An inquiry from a reader got me to look up these two safety letters from 1989, and I thought I'd post them in case they're useful to anyone working with iodine compounds now:

Iodosobenzene salt explosion (July 24, 1989, page 2):

Iodosobenzene tetrafluoroborate was described in 1988 by V. V.. Zhdankin et al. [Tetrahedron Lett., 29, 3717 (1988)] as the first stable hypervalent iodine reagent without a nucleophilic ligand. After having prepared a few grams of the product according to the protocol in the cited literature without any difficulties, we decided to undertake a 0.5 mol preparation in the same manner. After drying the yellow crude product (about 75 g) in vacuo at room temperature for about 20 hours, it exploded while still in the flask in vacuo, producing a considerable amount of pungent fumes. The blast from the explosion was such that fragments of the flask penetrated glassware standing nearby as well as a security shield located in front of the flask. Fortunately, no one was in the laboratory at the time of the explosion. We are unable to say whether the explosion was caused by impurities or by the instability of the product itself. However we would like to warn anyone preparing or handling iodosobenzene tetrafluoroborate to be extremely careful.

Gert van Look
Fluka Chemie A.G.
Buchs, Switzerland

Iodonium salt explosions (Aug. 21, 1989, page 4):

The current renaissance in polycoordinate iodine chemistry and in particular iodonium salt species, along with the recent report by G. v. Look (C&EN, July 24, page 2) on the explosion of iodosobenzene tetrafluoroborate, prompts us to report our own experiences with these compounds. We have also had an explosion (fortunately with no injury and only very minor damage) in the attempted preparation of an iodonium perchlorate, specifically PhC=E2=89=A1C-I+-Ph=E2=80=A2ClO4-. In contrast, we have prepared numerous, diverse iodonium triflates and tosylates (up to 25 g scale), including PhC=E2=89=A1C=E2=80"I+-Ph=E2=80=A2-OTf(-OTs), and to date experienced no difficulties in the formation, handling, and subsequent transformations of these salts.

It is generally recognized that perchlorates tend to be explosive, whereas fluoroborates are much less explosive, and triflates are considered even more stable. Hence, we suggest the substitution of triflate (CF3SO3-) for ClO4- in all cases where a nonnucleophilic counterion is desired and perhaps even in some cases as substitution for BF4-. Even with triflates, ordinary caution should be exercised in the handling of all potentially explosive compounds-including the above and related polycoordinated iodine species.

Peter J. Stang
Professor of Chemistry
University of Utah