Solar Photolysis of Ozone to Singlet D Oxygen Atoms, O(1-D) (Michigan).

Abstract

Ground level solar photolysis rate coefficients (jO(,3)) were measured for the photolysis of ozone by sunlight, {O(,3) + h(nu)(< 320nm) (--->) O(,2) + O(('1)D)}. The O(('1)D) atoms produced react with nitrous oxide (N(,2)O) carrier gas to form higher oxides of nitrogen (NOx). Computer model predictions show that these are mainly N(,2)O(,5) and NO(,3). These oxides (NOx) were detected by mixing with methanol and determining the increase in electrical conductivity with a continuous flow dual conductivity cell. The difference between the methanol conductivity before and after gas contact is a measure of the amount of NOx produced, and hence a measure of the amount of O(('1)D) generated from ozone photolysis. Seventy five days of data were taken during the summer of 1983, at Ann Arbor, Michigan, and are presented in the appendix. Over 390 clear air jO(,3) values are correlated with effective ozone column densities, and 500 nm aerosol optical depths. Higher atmospheric aerosol content decreases jO(,3) values for a given effective ozone column density. The solar direct beam component of ozone photolysis was measured for three different aerosol optical depths, over two entire days from sun-up to sun-down. For 500 nm aerosol depths of 0.345, 0.377 and 0.493, the maximum % direct jO(,3) component during the day was 45%, 30% and 22%, respectively. Peak hourly average jO(,3) values near solar noon (12:00-13:00, EST) for the months of May, June and July are 2.64 x 10('-5), 2.80 x 10('-5) and 2.74 x 10('-5) sec('-1), respectively. The largest diurnally averaged values for each of the months above are 1.51 x 10('-5), 1.74 x 10('-5) and 1.59 x 10('-5) sec('-1), respectively. Ozone column densities ranged from 0.300 atm-cm to 0.390 atm-cm. Temperature dependence of jO(,3) was measured from 10(DEGREES)C to 39(DEGREES)C with good agreement to models. Comparisons of jO(,3) versus global and ultraviolet radiation are made under various ozone column densities and aerosol optical depths. Second-degree polynomial equations are least-squares fitted to this data with good correlations. jNO(,2) coefficients, derived from ultraviolet radiation data, are compared with jO(,3) coefficients and show a nonlinear relationship at low aerosol depths, and a more linear relationship under cloudy conditions. A jO(,3)-photometer was built using an interference filter to pass only ozone photolyzing light. A linear regression analysis of 2234 jO(,3) versus jO(,3)-photometer data points correlated well, having a correlation coefficient (r) of 0.9836. Improvements to instrumental parts are shown for balloon and aircraft flyable payloads.