12/22/2006

oxygen--part two


Back to something I started a while ago and didn't finish: how to generate singlet oxygen. Probably the best-known means is photosensitization--use a dye (Rose Bengal, Methylene Blue, etc.) and a lamp. The dye is photoexcited and transfers its energy to oxygen, giving the singlet.[1] While this can involve some pretty cool-looking lamps (see left), it doesn't have to--you can do it Dylan's way with a lamp from Home Depot. (This group has even managed to do it without solvent!) Photochemical reactions can have some scale-up issues, though, despite being super-cool.

Alternatively, you can use peroxide and things involving metals (ew, metals, ick). Molybdates, peroxotungstates and La(OH)3 seem to work alright.[2] CaO2 appears to be the most popular, but it sounds kinda tricky to make. (Anyone done this?)

And then there's a fun little reaction I did a while ago.[3] Uses a weird little hypervalent iodine compound--[Bis(trifluoroacetoxy)iodo]benzene, oft-abbreviated as "PIFA"--and catalytic H2O2 to generate singlet oxygen.

This was dead easy to set up and about the only thing I've done that's given a respectable yield. If a knuckle-dragging undergrad like me can manage not to screw this up, just about anyone can. As an added bonus, the benzofuran I started with was the color of yellow highlighter ink. (No picture, though. Sorry!)




[1] You really don't want to get me started on energy transfer, especially since I am already very excitable. Maybe a nice long post will appear about it in the future.

[2] Found a review on this a while back. Fairly short & sweet. Have fun.

[3] Here is the paper! Go try it! (Especially if it's a bad day and nothing else is working. This will make you feel better.)

2 comments:

Excimer said...

PIFA looks slightly familiar to another well-known hypervalent iodine species, the Dess-Martin periodinane, which is an excellent, mild oxidant- and one of my personal favorite reagents.

Ψ*Ψ said...

That's what I was thinking too. I've never used the Dess-Martin reagent, though.