Diazo and Azides - Hazards

Barry Sharpless's recent paper on "Click Chemistry" contains a footnote on "azidophobia" which is summarised below. "Azidophobia" seems to be largely based on the fact that a number of metallic azides are shock sensitive. However, certain organic azides, especially small ones, are also explosive. The following points are noteworthy. Sodium azide is relatively safe, especially in aqueous solution, unless acidified to form HN₃, which is volatile and highly toxic. For organic azides, the "rule of six" is very useful: six carbons (or other atoms of about the same size) per energetic functional group (azide, diazo, nitro, etc.) provides sufficient dilution to render the compound relatively safe. Note that the presence of acetylenic groups or other energy producers with an azide makes for increased hazard. Decomposition of organic azides can be catalysed by certain transition metal species (especially those from the Fe and Co triads) and by strong acids (that is, the same catalysts that are most effective for decomposition of organic peroxides). Azide groups attached directly to olefinic, aromatic, or carbonyl moieties are much less stable and, other things being equal, more dangerous than aliphatic azides (the approximate activation energies for unimolecular loss of N₂ being 29 and 49 kcal mol^-1, respectively). For these reasons, azides should not be distilled or treated in a careless fashion. However, when common sense is employed, they can be prepared, stored, and used without risk in the standard organic chemistry laboratory: We have never experienced a safety problem with these materials. See: M. Peer, Spec. Chem, 1998, 18, 256-263."
Of course process chemists should always consult the latest edition of Bretherick's Handbook of Reactive Chemical Hazards (Butterworth Heinemann) for information on individual azides. Kevin Cardwell of GlaxoSmithKline presented an interesting example of azide chemistry at Scientific Update's Chiral Chemistry symposium in Malta 2000. A synthetic route via the open chain azide proved potentially hazardous for scale up since the product has an onset of decomposition at 45^oC, whereas the reaction had an adiabatic temperature rise of 65^oC. The alternative pathway via the cyclic bromide and azide displacement, or using a telescoped approach, were much safer.