Reconstruction of the Microcanonical Rate Constant from Experimental Thermal Data
Pressure-dependent reactions are ubiquitous for gas-phase chemical processes like combustion, atmospheric chemistry, and chemical vapor deposition. For proper modeling and better understanding of these processes, their pressure-dependent rate constants k(T, p) need to be known or estimated. Apart from direct experimental measurements, k(T, p) is obtained from microscopic balancing employing the Master Equation (ME) approach, using microcanonical rate constants k(E) as indispensable ingredient. The state-of-the-art k(E) prediction is based on the Rice–Ramsperger–Kassel–Marcus (RRKM) theory. This method, however, is limited to small molecules and requires the explicit knowledge of the transition state geometry. An alternative approach for k(E) computation is the inverse Laplace Transform (ILT). This method applies to arbitrarily sized molecules and arbitrarily complex electronic structures. According to the general perception, these advantages are derogated by the low quality of k(E) data obtained with the ILT approach. This lousy reputation of the ILT approach for k(E) reconstruction stems from the method’s high sensitivity to the input data quality. As a result, practically useless microcanonical kinetic data are obtained after ILT when inaccurate or incomplete microcanonical kinetic data are provided. To redeem the reputation of k(E) reconstruction via ILT, we studied various input data formats and their impact on the quality of k(E) reconstruction. Our results sugest a new method for computing k(E) from well-established thermodynamic (NASA polynomials) and kinetic data formats (modified Arrhenius equations). The computed k(E) could be further used in ME simulations to get an accurate and reliable estimation of k(T, p).