A study of heterogeneous oxidation of SO2in summer clouds

Abstract

Experimental determination of in‐cloud oxidation of SO₂has been lacking because of an inability to distinguish SO₄in cloud water, derived from aerosol scavenging (i.e., produced from homogeneous gas phase oxidation), from that due to in‐cloud SO₂oxidation, (SO₄)_in. A tracer technique has been developed in this laboratory that uses trace elements (M) to resolve the two SO₄components. Extensive data are reported in this paper on the use of Se as a tracer. Preliminary results are also included to test As and Sb as potential tracers. The technique is based on simultaneous determination of SO₄and M in cloud water (cw) and in out‐of‐cloud air (aa), and the knowledge of the relative scavenging efficiencies α and β of SO₄and M‐bearing aerosols, respectively: (SO₄)_in= [(SO₄/M)_cw‐ α/β (SO₄/M)_aa] (M)_cw. Data from two events are presented which yield the scavenging ratio of SO₄/Se as 1.04±0.29 and 1.04±0.19. The technique has been used to quantitatively determine in situ SO₂oxidation in six summer cloud systems during 1987–1989 at Whiteface Mountain, New York. Initially, cloud water samples were collected hourly, but in later sampling campaigns this interval was reduced to 15 min to better delineate the relationship between chemistry and meteorology in clouds. Aerosol collection was also reduced from 12 to 2 hours. Five cloud systems with pH ranging from 2.8 to 4.0 processed air which, based on our understanding of elemental signatures, had passed through the high‐pollution emitting areas of the midwest. In individual samples, (SO₄)_invaried from 0 to 39%. The amount of (SO₄)_indid not vary dramatically from one sample to the next. The data appeared to be consistent with a flow‐through reactor at the site through which updrafted air is passing. Where available, SO₂data in clear air and within the cloud yielded a mass balance consistent with our experimentally determined in situ SO₄formation. A sixth cloud event, a storm system which processed air largely from the north‐northwest (“clean air”), yielded pH of 4.3 to 5.0 in the cloud water. Our tracer techniques based on both Se and As revealed no in situ oxidation of SO₂. This was confirmed by our observation that SO₂concentration was low, and values in the cloud and in cloud‐free air were equal. When data from all six summer cloud systems were pooled together, (SO₄)_indecreased with increasing pH with negligible oxidation above pH 4. Also, (SO₄)_indecreased smoothly with increasing aerosol SO₄/Se ratio. Coal combustion is the primary source of S and Se in rural atmosphere in the northeast. Relative proportions of SO₂and SO₄are dependent on atmospheric oxidation. Enhancement in oxidation yields higher SO₄/Se ratio and at the same time leaves less SO₂in the air for oxidation. A simple model is developed that enables an estimation of SO₂oxidation in clouds.