C3MechV4.0 Laminar Flame Speed (LFS) reaction mechanism

[Leonardo Giordo]

Dear Cantera users and developers,

I'm trying to calculate the LFS for a surrogate bio-fuel using the new C3MechV4.0.
In the Precompiled module description is written:

This directory contains precompiled chemical kinetic sub-models of C3MechV4.0, tailored to specific fuel compositions and conditions, and the full version of C3MechV4.0. The sub-models were compiled from the sub-modules in the directory SUBMODULES/. The reactions in the sub-modules are grouped based on the number of carbon atoms in the species (C0, C1-C2, C3-C4, C5, C6, C7, C8+, C5CY, C6CY), with aromatic species being managed separately. Here, 'CY' refers to non-aromatic, cyclic fuel components. Additionally, sub-modules for dimethyl carbonate and ethylene carbonate (DMC+EC), nitrogen-containing species (N), cross-reactions between nitrogen- and carbon-containing species (C-N), and polycyclic aromatic hydrocarbons (PAH) are available. The tables below list all precompiled sub-models, where in this subdirectory "N" refers to the combined N- and C-N sub-modules if a sub-model contains carbon atoms.
Many sub-models are available as smaller high-temperature (HT) versions, which are suitable for simulations of unstretched premixed and counterflow flames, high-temperature flow reactors, and shock tubes. Low-temperature (LT) chemistry is, for example, required to simulate jet-stirred reactors, flow reactors under lower-temperature conditions, and rapid compression machine (RCM) experiments. The columns NS(HT/LT-HT) and NR(HT/LT-HT) in the tables below refer to the number of species and reactions in the respective sub-models.
Each HT and LT-HT version is assigned a unique model ID (MID(HT/LT-HT)) that encodes the specific sub-module combination; if a single model is used across all temperatures, the counts (NS and NR) and MIDs are the same. If you need a combination not listed here, see COMPILER/ directory for an easy-to-use script to create custom sub-models.

C3MechV4.0/PRECOMPILED at main · C3Mech/C3MechV4.0 · GitHub

I use a reduce version (2000 species) of the Precompiled HT and LTHT mechanisms.

The results of the HT only are generally up to 10% higher in the temperature range 600-1000 Kelvins. But being the LTHT model much more computationally demanding than the HT one I get way less data from the calculation.

Should I expect the results coming from the LTHT model to be more accurate than the HT only one? Up to what temperature can I expect the results of the HT only model to be accurate?

Kind regards,

Leonardo