Akash Gutal
Graduate Student (PMRF) Department of Chemistry, Indian Institute of Technology Jodhpur
Graduate Student (PMRF) Department of Chemistry, Indian Institute of Technology Jodhpur
Hi, I am Akash, a research scholar at the Department of Chemistry, IIT Jodhpur. I conduct my research in the Reaction Dynamics Laboratory under the supervision of Dr. Manikandan Paranjothy.
We study the 'Jiggling and Wiggling' of atoms happening at the molecular level during the chemical reaction. The method we use is called direct dynamics, which involves the study of the time evolution of a chemical system under theoretically simulated experimental conditions. The purpose of these studies is to gain a detailed understanding of atomistic mechanisms and associated properties such as product branching ratios and reactive scattering. The method is quite accurate but comes with the expense of high computational cost.
The reactions we study using direct dynamics are limited to a few atoms (10-15), as going beyond that is very difficult due to the inherent 'curse of dimensionality'. We don't know how god is simulating ~1080 atoms of the entire universe for us! Fortunately, Scientists have successfully mimicked the human brain cells called 'neurons' into computers, termed as Artificial Neural Networks. These neural networks can learn from a large amount of high-dimensional data, such as the potential energy surface of thousands of atoms, enabling us to run dynamics simulations and extract properties of chemical systems.
Currently, I am developing a methodology to simulate the dynamics of unimolecular and bimolecular reactions using Artificial Neural Networks. These artificial neurons are employed to fit the potential energy surfaces of the reactions, which allows us to perform efficient integration of the equations of motion and facilitates the extraction of key kinetic and thermodynamic properties.
Research Supervisor: Dr. Manikandan Paranjothy (web page)
IIT Jodhpur
Direct Dynamics of Bimolecular Reactions
The nucleophilic substitution reactions (SN2) are important reactions in organic chemistry. The classical dynamics of these reactions are revealing increasingly complex behaviour.
Machine Learning Potentials
The method that enables us to simulate high-dimensional chemical systems. We recently developed a full-dimensional potential energy surface for a seven-atom bimolecular reaction.
Electronic Structure Calculations
DFT calculations on gas-surface interactions and transition metal catalysts revealed key thermochemical and mechanistic insights.
Developing interfaces for VENUS
I have a strong interest in computers and programming. I’ve developed interfaces for the classical dynamics software VENUS, with several electronic structure and machine learning packages, including MOPAC, QUICK, and aenet-PyTorch.
Contact
161- Arjun Hostel (G6) Indian Institute of Technology, Jodhpur, NH-62 Karwar, Rajasthan
Lab: Reaction dynamics group (211)
Email: akashpg@iitj.ac.in, agpk42@gmail.com
If my research has sparked your curiosity, I’d be delighted to hear from you. I’m always eager to dive into scientific discussions—after all, great ideas often begin with a simple conversation.
"Everything that living things do can be understood in terms of the jigglings and wiggling of atoms."- Richard Feynman