Raghunathan Ramakrishnan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Raghunathan Ramakrishnan has a total of 50 co-authors. Here are the top co-authors by number of shared publications:
Sabyasachi Chakraborty
Tata Institute of Fundamental Research, Hyderabad 500046, India
16 shared publications- Surajit Das
Tata Institute of Fundamental Research, Hyderabad 500046, India
6 shared publications Prakriti Kayastha
Tata Institute of Fundamental Research Hyderabad, Hyderabad 500046, India
5 shared publicationsAmit Gupta
Centre for Interdisciplinary Sciences, Tata Institute of Fundamental Research 1 , Hyderabad 500107, India
4 shared publicationsAtreyee Majumdar
Tata Institute of Fundamental Research Hyderabad India
4 shared publications- Atreyee Majumdar
Tata Institute of Fundamental Research, Hyderabad 500046, India
4 shared publications
Latest Publications
- 4 September 2026
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 130 (37) , 7469-7480
A Chemical Space Perspective on Diastereomeric Barriers in Alkylperoxy-to-Hydroperoxyalkyl Isomerization
Low-temperature hydrocarbon autoxidation involves radical intermediates whose reactivity depends not only on the stereochemistry of the intermediates themselves, but also on that of the transient species encountered along the reaction path. This study offers large-scale evidence for the importance of stereochemistry in low-temperature autoxidation by propagating stereochemical information from 498 C1–C7 hydrocarbons through radical formation, O2 addition, and the isomerization of alkylperoxy (ROO•) radicals to hydroperoxyalkyl (•QOOH) radicals. The resulting data set comprises density-functional-theory-level data for 5356 species, including 2324 cyclic diastereomeric transition states associated with 1162 unique ROO• → •QOOH isomerization reactions, with transition-state connectivity confirmed by intrinsic reaction coordinate analysis. Explicit stereochemical treatment reveals that diastereomeric pathways may be either nearly degenerate or separated by more than 60 kcal/mol, with the magnitude of these differences governed by local steric strain, ring constraint, and the environments of the peroxyl-bearing and H-abstraction centers. These results show that constitutionally collapsed molecular representations can systematically miss kinetically relevant reactive channels and provide a foundation for stereochemistry-aware mechanism generation, rate estimation, and predictive combustion modeling.
- 4 September 2026
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 130 (37) , 7469-7480
A Chemical Space Perspective on Diastereomeric Barriers in Alkylperoxy-to-Hydroperoxyalkyl Isomerization
Low-temperature hydrocarbon autoxidation involves radical intermediates whose reactivity depends not only on the stereochemistry of the intermediates themselves, but also on that of the transient species encountered along the reaction path. This study offers large-scale evidence for the importance of stereochemistry in low-temperature autoxidation by propagating stereochemical information from 498 C1–C7 hydrocarbons through radical formation, O2 addition, and the isomerization of alkylperoxy (ROO•) radicals to hydroperoxyalkyl (•QOOH) radicals. The resulting data set comprises density-functional-theory-level data for 5356 species, including 2324 cyclic diastereomeric transition states associated with 1162 unique ROO• → •QOOH isomerization reactions, with transition-state connectivity confirmed by intrinsic reaction coordinate analysis. Explicit stereochemical treatment reveals that diastereomeric pathways may be either nearly degenerate or separated by more than 60 kcal/mol, with the magnitude of these differences governed by local steric strain, ring constraint, and the environments of the peroxyl-bearing and H-abstraction centers. These results show that constitutionally collapsed molecular representations can systematically miss kinetically relevant reactive channels and provide a foundation for stereochemistry-aware mechanism generation, rate estimation, and predictive combustion modeling.
