| Carbon nanotubes (CNTs), featuring unique structure and amazing properties, have been considered in numerous studies owing to their diverse potential applications. Recently, many studies suggest that CNTs could serve as nanoscale pipes to deliver fluids and molecular species. Due to the deep potential well inside the CNT, the nanotube is more likely to encapsulate molecules, which makes it a unique system for studying nanofluidics and molecular transport. In particular, its ability to serve as biocompatible transporters in medicine and drug delivery has received more and more attention. In addition, both covalent and noncovalent functionalizations have opened up opportunities for medical diagnostic purposes and drug delivery applications. So far, many biomolecules and drugs have been found to be encapsulated into the inner space of CNTs spontaneously from both computational and experimental shudies, in case that these guest molecules have suitable size. In spite of these fascinating observations on various molecules encapsulated into CNTs, the dynamic mechanisms of these encapsulation processes at the molecular level remain obscure, which limits the biological and biomedical applications of CNTs. To this end, the molecular simulation helps ones to better understand the dynamic mechanism in the bio-nano-systems, and establish new concepts for controlling/tuning the performance of such systems to facilitate the design and optimization of CNT-based functional nanoscale devices.In this thesis molecular dynamics (MD) simulation and steered MD simulation were performed to investigate the diffusion of small molecules and the encapsulation of proteins/peptides in CNTs, as well as the the atomic details of the interactions taking place at the molecular level, and the dynamic mechanisms of the biomolecules-CNT systems.1. Carbon nanotube structures provide a means to systematically investigate the diffusion mechanism of simple Lennard-Jones fluid mixtures under confinement. We have used a stochastic thermostat within these carbon nanotube structures to provide models of tubes with effective corrugated walls, to model the random walk of particles diffusing in three-dimensions (Fickian diffusion) and one-dimension (single-file diffusion) within cylindrical pores. The random walk within cylindrical pores has been shown to be needed to give rise to the square root of time dependence of the mean-squared displacement (MSD) for single-file diffusion. Within this framework we showed that for Lennard-Jones mixtures, transition diameters of carbon nanotubes (CNTs) existed for each component, the crossover from single-file to Fickain diffusion occurring at D～2.6σ-2.8σ, measured geometrically with respect to the largest component in the mixture. For Ar and Kr the transition diameter CNTs were the (7,7) CNT which gave rise to single-file diffusion, where the mean-squared displacement was proportional to the square root of time and the (8,8) CNT which gave rise to Fickian diffusion, where the mean-squared displacement was proportional to time. For Ne, the (6,6) CNT gave rise to single-file diffusion while the (7,7) and larger diameter CNTs gave rise to Fickian diffusion. For Xe, the (6,6),(7,7), and (8,8) CNTs gave rise to single file diffusion while CNTs larger in diameter gave rise to Fickian diffusion. The diffusion mechanism (single-file or Fickian) did not depend on the composition of the mixture.2. MD simulation suggests that in small (6,6) CNT, hydrogen bonds tend to aggregate water into a molecule chain and lead to collective behavior of rapid normal diffusion drift. On the contrary, in larger nanotubes, hydrogen bonding network doesn’t change the water diffusion mechanism and allows the water to form regional concentrated clusters. Several factors have great impact on water diffusion including diameter, temperature and loading. As the diameter increases, diffusion rate becomes faster. Confinement is able to stabilize the hydrogen bonds, allowing the lifetime one to three orders longer than that of the bulk water. As temperature decreases, the diffusion in (7,7) CNTs jumps from Fickian to single-file and the average lifetime of H-bond in (7,7) CNT increases. Moreover, the diffusion rate is highly density-dependent. It is found that water in extremely low density undergoes slow self-diffusion motion because water clusters provide limited space for molecule to diffuse.3. MD simulations and SMD simulations demonstrate the ability of carbon nanotubes (CNTs) to encapsulate peptides or proteins. Protein/peptide tends to be trapped inside the CNT and oscillate around the center of the tubes. The van der Waals attractive force between the CNT and the peptide is the driving force for the encapsulation process. The diameter and length of the CNTs play important roles during the encapsulation process. Longer nanotubes provide a broader area to trap the peptide while the CNTs with smaller diameters are able to encapsulate the peptide with a deeper interaction energy well. It will be important to balance the diameter and length of CNTs if they are to be used as drug delivery devices. This research can provide new insights into the design and control of nanoscale pharmaceutical delivery processes.4. The delivery of peptide drug is mostly associated with potential wells. Because peptide is more likely to be trapped inside carbon nanotubes, it is of great necessity to apply external force to release the peptide from the carbon nanotube. Thus the electric field has been applied to simulate the peptide released from the carbon nanotube. It has been found that with the increase of diameter, peptide is more likely to be released from the carbon nanotube. When the length of the carbon nanotube becomes greater, the peptide, due to the long-time transportation, is more subjected to conformational change under the electric field, which allows the peptide absorbed inside the carbon naotube inner wall. We could change the electric field to control the process of peptide released from the carbon nanotube. |
PDF Full Text Request