The steady-state approximation: Fact or fiction?

The kinetic equations for an 81-reaction model of a photochemical smog chamber have been solved using a complete numerical integration as well as a quasi-steady-state approximation (QSSA) procedure. The two sets of results differ markedly in their prediction of experimentally significant factors such as the hydrocarbon depletion rate and the ozone and (NO)x peaking times. The sources of the discrepancy are traced to the fact that the assumed steady-state conditions were not satisfied, leading to errors in the concentrations of intermediate radicals which in turn affect critical rates in the reaction model. The occurrence of such discrepancies in various types of reaction models, and with different QSSA strategies, is discussed, and it is concluded that the extent of such errors in QSSA calculations cannot be reliably predicted. Their impact on conclusions regarding reaction mechanisms and rate constants can surpass uncertainties in experimental data; conversely the credibility of predictions derived through QSSA calculations becomes highly suspect. Since recently devloped methods for complete numerical integration of systems of kinetic equations are now available, it is recommended that these be adopted in future work, and that the use of QSSA be abandoned.