- Open Access
- 5 February 2026
- Published by Wiley in Chemistry – A European Journal
- Vol. 32 (15) , e03557
Insights Into Symmetry and Substitution Patterns Governing Singlet–Triplet Energy Gap in the Chemical Space of Azaphenalenes
Molecules that violate Hund's rule by exhibiting an inverted singlet–triplet gap (STG), where the first excited singlet () lies below the triplet (), are rare but hold great promise as efficient fifth‐generation light emitters. Azaphenalenes (APs) represent one of the few known molecular classes capable of such inversion of the / energy ordering, yet a systematic exploration of all unique APs is lacking. Here, we investigate 104 distinct APs and classify them based on their adherence to or deviation from Hund's rule using – gaps computed with the second‐order coupled‐cluster method employing the Laplace transform (L‐CC2). To capture substitution‐dependent pseudo–Jahn–Teller distortions that are inadequately described by MP2 and DFT methods, we employ a focal‐point extrapolation scheme to obtain near‐CCSD(T)/cc‐pVTZ‐quality geometries. We find 3 APs to undergo and 10 to show symmetry lowering, leading to a total of 117 configurations of 104 unique APs. Our study identifies top candidates with inverted STGs, revealing how substitution and symmetry‐lowering modulate these gaps to uncover new stable AP cores that provide promising targets for designing molecular light‐emitters.
- 5 February 2026
- Published by Wiley in Chemistry – A European Journal
- Vol. 32 (15) , e03557
Open AccessInsights Into Symmetry and Substitution Patterns Governing Singlet–Triplet Energy Gap in the Chemical Space of Azaphenalenes
Molecules that violate Hund's rule by exhibiting an inverted singlet–triplet gap (STG), where the first excited singlet () lies below the triplet (), are rare but hold great promise as efficient fifth‐generation light emitters. Azaphenalenes (APs) represent one of the few known molecular classes capable of such inversion of the / energy ordering, yet a systematic exploration of all unique APs is lacking. Here, we investigate 104 distinct APs and classify them based on their adherence to or deviation from Hund's rule using – gaps computed with the second‐order coupled‐cluster method employing the Laplace transform (L‐CC2). To capture substitution‐dependent pseudo–Jahn–Teller distortions that are inadequately described by MP2 and DFT methods, we employ a focal‐point extrapolation scheme to obtain near‐CCSD(T)/cc‐pVTZ‐quality geometries. We find 3 APs to undergo and 10 to show symmetry lowering, leading to a total of 117 configurations of 104 unique APs. Our study identifies top candidates with inverted STGs, revealing how substitution and symmetry‐lowering modulate these gaps to uncover new stable AP cores that provide promising targets for designing molecular light‐emitters.
- 23 January 2026
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 164 (4)
Enhancing NMR shielding predictions of atoms-in-molecules machine learning models with neighborhood-informed representations
Accurate prediction of nuclear magnetic resonance (NMR) shielding with machine learning (ML) models remains a central challenge for data-driven spectroscopy. We present atomic variants of the Coulomb matrix (aCM) and bag-of-bonds (aBoB) descriptors and extend them using radial basis functions (RBFs) to yield smooth, per-atom representations (aCM-RBF and aBoB-RBF). Local structural information is incorporated by augmenting each atomic descriptor with contributions from the n nearest neighbors, resulting in the family of descriptors, aCM-RBF(n) and aBoB-RBF(n). For 13C shielding prediction on the QM9NMR dataset (831 925 shielding values across 130 831 molecules), aBoB-RBF(4) achieves an out-of-sample mean error of 1.69 ppm, outperforming models reported in previous studies. While explicit three-body descriptors further reduce errors at a higher cost, aBoB-RBF(4) offers the best balance of accuracy and efficiency. Benchmarking on external datasets comprising larger molecules (GDBm, Drug12/Drug40, and pyrimidinone derivatives) confirms the robustness and transferability of aBoB-RBF(4), establishing it as a practical tool for ML-based NMR shielding prediction.
- 23 January 2026
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 164 (4)
Enhancing NMR shielding predictions of atoms-in-molecules machine learning models with neighborhood-informed representations
Accurate prediction of nuclear magnetic resonance (NMR) shielding with machine learning (ML) models remains a central challenge for data-driven spectroscopy. We present atomic variants of the Coulomb matrix (aCM) and bag-of-bonds (aBoB) descriptors and extend them using radial basis functions (RBFs) to yield smooth, per-atom representations (aCM-RBF and aBoB-RBF). Local structural information is incorporated by augmenting each atomic descriptor with contributions from the n nearest neighbors, resulting in the family of descriptors, aCM-RBF(n) and aBoB-RBF(n). For 13C shielding prediction on the QM9NMR dataset (831 925 shielding values across 130 831 molecules), aBoB-RBF(4) achieves an out-of-sample mean error of 1.69 ppm, outperforming models reported in previous studies. While explicit three-body descriptors further reduce errors at a higher cost, aBoB-RBF(4) offers the best balance of accuracy and efficiency. Benchmarking on external datasets comprising larger molecules (GDBm, Drug12/Drug40, and pyrimidinone derivatives) confirms the robustness and transferability of aBoB-RBF(4), establishing it as a practical tool for ML-based NMR shielding prediction.
