IMPROVED PHOTOSYNTHESIS OF PREVITAMIN-D BY WAVELENGTHS OF 280-300 NM

The photochemistry of 7-dehydrocholesterol yielding previtamin D, the precursor of vitamin D, is reinvestigated with respect to the optimum wavelength at which the best selectivity for previtamin D vs. side products is obtained. Long irradiation periods lead to a steady state whose wavelength-dependent composition can be calculated by a kinetic model employing the absorption cross-sections of the components and the quantum yields of the reactions. The maximum stationary concentration of the desired isomer (greater than 60%) is obtained at 296 nm. Experimental results at several wavelengths show good agreement with the calculated kinetics. These investigations show that the process is more selective and more economic if the irradiation is stopped before the steady state is reached. The possible light sources are discussed. Lasers can easily generate the desired wavelengths, e.g. a KrF laser twice Raman shifted in deuterium (292 nm). However, their photons are still too expensive for production of vitamin D itself. Recently, a Br2 "excimer" lamp has been demonstrated. With its emission maximum at 292 nm and its efficiency it is probably ideal for vitamin D production.