A photochemical process for complex pharmaceutical intermediate manufacturing
Eric Fang , chief scientific officer at Snapdragon Chemistry , shares a case study in the use of photochemistry
a b
Figure 1 - Spectra of a typical commercial mercury lamp used in photochemistry ( a ) & commercial LED emission spectra ( b )
Since its birth on a rooftop of Università di Bologna over 100 years ago , generations of organic chemists have envisioned photochemistry as a general means for the largescale production of chemicals . Photochemical excitation enables molecular activation via a ‘ quantum leap ’, thereby achieving types of molecular architectures that are difficult to achieve using conventional means , such as heating or chemical reagents . There are many examples of using photochemistry to make milligramscale compounds . One is vitamin D3 production , which uses UV light generated by mercury lamps . Yet very few examples exist that are greater than a few grams in scale , a far cry from the kilogram to tonne production scale required for commercial production .
Hurdles to clear
There are many challenges to the scale-up of a photochemical process , even though photons can be simplistically considered an unconventional ‘ reagent ’ that is available on demand via a wall plug . First , the photon energy level needs to be matched with the substrate electron excitation spectrum so that there is sufficient interaction between photons and molecules . If the light source and reaction wavelengths are mismatched , photons might simply pass through the substrate without anything happening , or worse , they could become destructive . For example , in the case of mercury lamps , mercury emits ' mixed reagents ' of mixed wavelengths ( Figure 1a ). The ‘ yield ’ of a specific wavelength photon ( e . g . 365 nm ) from a medium pressure mercury is generally low (< 10 %). The unwanted wavelength
( e . g . high energy UVC 254 nm and UVB lines ) could cause substrate or product degradation , leading to reactor discolouration and fouling . Second , photons must be introduced into a reactor containing starting materials . This is easier said than done when dealing with nontransparent reactor vessel walls and very low surface-to-volume ratios of conventional chemical reactors . Beer ’ s law of exponential light transmittance also dictates the physics of light penetration into reaction solution , often fully absorbed within < 1 mm penetration depth under production concentrations . This creates locally high photon concentration , often leading to undesirable outcomes if mixing is inadequate . Third , quantum yield , which is defined as productive chemical turnovers per photon absorbed , often varies by orders of magnitude . In
58 SPECIALITY CHEMICALS MAGAZINE ESTABLISHED 1981