Particles are formed within the plumes from industrial plants from condensation of the oxidation products of SO₂ and other compounds. A quantitative understanding of the relationship between emissions of gas-phase particle precursors and downwind aerosol properties is needed for mitigation efforts.

During the Southern Oxidants Study 1999 and the Texas 2000 Air Quality Study, airborne measurements of particle size distributions were made in a variety of power plant, petrochemical, and urban plumes. Data from a 5-channel condensation condensation counter, a laser optical particle counter, and a white light optical particle counter were combined to determine particle size distributions from 0.004-10 µm diameter with 1 second resolution.

In all plumes from power plant and petrochemical industry sources, substantial enhancements in particle mass were found in only those plumes rich in SO₂. Even in the plumes of petrochemical complexes with large emissions of NO_x, propene and ethene near Houston, Texas, particle enhancements were not detected in the absence of SO₂. Nearby, modest sources of SO₂ produced detectable enhancements in particle volume.

In the absence of in-cloud oxidation, particulate sulfate is produced by condensation of the oxidation products of SO₂ and OH. Gas-phase HNO₃ is produced by the oxidation of NO₂ by OH; thus the ratio of HNO₃ to total reactive nitrogen, NO_y, indicates approximate total parcel exposure to OH. Assuming a composition of 50% (NH₄)₂SO₄, the fraction of plume S found in the particulate phase was calculated as function of HNO₃/NOy and compared with a 2-dimensional numerical model of plume chemistry and dynamics. In the plume of the Parish power plant near Houston, the increase in plume particulate S fraction with increasing HNO₃/NO_y was well simulated by the model. In the merged industry plume from the Houston ship channel, the apparent increase in the fraction of plume S in the particulate phase was underpredicted by the model. This difference may be due to model formulation errors, unaccounted SO₂ losses, or to particle growth by nonsulfate species. Similar analyses on other days produce very similar results for both the Parish power plant and the coalesced ship channel plumes. These observations point to the oxidation products of SO₂ as the primary compounds controlling particulate mass formation from the studied industrial sources. Condensation of other species, particularly semi-volatile organics from petrochemical sources, is not excluded by this analysis, but does not appear to occur in the absence of substantial SO₂ oxidation.