Login / Signup

Predicting wavelength-dependent photochemical reactivity and selectivity.

Jan Philipp MenzelBenjamin B NobleJames P BlincoShivshankar R Mane
Published in: Nature communications (2021)
Predicting the conversion and selectivity of a photochemical experiment is a conceptually different challenge compared to thermally induced reactivity. Photochemical transformations do not currently have the same level of generalized analytical treatment due to the nature of light interaction with a photoreactive substrate. Herein, we bridge this critical gap by introducing a framework for the quantitative prediction of the time-dependent progress of photoreactions via common LEDs. A wavelength and concentration dependent reaction quantum yield map of a model photoligation, i.e., the reaction of thioether o-methylbenzaldehydes via o-quinodimethanes with N-ethylmaleimide, is initially determined with a tunable laser system. Combined with experimental parameters, the data are employed to predict LED-light induced conversion through a wavelength-resolved numerical simulation. The model is validated with experiments at varied wavelengths. Importantly, a second algorithm allows the assessment of competing photoreactions and enables the facile design of λ-orthogonal ligation systems based on substituted o-methylbenzaldehydes.
Keyphrases
  • light emitting
  • machine learning
  • structural basis
  • high resolution
  • electronic health record
  • diabetic rats
  • big data
  • energy transfer
  • data analysis
  • electron transfer
  • amino acid
  • virtual reality
  • monte carlo