A collaboration between the groups of Professor Mónica H. Pérez-Temprano at the Institute of Chemical Research of Catalonia (ICIQ) and Professor Anat Milo at Ben-Gurion University of the Negev has uncovered how the characteristics of specific substrates require certain reaction conditions that determine the course of a chemical reaction, in the context of C–H deuteration reactions. The study, published in Nature Catalysis, combines detailed experiments with data science rooted in reaction intermediates. By correlating molecular features with reaction outcomes, the researchers reveal that the choice of deuterium source—such as heavy water (D₂O), deuterated methanol (CD₃OD), or acetic acid-d₄ (AcOD-d₄)—does more than merely influencing the degree of deuterium incorporation. It can actively alter the reaction pathway, revealing hidden mechanistic complexity that intuition alone could not predict.

“Even a seemingly simple transformation such as hydrogen isotope exchange can proceed via distinct mechanistic pathways under different reaction conditions,” explains Prof. Pérez-Temprano. “Our work argues against assuming a single mechanistic model across substrates or metals and underscores the importance of evaluating multiple deuterium sources.”

Hydrogen isotope exchange (HIE) reactions are of particular interest in both fundamental and applied chemistry. Transition metal-catalysed HIE provides an efficient route to introduce isotopic labels into organic molecules—essential tools in medicinal chemistry, analytical studies, and mechanistic research. Deuterium-labelled compounds are increasingly valuable in the pharmaceutical industry, where they are used to probe drug metabolism, reduce toxicity, and enhance pharmacokinetic profiles. The approval in last years of the first deuterated drug by the US Food and Drug Administration underscores the growing relevance of these transformations. Yet, despite their widespread use, the choice of isotope source in HIE reactions has remained largely empirical and poorly understood.

To address this gap, the authors combined experimental mechanistic studies with computational analyses based on molecular descriptors—numerical values that represent molecular features and allow quantitative correlations between structure and reactivity. The findings reveal a strong and unexpected link between the substrate structure and the preferred deuterium source. The isotope source itself can act as a mechanistic trigger, directing the reaction along distinct pathways. Remarkably, the team found that the same cobalt catalyst could undergo three different mechanisms under three sets of conditions—highlighting the system’s sensitivity to subtle changes.

The implications of this research extend beyond HIE. “This method complements traditional computational mechanistic studies,” notes Prof. Milo. “While computations focus on a full reaction trajectory using one substrate, our approach captures how mechanistic behaviour varies across broader substrate classes and conditions—offering new tools for interrogating how catalytic systems operate in practice.”

Recognising this sensitivity reshapes how chemists interpret catalytic behaviour: a simple change in reaction conditions, substrate, or metal can decisively alter the reaction’s course. Building on the concept of linear free energy relationships (LFER), the researchers developed a statistical methodology to explore whether different reaction intermediates could shed light on divergent outcomes depending on the substrate and reaction conditions. By moving beyond the traditional practice of examining a single “model substrate,” this work discloses a broader, data-driven analysis to map and understand mechanistic diversity

Reference publication

Charting the influence of deuterium sources in hydrogen isotope exchange using a cobalt(III) catalytic platform
Barranco, S.; Eshel, I. L.; Zhang, J.; Di Matteo, M.; Milo, A.; Pérez-Temprano, M. H.
Nat. Catal. 2025
DOI: 10.1038/s41929-025-01447-x

La entrada When substrates dictate the route: Reaction conditions shape mechanistic pathways se publicó primero en ICIQ.