- 30 December 2025
- Published by IOP Publishing in Journal of Physics: Condensed Matter
- Vol. 38 (1) , 013002
Machine-learned potentials for solvation modeling
Solvent environments play a central role in determining molecular structure, energetics, reactivity, and interfacial phenomena. However, modeling solvation from first principles remains difficult due to the complex interplay of interactions and unfavorable computational scaling of first-principles treatment with system size. Machine-learned potentials (MLPs) have recently emerged as efficient surrogates for quantum chemistry methods, offering first-principles accuracy at greatly reduced computational cost. MLPs approximate the underlying potential energy surface, enabling efficient computation of energies and forces in solvated systems. They are also capable of accounting for effects such as hydrogen bonding, long-range polarization, and conformational changes. This review surveys the development and application of MLPs in solvation modeling. We summarize the theoretical basis of MLP-based energy and force predictions and present a classification of MLPs based on training targets, model types, and design choices related to architectures, descriptors, and training protocols. Integration into established solvation workflows is discussed, with case studies spanning small molecules, interfaces, and reactive systems. We conclude by outlining open challenges and future directions toward transferable, robust, and physically grounded MLPs for solvation-aware atomistic modeling.
- 30 December 2025
- Published by IOP Publishing in Journal of Physics: Condensed Matter
- Vol. 38 (1) , 013002
Machine-learned potentials for solvation modeling
Solvent environments play a central role in determining molecular structure, energetics, reactivity, and interfacial phenomena. However, modeling solvation from first principles remains difficult due to the complex interplay of interactions and unfavorable computational scaling of first-principles treatment with system size. Machine-learned potentials (MLPs) have recently emerged as efficient surrogates for quantum chemistry methods, offering first-principles accuracy at greatly reduced computational cost. MLPs approximate the underlying potential energy surface, enabling efficient computation of energies and forces in solvated systems. They are also capable of accounting for effects such as hydrogen bonding, long-range polarization, and conformational changes. This review surveys the development and application of MLPs in solvation modeling. We summarize the theoretical basis of MLP-based energy and force predictions and present a classification of MLPs based on training targets, model types, and design choices related to architectures, descriptors, and training protocols. Integration into established solvation workflows is discussed, with case studies spanning small molecules, interfaces, and reactive systems. We conclude by outlining open challenges and future directions toward transferable, robust, and physically grounded MLPs for solvation-aware atomistic modeling.
- Open Access
- 17 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 387-395
Indistinguishability domains of neural microcircuit motifs mapped through classification scores of postsynaptic spike counts
- 17 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 387-395
Open AccessIndistinguishability domains of neural microcircuit motifs mapped through classification scores of postsynaptic spike counts
- Open Access
Indistinguishability domains of neural microcircuit motifs mapped through classification scores of postsynaptic spike counts
In the kinetic modeling of a chemical synapse, the reversal potential ( E syn ) and conductance amplitude ( ) of the synapse jointly regulate the spike-induced activity of the postsynaptic neuron. We simulate two-neuron microcircuit motifs of feedforward ( i → j ) and feedback ( i ⇌ j ) types using random external current pulses provided only at the presynaptic neuron. For combinations of E syn and , corresponding to an excitatory synapse, a supervised machine learning method for classification distinguishes the motifs perfectly with a score of 1.0 when using binned counts of postsynaptic spikes as the input. Strongly inhibitory combinations of these parameters result in no postsynaptic response in both types of microcircuit motifs; hence, the classifier does not improve upon a random assignment with a score of 0.5. In other domains of the parameter space spanned by E syn and , corresponding to diminished excitation/inhibition, the classifier fails (0.5 < score < 1), indicating the challenge in identifying the nature of the synapse inferred exclusively from postsynaptic spikes. For this task, the gradient boosting classifier gives higher classification accuracies that improve with further training, compared to other classifiers explored in this study.
