful coarse-grained description.
Keywords: entropic surfactant; pristine graphene; interfacial tension; structure formation; directed self-assembly; Cell membrane; evolutionary algorithms; Membrane properties