Assessment of Semi-Empirical Soot Modelling in Turbulent Buoyant Pool Fires from Various Fuels
Round 1
Reviewer 1 Report
The paper requires significant revisions in various sections. Notably, the description of the experimental work used to validate the model is missing. The literature review lacks depth and contains unrelated points. The numerical simulation inputs are inadequately presented, with no cell sensitivity analysis. Additionally, the numerical models were directly copied from the FDS technical guide, rendering their inclusion unnecessary. The main methodology of the study is unclear, and the conclusions read more like a summary rather than the conclusions of a scientific paper. This overall assessment suggests that the paper has potential, but major revisions are necessary to address the aforementioned issues. Furthermore, the following additional points should be considered:
-
Writing Style Enhancement: The paper would benefit from improved writing throughout its content.
-
Abstract Clarification: The abstract should be revised to provide clearer information about the experimental data utilized and the key contributions made by this study.
-
Definition of CFD and FDS: The authors should provide explicit explanations of the abbreviations CFD (Computational Fluid Dynamics) and FDS (Fire Dynamics Simulator) to ensure clarity for readers.
-
Introduction: The introduction should focus on presenting the methodology employed, rather than discussing literature on jet flames, as the study utilizes FDS, which is primarily designed for buoyancy-driven fluids.
-
Numerical Modelling: The presentation of the smagorinsky similarity assumption and other details sourced from the FDS user guide is unnecessary and can be omitted.
-
Figure 1: To maintain consistency with the English language used throughout the manuscript, please replace the French words in Figure 1 with their English equivalents.
-
Cell Size Justification: The authors mention choosing the appropriate domain and cell sizes but fail to provide a cell sensitivity analysis. A proper justification for the selected cell size should be presented.
-
Experimental Setup: The paper lacks information regarding the experimental setup employed. The authors should include pertinent details to provide context for the model validation.
-
Figure 7: In Figure 7, "kW" should be used instead of "Kw" to represent Heat Release Rate. Furthermore, the authors should clarify the difference between the red and black lines in the figure.
-
Reference Formatting: The authors should ensure that the font and size of the references are consistent with the rest of the paper.
In summary, this journal paper review highlights several areas that require significant revisions and improvements. The authors should address the mentioned points and carefully revise the manuscript to enhance its overall quality and scientific rigor.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper is aimed at comparing two soot models in FDS for the simulation of turbulent jet flames/pool fires. This topic is not new as there is a substantial amount of work on this that has already been published and the semi-empirical soot model discussed in this paper was developed by other researchers. One could argue that the latest version of FDS is used in this work. However, my main concern is that the soot model presented is only applicable for laminar flames, which may be fine for the smaller-scale turbulent flames, but should certainly not be directly used for turbulent pool fires – some sort of averaging (flamelet or conditional moment closure) is needed, since the temperature in the turbulent pool fire will be lower than that in a laminar flame, which would result in the soot formation and oxidation rates incorrectly calculated. Other minor suggestions, 2D contour of soot formation and oxidation rates would be useful. For modelling the evaporation of the liquid fuel, the thermal properties, absorption coefficient and heat of vaporization used in the model should be specified.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
In this paper, the authors implemented 2 semi-empirical models for soot formation and oxidation in NIST Fire Dynamics Simulator software and compared the predictions against experimental data for methane, ethylene and n-heptane. The authors should justify better the novelty of this investigation, since semi-empirical soot models are widely available in commercial CFD software. Moreover, the models tested in this investigation are not sophisticated enough to properly predict soot (especially for large liquid hydrocarbons, see Figure 10 of the paper).
Specific comments from the reviewer:
11) Please, avoid using acronyms in the abstract, such as HRR.
22) The introduction needs to highlight previous efforts from other researchers to model soot in fire dynamics simulations.
33) What is the advantage of a simple soot model such as the one used in this paper compared to a comprehensive soot model (Particulate Mimic model or Particulate Size Mimic model) based on detailed PAH chemistry? Results have shown repeatedly that simple semi-empirical soot models fail when applied at conditions different from those used in the calibration.
44) Mesh quality in LES is typically characterized by the fraction of the turbulence that is resolved instead of modeled. Note that, due to the intrinsic variability of LES, grid size dependency studies are not typically performed in LES (two LES cases with exactly the same initial and boundary conditions will give different results due to the stochastic and chaotic nature of turbulence). What fraction of the turbulence are the authors solving in their simulations?
55) In Figure 3. Why the soot model works well for 3D and 4D, works reasonably well for 1D, but it fails for 2D? Is there anything special about height 2D that may be causing the deviation? Is the experimental uncertainty known?
66) Figure 6. Why are the results in FLUENT not included? The Conditional Source-term Estimation method also uses a semi-empirical model for soot. Did the authors consider a detailed soot chemistry model? This is especially important for the n-heptane simulations, where the LDP model and soot yield approach are not able to reproduce the experimental data.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The paper has been improved. I don't agree with the comment that most of previous studies are for 2-D cases. I appreciate the difficulties to implement additional models in limited time, but some discussions should be provided on the potential uncertainties/limitations in using these models for turbulent or large scale fires.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors have addressed all my questions, comments and concerns and I recommend the paper for publication in this journal.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
