摘要：Odors from wastewater treatment plants (WWTPs) have frequently been attributed primarily to hydrogen sulfide (H2S). Low-to-medium cost hydrogen sulfide sensors have been utilized as odor indicators. However, other odorous species are usually present that may have lower odor thresholds than hydrogen sulfide. Hydrogen sulfide is not always present in odorous environments and the correlation of hydrogen sulfide to odor at a treatment facility is inconsistent. Such factors determine hydrogen sulfide an inconsistent indicator and more sophisticated measurement techniques are required to accurately predict odor intensity from complex gaseous mixes. In this paper, the performance of a direct mass spectrometric technique, selected ion flow tube mass spectrometry (SIFT-MS), is evaluated for analysis of odors from diverse sources at a modern WWTP. The soft chemical ionization employed in SIFT-MS provides detection and quantification of a wide range of potential odorants to below, or close to, the human odor detection threshold (ODT). The results presented demonstrate that methyl mercaptan is almost always a more significant odorant at this WWTP than hydrogen sulfide and confirm that the relative abundances of these odorants vary significantly. Parallel SIFT-MS chemical analysis and human sensory analysis (olfactometry) was conducted in this study. Good agreement was observed for samples of moderate to strong “sewage” or “chemical” character. However, in samples that were otherwise low in odor intensity, sensory analysis did not attribute “sewage” odor notes as the predominant odor character. Chemicals attributed with this odor character were present significantly above the ODTs in the mixed samples and were detected by SIFT-MS. A weak correlation was obtained between total odor activity values measured using SIFT-MS and the odor concentration (in odor units per cubic meter) determined using dilution olfactometry. The complexity of the wastewater matrix and complexity of human odor recognition from mixed samples is thought to be the underlying cause of less-than-ideal correlation, perturbing both olfactometry and SIFT-MS analyses.
