I've just stumbled at Ushio on a
UV irradiation panel that uses Led instead of gas discharge lamps:
http://www.ushio.co.jp/en/products/light_source/led/index.htmlthis can't be very old, since UV leds are rather recent, especially at 365nm. The panel delivers ~
60W light at 365, 385, 395 or 405nm (as much as an excimer lamp drawing 800W). It must still be expensive.
Sold for photolithography presently, but if Ushio find enough customers, they will probably sell it for chemical reactors as well. Do not underestimate the capacity of the semiconductor industry to improve performances and collapse prices: <5 years passed between GaN laser diodes and putting one in each Dvd burner.
Presently GaN at 405nm is banal, and many-watt sources are in catalogues. It's the first time I see many-watt sources below 405nm in a catalogue, which implies AlGaN is production-ready at least at Ushio. Since semiconductor processes can adjust the emission wavelength, they will target specifically bromides, iodides and so on. It can become a nice synthesis tool.