Nitrosyl coordination compounds: a comprehensive review of synthetic strategies, bonding elucidation, spectroscopic diagnostics, and computational intelligence from DFT to machine learning and generative AI

Abstract Nitrosyl coordination compounds have undergone a remarkable evolution from Priestley’s serendipitous discovery of nitric oxide to their current standing as versatile molecular platforms at the forefront of modern coordination chemistry, bioinorganic signaling, catalysis, and materials science. This comprehensive review presents an integrated and forward-looking examination of metal–nitrosyl chemistry that systematically bridges classical synthetic methodologies, fundamental bonding paradigms, advanced spectroscopic diagnostics, and cutting-edge computational intelligence. The synthetic landscape is critically surveyed as a decision-oriented framework encompassing direct NO gas reactions, nitrosonium salts, hydroxylamine-mediated protocols, nitrite-based routes, coordinated NO transfer strategies, and emerging green chemistry approaches, with particular emphasis on method selection criteria based on metal identity, ligand sensitivity, scalability, and sustainability imperatives. The bonding narrative is rigorously dissected through the classical lenses of valence-bond theory and molecular orbital analysis, culminating in the modern DFT-based understanding that has revolutionized our comprehension of the elusive M−NO interaction; the linear–bent dichotomy, elegantly captured by the Enemark–Feltham formalism, is established as a predictive tool for rationalizing electron counting, donor capacity, and reactivity patterns, while the non-innocent behaviour of NO-manifesting as valence tautomerism, electron-sink functionality, and dynamic geometric isomerism, is underscored as a fundamental property underpinning catalytic and signalling functions. Spectroscopic characterisation is presented as an integrated diagnostic toolkit wherein complementary techniques, including IR and Raman vibrational spectroscopy, UV–Vis–NIR electronic spectroscopy, electron paramagnetic resonance (EPR), 15 N nuclear magnetic resonance (NMR), X-ray photoelectron spectroscopy (XPS), and Mössbauer spectroscopy, collectively resolve structural, electronic, and magnetic ambiguities through a logically tiered workflow. A distinctive and transformative feature of this review is the extensive coverage of computational intelligence paradigms that have reshaped nitrosyl chemistry: density functional theory (DFT) and time-dependent DFT (TD-DFT) for accurate geometry optimization, vibrational prediction, spectral simulation, and mechanistic elucidation; machine learning algorithms including random forests, support vector machines, XGBoost, and Gaussian processes for rapid high-throughput property prediction of ν (NO) frequencies, M−N bond lengths, M–N–O bond angles, and magnetic properties; generative artificial intelligence through variational autoencoders and generative adversarial networks for de novo molecular design with targeted property constraints; physics-informed neural networks (PINNs) for inverse problem-solving and kinetic modelling; and Bayesian uncertainty quantification for robust decision-making through aleatoric and epistemic uncertainty assessment. This work establishes nitrosyl chemistry as a vibrant, interdisciplinary frontier where molecular innovation converges with computational intelligence, and it serves as an authoritative, decision-oriented reference for both newcomers and experienced practitioners across inorganic chemistry, catalysis, bioinorganic chemistry, and materials science.