The nitric acid shootout: Field comparison of measurement methods

Abstract

Eighteen instruments for measuring atmospheric concentrations of nitric acid were compared in an eight day field study at Pomona College, situated in the eastern portion of the Los Angeles Basin, in September 1985. The study design included collocated and separated duplicate samplers, and the analysis by each laboratory of a set of quality assurance filters, so that the experimental variability could be distinguished from differences due to measurement methods. For all sampling periods, the values for nitric acid concentrations reported by the different instruments vary by as much as a factor of four. The differences among measurement techniques increase with nitric acid loading, corresponding to a coefficient of variation of 40%. In contrast, samplers of the same design operated by the same group show variability of 11–27 %. Overall, the highest reported concentrations are observed with the filter packs and lower concentrations are observed by the annular denuders and tunable diode laser absorption spectrometers. When the nitric acid concentrations are high enough to be detected by the FTIR, the FTIR values are close to those obtained by the denuder difference method and to the mean value from the other sampler groups. In the absence of a reference standard for the entire study, measurement methods are compared to the average of four denuder difference method samplers (DDM). Filter pack samplers are higher than the DDM for both daytime and night-time sampling. Two different filter packs using Teflon® prefilters are higher than the DDM by factors of 1.25 and 1.4. The results from the three annular denuders do not agree; the ratios of means to the DDM value are 1.0,0.8 and 0.6. For the transition flow reactor method and for two dichotomous samplers operated as denuder difference samplers, the ratio of means to the DDM are 1.09 and 0.93, respectively. The tunable diode laser absorption spectrometers gave lower daytime and higher night-time readings compared to the DDM, especially during the last three days of the study. Averaged over the entire measurement period, the daytime ratio of TDLAS to DDM is 0.8 and the night-time ratio is 1.7.