We report on a new instrument developed for rapid automated on-line and continuous measurement of ambient aerosol bulk composition. The general approach is based on earlier devices (Khlystov et al. 1995; Simon and Dasgupta 1995) in which ambient particles are mixed with saturated water vapor to produce droplets easily collected by inertial techniques. The resulting liquid stream is analyzed with an ion chromatograph to quantitatively measure the bulk aerosol ionic components. In this instrument, a modi ed version of a particle size magni er (Okuyama et al. 1984) is employed to activate and grow particles comprising the ne aerosol mass. A single jet inertial impactor is used to collect the droplets onto a vertical glass plate that is continually washed with a constant water diluent ow of nominally 0.10 ml min ¡ 1. The ow is divided and then analyzed by a dual channel ion chromatograph. In its current form, 4.3 min integrated samples were measured every 7 min. The instrument provides bulk composition measurements with a detection limit of approximately 0.1 ¹g m ¡ 3 for chloride, nitrate, sulfate, sodium, ammonium, calcium, and potassium. Results obtained during its deployment in the August 1999 Atlanta EPA Supersite study are presented. BACKGROUND Process studies aimed at understanding sources, atmospheric transformation, and fate of ambient aerosols require knowledge of particle chemical composition. To date, quantitative composition measurements are typically performed off-line on particles collected onto substrates by ltration or inertial impaction. The collected sample is then manually extracted and analyzed. For example, measurements of ionic aerosol components involve collection on denuder-lter pack assemblies, extraction of the collected aerosol into water, and analysis of

Trace gas measurements pertinent to understanding the transport and photochemical formation of 03 were made at a surface site in rural Georgia as part of the Southern Oxidant Study during the summer of 1991. It was found that there was a strong correlation between 03 and the oxidation products of NOx: O3(ppb) = 27 + 2 11.4 (NOy(ppb) -NOx(ppb)), r = 0.78. This fit is similar to that observed at other rural sites in eastern North America and indicates a nominal background 03 level of 27 ppb; values higher than 27 ppb are due to photochemical production in the recent past, which varied from near zero to -•50 ppb. The origin of the 03 above background was investigated by using a free radical budget equation to calculate an in situ 03 production rate in terms of measured concentrations of NO and free radical precursors (03, HCHO, peroxides, and other carbonyls). A comparison of observed and predicted diurnal trends in 0 3 indicates significant 03 production in the afternoon at a time when 03 concentration is either steady or decreasing. The afternoon near-surface layer is thereby a source region for 03 which can be exported. In situ production accounts for approximately one half of the morning increase in 03 concentration on days with high 03; the remainder is due to entrainment of dirty air aloft by the growing convective boundary layer. Additional evidence for the role of vertical transport in controlling the hour-to-hour changes in 03 is found in the diurnal cycles of SO2 and HNO3 which also have rapid increases in the morning. The day-to-day variability of 03 was investigated using a back trajectory model. NOy concentration at the measurement site could be reasonably accounted for by considering NOx emission sources located within 1-day transport distance. In as much as there is a strong correlation between 03 and NOy, the coincidence between trajectory location and NOx emission sources appears to t•e an important factor influencing midday 03 concentration. Hydrocarbon measurements are consistent with NOx being the limiting factor for formation of 03. 20-30 ppb. During pollution episodes, 03 levels in excess ofthe 120 ppb National Ambient Air Quality Standard have been measured at rural sites [Meagher et al.southeastern United States differs from more industrialized and populated regions in that NOx emissions are lower and natural HC emissions are higher. In addition to precursor emission rates the formation of 03 depends on meteorological conditions. An active photochemistry is favored by high solar intensity, temperature, and absolute humidity which are common summertime conditions in the southeastern United States. Stagnation episodes, which are also common, allow emitted pollutants and their photochemically produced reaction products to accumulate over a several day period. Considerable progress has been made in using photochemical models to simulate the production of 03 and the effects of emission changes [e.g., Seinfeld, 1988; McKeen et al., 1991a, b; NRC, 1991; Roselle et al., 1991]. However, the coupled em...

[1] Five new instruments for semicontinuous measurements of fine particle (PM2.5) nitrate and sulfate were deployed in the Atlanta Supersite Experiment during an intensive study in August 1999. The instruments measured bulk aerosol chemical composition at rates ranging from every 5 min to once per hour. The techniques included a filter sampling system with automated water extraction and online ion chromatographic (IC) analysis, two systems that directly collected particles into water for IC analysis, and two techniques that converted aerosol nitrate or sulfate either catalytically or by flash vaporization to gaseous products that were measured with gas analyzers. During the one-month study, 15-min integrated nitrate concentrations were low, ranging from about 0.1 to 3.5 mg m À3 with a mean value of 0.5 mg m À3 . Ten-minute integrated sulfate concentrations varied between 0.3 and 40 mg m À3 with a mean of 14 mg m À3 . By the end of the one-month study most instruments were in close agreement, with r-squared values between instrument pairs typically ranging from 0.7 to 0.94. Based on comparison between individual semicontinuous devices and 24-hour integrated filter measurements, most instruments were within 20-30% for nitrate ($0.1-0.2 mg m À3 ) and 10-15% for sulfate (1-2 mg m À3 ). Within 95% confidence intervals, linear regression fits suggest that no biases existed between the semicontinuous techniques and the 24-hour integrated filter measurements of nitrate and sulfate;, however, for nitrate, the semicontinuous intercomparisons showed significantly less variability than intercomparisons amongst the 24-hour integrated filters.

Formation constants have been determined for the mixed-metal complexes between indium(III) and uranium(VI) with malic, citric, and tartaric acids. In the cases of malic and tartaric acids the equilibrium can be written as U022 + + In3 + + 2HaL = U02InL2~+ 6H+. The logarithms of the equilibrium constants are -7.62 for the malate system and -7.14 for the tartrate.In the case of citric acid the equilibrium is U022+ + In3+ + 2H,L = U02InL23_ + 8H + where two of the protons come from the hydroxy groups. The logarithm of the constant is -11.58. Infrared measurements on trilaurylamine extracts of the mixed-metal complexes have been made and are discussed. The dimerization constant for the formation of (uranyl citrate)2 was determined poten tiometrically and found to agree with the literature values. However, arguments are presented for the necessity of recalculating the values presented in the literature for the equilibrium U022+ + H8L = U02L~+ 3H+. The logarithms of the recalculated values are -7.40 for malate, -6.30 for citrate, and -6.85 for tartrate.

No abstract