Molecular Simulations

Gelation and network formation in polymer-grafted nanoparticles

Some initial work in our group, and from other groups suggests that polymer-grafted nanoparticles can for networks and equilibrium gels under the right conditions.  This is remarkable, since while gels are useful most gels represent non-equilibrium states that age, and disintegrate with time.  The idea of forming equilibrium gels which are non-perishable, is therefore attractive.  In this study we determine the conditions for the formation of equilibrium gels by grafted nanoparticles.

A basic understanding of coding is required.

The phase behavior of connected hard and soft particles.

A surprising new development in materials science and chemical engineering is the finding that mixtures of hard (colloidal), and soft (polymeric, or micellar) particles can self organize on length scales much larger than the diameter of either species.  In this project we explore the behavior of connected hard- and soft particles.  An elementary knowledge of coding is sufficient.

Design and synthesis studies of porous/catalytic materials

The synthesis of porous catalytic materials has profound impact in the chemical industries. The effectiveness of these materials is governed by the structure and surface morphology which is controlled by the synthesis parameters (such as temperature, synthesis time, pH, additives). This project is aimed at understanding role of synthesis parameters for the better control over porosity, surface morphology and structure of porous catalytic materials using simulations and possible experiments.

Simulation study of Enhance Oil Recovery

The crude oil in direct contact with mineral surface needs to be displaced using external medium (solvent + additives) in the secondary and tertiary phase of recovery. The mechanism of replacement is governed by the structural and energetic behaviour of interfacial system (solvent + additives + hydrocarbon oil) at the mineral surface. This project is aimed to obtained molecular understanding of the interfacial system (crude oil+solvent+mineral) to design better displacing agents for the economic recovery of oil.

Materials for water purification and desalination (TA or FA only)

Although earth is covered with 70% of water, only 2% of it is available as fresh drinkable water. Access to this fresh water is scarce in many parts of the country. The groundwater contamination due to industrial pollution and geological minerals leads to many health issues, especially in childrens and women. Conversion of sea-water to fresh water is an expensive and energy-intensive process. The aim of this project is to find organic and inorganic porous materials for water purification.

Design of nanoporous materials for gas separation (TA or FA only)

Natural gas meets around 20-25% of world energy demands. Overall the world has around 200 trillion cubic meters of natural gas reserve and new reservoirs are being found. Methane gas constitutes around 80-90% of natural gas and for economical utilization of methane as fuel, efficient separation technology is required. The aim of this project is to design new nanoporous materials for methane separation and storage from natural gas.

Computational design of bimetallic catalysts

Bimetallic catalysts are synthesized using two different metals. For instance, Ag-Au, Pt-Ni, Au-Pt are examples of bimetallic catalysts. For some reactions these materials are more promising than pure metal catalyst. A question arises about the role of the individual metal species. We shall employ a combination of state-of-the-art density functional theory, molecular dynamics and kinetic Monte Carlo simulations to study the effect of Au-Pt composition, surface arrangement and other experimental parameters on the rate of methanol electro-oxidation reaction.

Machine learning techniques applied to molecular simulations

Molecular simulations of catalytic and catalyst support materials are often computationally expensive. These simulations can directly provide information about how chemical rates may depend on atomic position, or the interactions between atoms. Here we explore the use of machine learning techniques. Machine learning techniques (neural network, random forest, clustering, gaussian process regression) will be used to develop atomic scale models for catalytic and fast ion conductors.