Research

Classical Molecular Dynamics Simulation

During my second year as a Master's student (2012-2013), I worked on a group project regarding molecular dynamics.

Here is the abstract for my group called Classical Molecular Dynamics Simulation:

This project focuses on classical molecular dynamics studied by numerical simulations which interact via a physics-based potential. These computations show how thermodynamic properties are derived from mechanics. In particular, the simulated molecular systems transition between phases depending on the prescribed density and temperature.

Stochastic Power Grid Modeling

During my first year as a Master's student (2011-2012), I worked on a group project regarding power grids.

Here is the abstract for my group called Stochastic Power Grid Modeling:

We seek to minimize the cost incurred during a blackout by optimizing the grid operator's response to different contingencies. To do this, a stochastic model for the dynamics of an electrical transmission network is simulated. Stochastic components include random line failure, overloaded lines, and restoration times. With this model, the cost associated with different combinations of response probabilities is estimated, and from there, the optimal strategy that would minimize the blackout cost is determined.

Monte Carlo Methods and Ising Spin Systems

Throughout the summer (2011) after my Chemical Engineering bachelors degree and before my Applied Mathematics masters degree, I worked on a research project under Professor Jon Machta regarding statistical physics called Monte Carlo Methods and Ising Spin Systems. Here is a brief summary of the work:

I explored annealing algorithms using C++ to find the energy minimum of rough free energy landscapes in spin glasses. A spin glass is a magnet with frustrated interactions because ferromagnetic and anti-ferromagnetic bonds are randomly distributed. In ferromagnetic materials, the spins of the outermost electron tend to align, while the opposite is true for anti-ferromagnetic bonds. It was found that an evolutionary annealing algorithm with elements from statistical physics was optimal to find the energy minimum of the system.

An Analysis for the Production of Jet Fuel from Soft Wood

While I was still an undergraduate in Chemical Engineering, during the first semester of my senior year (2010) I took a capstone Chemical Engineering Design course. A central part of the course was a design group project to reflect the design methods taught in the course along with key engineering concepts from previous courses. Here is a summary of my group's project titled An Analysis for the Production of Jet Fuel from Soft Wood:

Based on outside research, we developed a process flow diagram with material balances around each unit and a table of components in each stream. I was responsible for coordinating team efforts to meet the project deadlines for design level specifications. As a result, we compiled a detailed report including but not limited to a market analysis, an economic analysis, design specifications using Aspen, reaction chemistry, a heat exchange network, and thermophysical properties of the materials in the process. One important result is that the discounted cash flow rate of return of our process was projected to be 8.09%.

Photo Patterning the Creasing Instability of Surface Attached Hydrogels

Throughout the summer (2010) before my senior year as a Chemical Engineering undergraduate, I worked with Professor Ryan Hayward from the Polymer Science Department in the Conte building at UMass Amherst to study hydrogels.

I crosslinked dropcasted thermoresponsive hydrogels by exposing them to UV light through a photomask created in AutoCAD and then added a drop of saline solution to observe crease formation. Applications of this research are to develop spatiotemporal control over catalytic activity on a surface with dynamic biomolecular patterns and the creation of a lab on chip array.

Predicting Optimal Experimental Routes to a Desired Drop Size Distribution for Pharmaceutical Emulsions

Throughout the summer (2007) as a Chemical Engineering undergraduate, I worked with Professor Surita Bhatia, on a research project for MassNanoTech at the SURE program in UMass Amherst. The abstract is as follows:

The goal of this project is to use a mathematical model to predict drop diameter distributions coming out of a homogenizer from initial data. A coarse emulsion is made using a shear mixer, and then a fine emulsion is prepared using the homogenizer. The drop distribution is measured using various laser techniques. The pharmaceutical value of emulsions is that it is an economical and safe method for hydrophobic drug delivery via the inner hydrophobic part of the drops. The mathematical model will facilitate scale-up in the pharmaceutical industry since drop distribution is integral to determine correct biodistribution of drugs. The problem is that it is far too expensive and time consuming to carry out all the necessary experiments since there are far too many variables involved in predicting drop distribution. A working mathematical model would be a great advancement in emulsion technology. I investigated the effect of pressure, surface tension, viscosity, initial mean distribution, initial width distribution, and volume fraction on drop size distribution theoretically. From the MATLAB program, I gathered mean and variance data. Then I graphed the correlations leaving all values constant each time and changing just one variable. Next they were analyzed and only the most important results were chosen to be presented. General correlations show that increasing pressure in a homogenizer decreases the mean diameter of the drop distribution. Listed from most effective to least effective in increasing drop diameter are surface tension, volume fraction, initial variance, viscosity and initial mean distribution.