Research Topics

Explore the key research themes, active projects, and scientific investigations carried out at the Renjith Thomas Lab.
Software we use are listed (incomplete list)
Quantum Chemistry: Gaussian 09W (licence), Orca (open access)
Turbomole (licence)
Wavefunction: QTAIM (licence), Multiwfn  (open access)
Conformations: GFNn-xTB (n=0,1,2), GFN-FF, and g-xTB (through atomistica.online), Vconf
Visualisation: Gaussview 5.0, Chemcraft (licence), Avogadro (open access)
Docking: Autodock Vina4 (open acess)
Hardware: Xeon 20 core with 128 GB RAM, Xeon 16 core with 64 GB RAM, I9 20 core with 64 GB RAM, I7 10 core with 32 GB RAM
Cluster access through BRAF, PARAM Utkarsh (CDAC), ACCESS (harware and software)
Data Ananlysis and ML: Jamovi (open access) and Python
Online platforms: www.atomistica.online, XEDA

Digital Chemistry- ML and AI for Sustainable Chemistry

Current:
  • Dr Ali Khiarbek (Post Doc)
  • Manjesh Mathew (PhD student)
  • Niyah Mary Eapen (PhD student)

Digital chemistry, as practised in this group, treats machine learning not as a substitute for electronic-structure theory but as a layer built upon it, with the emphasis on models whose predictions trace back to physically meaningful descriptors and whose limits of validity are stated rather than assumed. Recent work has applied interpretable ensemble learning to the open Quantum Cluster Database, reproducing DFT energies per atom for coinage-metal nanoclusters to within roughly 15 meV per atom and recovering both the classical magic numbers and the gold-specific signatures of relativistic stabilisation, with a geometry-only variant that retains almost all of this accuracy from atomic coordinates alone. A companion study extended the approach to boron, silicon, arsenic and tellurium clusters, where SHAP and partial-dependence analysis exposed the coupling between cluster shape and electronic structure. In the f block, relativistic density-functional calculations combined with explainable learning have shown that the electron affinities of actinide(IV) carboxylate complexes are governed by a single physical variable, the occupation of the 5f shell. Running through all of this is a deliberate stress-testing of transferability, by leave-one-metal-out, leave-one-element-out and size-grouped cross-validation, which establishes where such models can be trusted and, equally important, where they cannot. An Indian utility patent covering a machine-learning system for the prediction of gas-phase proton affinity was published in July 2026, pending final granting process (Manjesh Mathew).
Our future plans are on the study of storage devices for green hydrogen production and sustainable chemistry.

Please read the following papers. A.A. Khairbek, M.I. Al-Zaben, A.Y.A. Alzahrani, R. Thomas, Interpretable machine-learning prediction of DFT energies per atom and identification of magic numbers in coinage-metal nanoclusters (N ≤ 55) from the open quantum cluster database, Phys. Chem. Chem. Phys. (2026). https://doi.org/10.1039/d6cp01474g R. Thomas, A.A. Khairbek, S.A. Sunny, A.Y.A. Alzahrani, Machine learning analysis of metalloid nanoclusters from the Quantum Cluster Database: Structural-phase transitions, electronic-geometric coupling, and exploratory binary nano-alloy predictions, Calphad 94 (2026) 102978. https://doi.org/10.1016/j.calphad.2026.102978 A.A. Khairbek, M. Abd Al-Hakim Badawi, R. Puchta, D.I. Saleh, S.F. Mahmoud, R. Thomas, Electron affinity of actinide(IV) carboxylate complexes from MN12-L density-functional calculations and explainable machine learning, J. Comput. Chem. 47 (2026) e70469. https://doi.org/10.1002/jcc.70469

Berchmans Protocol for Modelling Non-Covalent Interactions 1.0 (BerchNCI 1.0)

Please read our article at https://publishing.aidasco.org/journals/index.php/aire/article/view/84/152

Modeling of Polymer-Solvent Systems & Blends

Current:
  • Dr Athira Maria John (Post Doc)
  • Dr Rehin Sulay (Post Doc)

Aromaticity and Proton Sponges

Current:
  • Manjesh Mathew (PhD student)
  • Prince Sebastian (MSc student)

Electron Upconversion (EUC) in molecular catalysis

Current:
  • Meera Kattoor PhD Student
Past:

With active support of our collaborator Prof Igor Alabugin from Florida State University, USA ,  we have examined the fundamental understanding of electronic structures to drive practical innovation, as exemplified by our work on Electron Upconversion (EUC) in molecular catalysis. We establishes the EUC  as a  computational framework for characterizing catalysts that drive high-energy chemical transformations through electronic excitation. Traditional energy profiles are insufficient for a complete understanding of catalytic efficiency; instead, the EUC protocol successfully maps the electronic anatomy of a reaction by integrating IGMH and NBO analysis to resolve hidden electronic rearrangements and donor-acceptor interactions. A key result is the protocol’s ability to precisely quantify the contributions of individual atomic fragments to the stabilization of transient states, providing a standardized, predictive methodology for the rational design of light-harvesting systems. These high-fidelity electronic descriptors provide the essential data foundation required to train future machine learning models for the autonomous discovery of sustainable catalysts.
Please see our paper in the Journal of Computational Chemistry – Electron Upconversion Enables C-P and C-S Bond Formation Under Mild Oxidative Conditions: A Theoretical Study

Computational Study of Catalytic Reactions

Current:
  • Dr Ali Khairbek (Post Doc)
Past:
  • Dr Zakir Ullah (collaborator)

Chemical Bond – Sulphur centered H Bond, Halogen bond & Tetral bond

Current:
  • Sneha Anna Sunny (PhD student)
  • Alen Binu Abrahan (Visiting researcher from Austonomous Universityof Madrid, Spain and Erasmus Mundus Fellow)
Past:
  • Arnav Paul (NIISER , now PhD Student at University of Illinois at Urbana-Champaign)
  • Mebin Varghese (Visitor, now PhD Student at VIT)
  • Dr Aristote Matando (co-supervised PhD student, University of Kingshasa, Congo)

Weak Interactions, Microsolvation and its dynamics

Current:
  • Sneha Anna Sunny (PhD student)
  • Francis Thomas (PhD student)
Past:
  • Dr. T. Pooventhiran (post doc, Currently at IISER and KR College)
  • Dr Jisha Mary Thomas (Post Doc, currently at CHRIST)

Synthesis of bioactive compounds

Current:
  • Rajimon KJ
Past:
  • Dr N Elangovan (Post Doc)

Get In Touch

We welcome research collaborations and enquiries from theoreticians, experimental chemists, and interdisciplinary researchers. Prospective PhD, postdoctoral, and student researchers are encouraged to contact us via the details below or the message form.

Dr. Renjith Thomas

Department of Chemistry

St. Berchmans College (Autonomous)

Changanassery, Kerala, India – 686101


[email protected]

+91 95446 58314