Vaneet Saini

A neural network architecture efficiently predicted TC ⁺ DA values using five easily computed and interpretable Mordred-based structural descriptors.

Understanding and predicting chemical reaction behavior is a fundamental challenge in chemistry. The Hammett equation, introduced in 1935, has been a cornerstone in modelling structure–activity relationships, particularly in physical organic chemistry. In this study, machine learning (ML) is employed to predict Hammett constants (σ_m and σ_p) for a diverse collection of benzoic acid derivatives, based on a dataset previously developed by our group that includes over 900 molecules spanning meta-, para-, and symmetrically substituted variants. Quantum chemical descriptors, combined with Mordred-based electronic, steric, and topological descriptors, were used to train models such as Extra Trees (ET) and Artificial Neural Networks (ANNs). The ANN model achieved the highest accuracy, with a test R² of 0.935 and an RMSE of 0.084, outperforming other models and a previously developed graph neural network. Feature importance analysis revealed key descriptors, including NBO charges and HOMO energies, driving the predictions. Applicability domain (AD) analysis identified outliers and compounds outside the AD, ensuring model reliability. This work highlights the potential of ML in predicting Hammett constants, offering a robust tool for chemical reactivity analysis and molecular design.

Metal–organic frameworks (MOFs) are emerging as a key solution for detecting and adsorbing hazardous nitroaromatic compounds, driven by environmental and security concerns. In this paper, a high pH and water-stable fluorescent MOF (PUC-10) with distorted pentagonal bipyramidal geometry and interesting hcb topology was prepared by a solvothermal method. Furthermore, a PUC-10 NP based composite (PUC-10@HNT) was fabricated and applied for the sensitive detection of hydroxyl group-containing nitroaromatic compounds (NACs). In particular, trinitrophenol (TNP), dinitrophenol (DNP) and para-nitrophenol (4-NP) rapidly quench the fluorescence of PUC-10@HNT, with a detection limit of 0.027 μM, 0.031 μM and 0.036 μM, respectively. The sensing ability of PUC-10 and PUC-10@HNT was compared to signify the composite formation. This method effectively analyzes the respective NACs in natural water and a Whatman-paper-strip sensing element was also developed for efficient visual detection of analytes. Moreover, PUC-10@HNT showed excellent recyclability over four cycles. Based on the DFT calculations, zeta-potential, absorption and photoluminescence spectra of the analytes and probe, the fluorescence quenching mechanism was analyzed and found to be due to a combination of PET, FRET, and competitive light absorption between the analytes and PUC-10@HNT.

The primary objective of this work is to delve into the intricacies of allylic fluorination reactions through the application of density functional theory (DFT) calculations. These reactions hold significant importance in the realm of synthesizing organofluorine compounds. The specific focus lies on comprehending the interaction mechanisms when styrenes, a class of organic molecules, come in contact with an electrophilic fluorinating reagent known as Selectfluor. Notably, this interaction pathway demonstrates remarkable efficiency in yielding allylic fluoride products. The proposed mechanism for this transformation involves a sequential process. To unveil the microcosmic intricacies governing this reaction between the alkene substrate and Selectfluor, advanced computational methodologies are employed. The paper systematically outlines the computational strategies harnessed to probe the minute details of the reaction mechanism. The outcomes of these computations are subsequently subjected to thorough analysis, encompassing crucial facets such as transition states and energy barriers. This analytical depth enhances the fundamental understanding of the reaction mechanism and sheds light on the underlying factors influencing its feasibility and efficiency. In a broader context, the insights garnered from this study carry significant utility. They provide pivotal guidance for the optimization of reaction conditions, facilitating the fine-tuning of experimental setups. Moreover, the elucidated mechanism serves as a platform for the design of even more efficient and selective allylic fluorination reactions. This paper, by amalgamating theoretical insights with practical synthetic objectives, contributes to the broader advancement of organofluorine compound synthesis and allied fields.

While the carbonyl olefination has been extensively studied and well documented, use of α-hydroxy ketones as precursors for the carbonyl olefination is not reported, till date. Herein, a transition-metal-free and selective Knoevenagel-type deconstructive carbonyl olefination of α-hydroxy ketones using arylacetonitriles under mild reaction conditions is presented. The reaction affords valuable scaffolds of acrylonitriles with the use of α-hydroxy ketones as precursors for carbonyl olefination.

Machine learning based statistical models have played a significant role in increasing the speed and accuracy with which the chemical and physical properti