Open AccessIndistinguishability domains of neural microcircuit motifs mapped through classification scores of postsynaptic spike counts
In the kinetic modeling of a chemical synapse, the reversal potential ( E syn ) and conductance amplitude ( ) of the synapse jointly regulate the spike-induced activity of the postsynaptic neuron. We simulate two-neuron microcircuit motifs of feedforward ( i → j ) and feedback ( i ⇌ j ) types using random external current pulses provided only at the presynaptic neuron. For combinations of E syn and , corresponding to an excitatory synapse, a supervised machine learning method for classification distinguishes the motifs perfectly with a score of 1.0 when using binned counts of postsynaptic spikes as the input. Strongly inhibitory combinations of these parameters result in no postsynaptic response in both types of microcircuit motifs; hence, the classifier does not improve upon a random assignment with a score of 0.5. In other domains of the parameter space spanned by E syn and , corresponding to diminished excitation/inhibition, the classifier fails (0.5 < score < 1), indicating the challenge in identifying the nature of the synapse inferred exclusively from postsynaptic spikes. For this task, the gradient boosting classifier gives higher classification accuracies that improve with further training, compared to other classifiers explored in this study.
- Open Access
- 18 September 2025
- Published by Wiley in Journal of Computational Chemistry
- Vol. 46 (25) , e70228
Leveraging the Bias‐Variance Tradeoff in Quantum Chemistry for Accurate Negative Singlet‐Triplet Gap Predictions: A Case for Double‐Hybrid DFT
Molecules that have been suggested to violate the Hund's rule, having a first excited singlet state () energetically below the triplet state (), are rare. Yet, they hold the promise to be efficient light emitters. Their high‐throughput identification demands exceptionally accurate excited‐state modeling to minimize qualitatively wrong predictions. We benchmark twelve energy gaps to find that the local‐correlated versions of ADC(2) and CC2 excited state methods deliver excellent accuracy and speed for screening medium‐sized molecules. Notably, we find that double‐hybrid DFT approximations (e.g., B2GP‐PLYP and PBE‐QIDH) exhibit high mean absolute errors () despite very low standard deviations (). Exploring their parameter space reveals that a configuration with 75% exchange and 55% correlation, which reduces the mean absolute error to below 5 meV, but with an increased variance. Using this low‐bias parameterization as an internal reference, we correct the systematic error while maintaining low variance, effectively combining the strengths of both low‐bias and low‐variance DFT parameterizations to enhance overall accuracy. Our findings suggest that low‐variance DFT methods, often overlooked due to their high bias, can serve as reliable tools for predictive modeling in first‐principles molecular design. The bias‐correction data‐fitting procedure can be applied to any general problem where two flavors of a method, one with low bias and another with low variance, have been identified a priori.
- 18 September 2025
- Published by Wiley in Journal of Computational Chemistry
- Vol. 46 (25) , e70228
Open AccessLeveraging the Bias‐Variance Tradeoff in Quantum Chemistry for Accurate Negative Singlet‐Triplet Gap Predictions: A Case for Double‐Hybrid DFT
Molecules that have been suggested to violate the Hund's rule, having a first excited singlet state () energetically below the triplet state (), are rare. Yet, they hold the promise to be efficient light emitters. Their high‐throughput identification demands exceptionally accurate excited‐state modeling to minimize qualitatively wrong predictions. We benchmark twelve energy gaps to find that the local‐correlated versions of ADC(2) and CC2 excited state methods deliver excellent accuracy and speed for screening medium‐sized molecules. Notably, we find that double‐hybrid DFT approximations (e.g., B2GP‐PLYP and PBE‐QIDH) exhibit high mean absolute errors () despite very low standard deviations (). Exploring their parameter space reveals that a configuration with 75% exchange and 55% correlation, which reduces the mean absolute error to below 5 meV, but with an increased variance. Using this low‐bias parameterization as an internal reference, we correct the systematic error while maintaining low variance, effectively combining the strengths of both low‐bias and low‐variance DFT parameterizations to enhance overall accuracy. Our findings suggest that low‐variance DFT methods, often overlooked due to their high bias, can serve as reliable tools for predictive modeling in first‐principles molecular design. The bias‐correction data‐fitting procedure can be applied to any general problem where two flavors of a method, one with low bias and another with low variance, have been identified a priori.
