Dissipative Particle Dynamics Simulations to Model Amorphous Nanoparticles

Open Access
- Author:
- Weitzner, Stephen E
- Area of Honors:
- Materials Science and Engineering
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Coray M. Colina, Thesis Supervisor
Dr. Robert Allen Kimel, Thesis Honors Advisor - Keywords:
- dissipative particle dynamics
DPD
calcium phosphosilicate
nanoparticle
calcium
phosphate - Abstract:
- Targeted nanodelivery vehicles have the potential to revolutionize oncology through the simultaneous treatment and imaging of cancer. The success associated with many of these delivery systems is the small scale of the vehicles, the prevention of chemical degradation by encapsulation, and the targeted delivery of chemical agents to cancerous cells. A novel delivery system currently in development consists of ~20 nm amorphous calcium phosphosilicate nanoparticles (CPSNPs) which possess the additional characteristics of being inherently non-toxic and bio-compatible. The future development of delivery systems such as this depend upon a deep understanding of the material's chemistry and its potential interactions in the human body. The effects of varying the silicate content in CPSNPs has yet to be stringently studied, but controlling the CPSNPs' compositions may open the door for higher molecular encapsulation efficiencies or even a broader range of possible cargo molecules. In this thesis, a model for simulating the solubility of amorphous calcium phosphate (ACP) and CPSNPs of variable silicate content was developed. ACP and CPSNPs are represented as consisting of ~1 nm randomly packed ionic clusters, known as Posner's clusters (PCs), where silicate substitutions occur at phosphate positions in the PCs. Dissolution was modeled as the disassociation and rearrangement of PCs. Due to the large number of particles in the model system and the desire to reproduce hydrodynamic effects, dissipative particle dynamics (DPD) was employed. Different sets of interaction parameters which scale the conservative force between neighboring beads were used to determine how the solubility of the simulated system can be tuned. Philic interactions between particle beads and fluid beads resulted in dissolution. Employing a philic interaction between silicate-modified beads and particle beads provided a driving force for PCs to remain associated with the initial particle. Solubility was calculated in the simulation by comparing distances between disassociated PCs in the simulation box. Disassociated PCs were identified as PCs with only fluid bead neighbors within 1 force cutoff distance. For the ACP model, scaling the simulation box size with constant particle concentration produced similar solubilities, indicating the simulation appears to be free of box size artifacts. Changing the particle concentration by increasing the box size and maintaining the same particle size increased the solubility. For the CPSNP model, three different sets of interaction parameters with philic interactions were used to demonstrate varying degrees of solubility in the CPSNP model. Each set showed that simulations with 19% silicate substitutions were able to almost fully suppress the particle's solubility. Although accurate solubility data for ACP and CPSNPs are not yet available, it was concluded that the solubility of the proposed model can be adjusted by changing the concentration of the nanoparticle, tuning the interaction parameters, or varying the silicate content to match experimental data, when available. The solubility calculation results were found to be inadequate as stand alone criteria, and thus simulation snapshots were use to assist in explaining system behavior. The calculation of radial distribution function or radius of gyration could provide a more quantitative description of the system and should be explored in the future. Furthermore, simulations must be run longer to verify that equilibrium behavior is being observed, as well as performing multiple runs to collect statistics. While more work on both the experimental and computational sides must be done, the development and subsequent refinement of such a model could greatly enhance the study of CPSNPs' material chemistry. The ultimate goal of this work is to foster the development of a predictive tool for studying the encapsulation efficiencies of CPSNPs.