We work on a range of problems in chemical reactivity, with particular emphasis on catalytic reactions and asymmetric transformations. To study the catalytic action, a combination of computational chemistry and machine learning methods are employed.
Catalytic reaction of our current interest is cooperative multi-catalytic transformations involving two or more catalysts. We examine such reactions by identifying various intermediates and transition states along the reaction path from reactants to products. Molecular insights, such as the weak interactions in the transition state, are subsequently employed for predicting new catalyst/reactions. Good number of examples on noncovalent interaction (NCI) driven asymmetric catalysis have been reported by group.
Another domain of our active research efforts is in the use of machine learning (ML) for chemical catalysis. We aim to identify optimal reaction conditions to maximize the yield and selectivity, in relatively small data settings. The ML models are trained to predict reaction outcomes (yield, enantio-, regio- selectivities). Reliable models, built on various deep learning protocols are also employed for generative applications. The idea is to deploy such models to expedite reaction discovery using an ML-guided approach.

Specific research interests:

1. Asymmetric multi-catalytic reactions
   • Asymmetric reactions involving transition metal catalysts and organocatalysts.
   • Origin of enantioselectivity and catalyst design.
2. Mechanistic studies on C-H bond activation reaction
   • Role of additives and solvents
   • Rational modifications to catalysts and substrates
3. Machine learning in catalysis
   • Prediction of reaction outcome
   • Artificial intelligence (AI)-enabled catalyst design