I'm working with trace amounts (say, 1 microgram) of sterols such as dinosterol, stigmasterol, and cholesterol, and have been trying to acetylate with acetic anhydride and pyradine. This is a standard procedure our lab has performed for years. Transfer the sterol to a vial insert with toluene, evaporate, add 20 ul of anhydride and 20 ul of pyridine, heat to 70C for half an hour, cool, evaporate reagents. For several months now, though, the sterol acetate yield from this step has been 20-30%, instead of our typical 90-100.

The confusing part is that it appears that sterol is being destroyed/converted, not just failing to react. When I acetylate a sterol standard and get 30% yield of the acetate, I don't see un-derivitized sterol in my chromatogram. Furthermore, if I sylilate the residue with BSTFA/TMCS, I do NOT recover the un-acetylated sterol as a TMS ether, either. If I sylilate an un-reacted sterol standard, however, I get 100% of what I expect as the TMS ether (so I know it's not a problem with my standard being more dilute than it should be). This does NOT happen with n-alkyl alcohols. nC21-OH, for example, acetylates with 100% yield just fine.

I have tried new containers of acetic anhydride and pyridine. I have tried diluting the mix with toluene and/or changing the ratios of anhydride/base. I have tried changing glassware, or using teflon. I have tried purging the reaction vial with nitrogen and argon. I have tried substituting triethylamine for the pyridine. I have tried water/DCM phase separations followed by drying of Na2SO4 as a workup instead of simply evaporating the reaction mixture. I have tried pre-deactivating my glassware. I have tried letting the reaction proceed at room temperature overnight. I have tried leaving the reaction at 70C overnight. I have tried a different method that uses toluene, acetic anhydride, and sodium bicarbonate as a sparingly-soluble base that keeps the byproduct acetic acid quenched. I've compared results on a GC-FID with a PTV inlet, a GC-MS with a split/splitless, and someone elses GC-FID with split/splitless in another lab.

Nothing has worked.

Does anyone, in the name of all that is good, have any idea where my sterol might be going? Or how to proceed to fix this? Simple anhydride:pyridine at 1:1 was always effective in the past, and the lack of un-reacted sterol tells me it's not a reaction rate problem that a catalyst will solve.

Regardless, I currently have test reactions running that include DMAP and n-methylimidazole as catalysts. I haven't worked them up yet, but for former turned deep yellow and the latter turned deep red. The reactions in which I used TEA as the base also turned yellow, and left behind a large amount of oily residue. Are these warning signs or telling indicators of some sort of oxidative problem?

In a couple of days I will have materials for methods using silver triflate or montmorillanite clay as catalysts, but I have no reason to suspect these will go any differently.

Edit: Edited to indicate that, in contrast, straight chain saturated alcohols DON'T seem to have any problem acetylating with 100% yield under normal conditions.