| 1 |
\chapter{\label{chapt:conclusion}CONCLUSION} |
| 2 |
|
| 3 |
The primary goal of this research has been to develop and apply |
| 4 |
computational methods to study the structure and dynamics of soft |
| 5 |
condensed matters. As the underlying physical law behind molecular |
| 6 |
modeling of soft condensed matter, statistical mechanical principles |
| 7 |
used in this dissertation are briefly reviewed in |
| 8 |
Chapter.~\ref{chapt:introduction}. Following that, an introduction to |
| 9 |
molecular simulation techniques including newtonian dynamics and |
| 10 |
Langevin dynamics was provided. Even though the motions of soft |
| 11 |
condensed systems are characterized by different ODEs between |
| 12 |
Newtonian dynamics and Langevin dynamics, they all preserve some |
| 13 |
underlying geometric properties. These properties are built into a |
| 14 |
geometric integration method, which gives the method remarkable |
| 15 |
performance and stability, especially during long simulations. Thus, |
| 16 |
the theory of geometric integration and the methods to construct |
| 17 |
symplectic integrators are also covered in |
| 18 |
~\ref{chapt:introduction}, as well as the mathematics behind |
| 19 |
the elegant symplectic integration scheme involving rigid body |
| 20 |
dynamics. |
| 21 |
|
| 22 |
In Chapter.~\ref{chapt:methodology}, the basic methods used in this |
| 23 |
work were discussed. An overview of the DLM method was given showing |
| 24 |
that DLM distinguished itself by its accuracy and efficiency during |
| 25 |
long time simulation. Following this, the DLM method was extended to |
| 26 |
produce canonical ensemble and isobaric-isothermal ensemble, as well |
| 27 |
as special ensembles like $NPAT$ ensemble and $NP\gamma T$ ensemble |
| 28 |
to alleviate the anisotropic effect of biological membrane systems. |
| 29 |
In order to study slow transport in membrane systems, a new method |
| 30 |
to study diffusion by measuring the constraint force was proposed |
| 31 |
and verified. |
| 32 |
|
| 33 |
Chapter.~\ref{chapt:lipid} provided a general background to transport |
| 34 |
phenomena in biological membranes. Atomistic simulations were |
| 35 |
applied to study the headgroup solvation for different |
| 36 |
phospholipids. A simple but relatively accurate and efficient |
| 37 |
coarse-grained model was developed to capture essential features of |
| 38 |
the headgroup-solvent interactions. It was then shown that the |
| 39 |
structural properties of the simulated membrane bilayers agreed well |
| 40 |
with experimental data. Further studies combining an external force |
| 41 |
dragging method and z-constraint method may provide insights into |
| 42 |
understanding of transport in large scale biological systems. |
| 43 |
|
| 44 |
The current status of experimental and theoretical approaches to |
| 45 |
study phase transition in banana-shaped liquid crystal system was |
| 46 |
first reviewed in Chapter.~\ref{chapt:liquidcrystal}. A new rigid body |
| 47 |
model consisting of three identical Gay-Berne particles was then |
| 48 |
proposed to represent the banana shaped liquid crystal. Starting |
| 49 |
from an isotropic configuration, we successfully explored an unique |
| 50 |
chevron structure. Calculations from various order parameters and |
| 51 |
correlation functions also confirmed this discovery. |
| 52 |
|
| 53 |
Lastly, Chapter.~\ref{chapt:langevin} summarized the applications of |
| 54 |
Langevin dynamics and the development of Brownian dynamics. By |
| 55 |
embedding hydrodynamic properties into the sophisticated rigid body |
| 56 |
dynamics algorithms, we developed a new Langevin dynamics for |
| 57 |
translation-rotation couplings systems. Molecular simulations with |
| 58 |
different viscosities demonstrated the temperature control ability |
| 59 |
of this new algorithm. It was also shown that the dynamics was |
| 60 |
preserved using this implicit solvent model in studying mixed |
| 61 |
systems of banana shaped molecules and pentane molecules. |
| 62 |
|
| 63 |
Overall, this work has shown the successful application of |
| 64 |
statistical mechanics to study structure, dynamics and phase |
| 65 |
behavior of soft condensed materials. Beginning by developing coarse |
| 66 |
grained models that could reproduce experimental observations, we |
| 67 |
have extended molecular simulations to study self-assembly in soft |
| 68 |
condensed systems. Finally, we have developed a new Langevin |
| 69 |
dynamics algorithm for arbitrary rigid particles which can be used |
| 70 |
as an implicit solvent model to explore slow processes in soft |
| 71 |
condensed systems. |
