The hydroxyl radical, OH, is the most important oxidant in tropospheric chemistry. Measurement of atmospheric hydroxyl radicals is extremely difficult due to the low concentration and short lifetime of this species. These same properties mean that OH concentrations are indicative of local conditions and measurements are therefore extremely sought-after by atmospheric chemists. A novel method for monitoring tropospheric hydroxyl radicals was examined that, unlike other methods, was reported to be inexpensive, portable and relatively simple.

Salicylic acid reacts with atmospheric OH to produce the extremely fluorescent 2,5-dihydroxy benzoic acid, 2,5-DHBA. High Performance Liquid Chromatography then separates the post reaction solution and the products are quantified using fluorescence detection. Since 2,5-DHBA is extremely fluorescent, typical daytime OH concentrations of 0.04 - 0.4 parts per trillion can be measured within a 30 minute integration time.

A laboratory investigation into optimum sampling and analysis conditions showed that the method was analytically feasible. Ambient measurements, however, revealed interferences from light, ozone and bacteria, which produced a signal much larger than that expected for ambient OH. The analytical technique was also found to be inadequate for field measurements. Systematic identification, characterisation and elimination of each of these interferences, coupled with optimisation of the analytical technique, drastically reduced this signal. An interference was however still observed during the SONTAS 2000 field campaign. Tests showed that this signal was probably not due to ozone, HO2, NOx, hydrogen peroxide, aerosols, light, bacteria or the analytical method. Laboratory studies following this campaign suggest that the interference could be due to methyl peroxy radicals. A simulation of the experiment that produced CH3O2 radicals showed that this was the only oxidant present in sufficient concentrations to account for the observed signal. Although further work would be required to confirm this interference, these results illustrate the non-selectivity of the method. The effect of many other components in the atmosphere, both individual and combined, must therefore be tested before the method can be used reliably in the field. The validity of previous reports of ambient hydroxyl measurements using this technique is therefore brought into question.