Rate of the Reaction of S(3p^{4 I}D₂) with OCS

Abstract

THE difference in both rate and mode of chemical reaction shown by the first electronically excited state of the group VI elements (that is, … np⁴(¹D²)), relative to that of their ground electronic state (… np⁴(³P_J)), has been of considerable interest^1–4. Furthermore, recent observation of the more highly excited singlet state of the sulphur atom S(… 3p⁴(¹S₀)) (S(3¹S₀) – S(3¹D₂ = 1.61 eV; S(3¹D₂) – S(3³P₂) = 1.15 eV) by kinetic absorption spectroscopy in the vacuum ultraviolet has permitted a comparison of the chemical behaviour of this state with that of the lower states². The S(3¹S₀) atoms were generated by vacuum ultraviolet flash photolysis of OCS in the presence of a large excess of the inert gas argon (pOCS = 10.6 Nm⁻²; p_Ar = 2.66 kNm⁻²). S(3¹D₂) atoms were not observed in this system, however, despite compelling evidence for their initial production². No quantitative data have been given for the efficiency of quenching of S(3¹D₂) by the inert gases, although such data are available for the analogous state of oxygen. Some measure of disagreement concerning the absolute rates for quenching of O(2¹D₂) appears to exist, however, and the most recent absolute estimates^4,5 imply that quenching occurs with an efficiency of one collision in 200 for argon. If quenching of S(3¹D₂) by argon occurs with a similar efficiency, then one reason for the failure to observe this state in the vacuum ultraviolet photolysis of OCS² is clear. We have therefore repeated the earlier experiments replacing argon by helium as the inert diluent gas, for helium is known to be particularly inefficient in quenching O(2¹D₂) atoms^3–5. The kinetics of the S(3¹S₀) atoms are unaffected by this change, in agreement with the previous conclusions², but the strong spectrum of S(3³P_J) observed in the presence of argon at short delay is virtually absent in the presence of helium. Furthermore, the rapid growth of a strong spectrum of S₂(g¹Δ_u←a¹Δ_g) is now seen, whereas in the presence of argon this spectrum was weak². At longer delays, the S₂(a¹Δ_g) state is apparently quenched to yield S₂(X³σ_g⁻) (Fig. 1). These observations are in agreement with the mechanism proposed by Strausz and Gunning et al.^>1,6. In the presence of helium, k₂ appears to be negligible; about twenty per cent of the sulphur atoms are observed in the S(3³P₂) state, which may represent the fraction formed in the primary photochemical process. Thus, provided k₃[M]≪k₁[OCS], the rate constant for reaction into the quantum state S₂(a¹Δ_g) may be obtained by monitoring the growth of the spectrum S₂(g¹Δ_u←a¹Δ_g). The magnitude of k₃[M] was derived from the decay of S₂(a¹Δ_g) at longer times up to 100 µs. A value of k₁[OCS] was then obtained by curve fitting the integrated form of the differential equation for the formation of S₂(a¹Δ_g), comprising terms due to reactions (1) and (3) and an analytical form for the photoflash output, with the observed intensity variation of the transition S₂(g¹Δ_u←a¹Δ_g). It was found that the experimental points gave the best fit for a value of k₁ > 4 × 10¹³ ml. mole⁻¹ s⁻¹ which corresponds to reaction for less than one collision in four. A lower limit for this rate constant has previously been given as 4 × 10¹¹ ml. mole⁻¹ s⁻¹ (ref. 6). Thus the rate constant for the reaction of S(3¹S₀) with OCS is significantly lower; it has been shown by direct observation of this state using kinetic absorption spectroscopy² to be 6 × 10¹² ml. mole⁻¹ s⁻¹.