- Open Access
- 30 July 2025
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 27 (30) , 16266-16267
Correction: Influence of pseudo-Jahn–Teller activity on the singlet–triplet gap of azaphenalenes
Correction for ‘Influence of pseudo-Jahn–Teller activity on the singlet–triplet gap of azaphenalenes’ by Atreyee Majumdar et al., Phys. Chem. Chem. Phys., 2024, 26, 26723–26733, https://doi.org/10.1039/D4CP02761B.
- 30 July 2025
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 27 (30) , 16266-16267
Open AccessCorrection: Influence of pseudo-Jahn–Teller activity on the singlet–triplet gap of azaphenalenes
Correction for ‘Influence of pseudo-Jahn–Teller activity on the singlet–triplet gap of azaphenalenes’ by Atreyee Majumdar et al., Phys. Chem. Chem. Phys., 2024, 26, 26723–26733, https://doi.org/10.1039/D4CP02761B.
- Open Access
- 8 July 2025
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 16 (31) , 14392-14407
Unlocking inverted singlet–triplet gap in alternant hydrocarbons with heteroatoms
BN-substituted polycyclic aromatic hydrocarbons exhibit negative singlet–triplet gaps.
- 8 July 2025
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 16 (31) , 14392-14407
Open AccessUnlocking inverted singlet–triplet gap in alternant hydrocarbons with heteroatoms
BN-substituted polycyclic aromatic hydrocarbons exhibit negative singlet–triplet gaps.
- Open Access
Unlocking Inverted Singlet-Triplet Gap in Alternant Hydrocarbons with Heteroatoms
Fifth-generation organic light-emitting diodes exhibit delayed fluorescence even at low temperatures, enabled by exothermic reverse intersystem crossing from a negative singlet-triplet gap (STG), where the first excited singlet lies anomalously below the triplet. This phenomenon -- termed delayed fluorescence from inverted singlet and triplet states (DFIST) -- has been experimentally confirmed only in two triangular molecules with a 12-annulene periphery and a central nitrogen atom. Here, we report a high-throughput virtual screening of 30,797 BN-substituted polycyclic aromatic hydrocarbons derived from 77 parent scaffolds (2--6 rings). Using a multi-level workflow combining structural stability criteria with accurate L-CC2 excited-state calculations, we identify 72 heteroaromatic candidates with STGs$<0$. Notably, this includes BN-helicenes, where inversion arises from through-space charge-transfer states. Several systems exhibit non-zero oscillator strengths, supporting their potential as fluorescent emitters. Our findings reveal new design motifs for DFIST beyond known frameworks, expanding the chemical space for next-generation emitters based on heteroatom-embedded aromatic systems.
Open AccessUnlocking Inverted Singlet-Triplet Gap in Alternant Hydrocarbons with Heteroatoms
Fifth-generation organic light-emitting diodes exhibit delayed fluorescence even at low temperatures, enabled by exothermic reverse intersystem crossing from a negative singlet-triplet gap (STG), where the first excited singlet lies anomalously below the triplet. This phenomenon -- termed delayed fluorescence from inverted singlet and triplet states (DFIST) -- has been experimentally confirmed only in two triangular molecules with a 12-annulene periphery and a central nitrogen atom. Here, we report a high-throughput virtual screening of 30,797 BN-substituted polycyclic aromatic hydrocarbons derived from 77 parent scaffolds (2--6 rings). Using a multi-level workflow combining structural stability criteria with accurate L-CC2 excited-state calculations, we identify 72 heteroaromatic candidates with STGs$<0$. Notably, this includes BN-helicenes, where inversion arises from through-space charge-transfer states. Several systems exhibit non-zero oscillator strengths, supporting their potential as fluorescent emitters. Our findings reveal new design motifs for DFIST beyond known frameworks, expanding the chemical space for next-generation emitters based on heteroatom-embedded aromatic systems.