Numerical Simulation of a Thermal Management System Using Composite Flame-Retardant Resin and Its Effect on Battery Life Span

Round 1

Reviewer 1 Report

The authors simulated the effect of a composite flame-retardant resin on battery life. This work is meaningful.

1.     The temperature of Li-ion cells has been simulated with different configuration. However, the difference between simulated temperature and real temperature should be discussed.

2.     Authors displayed the superiority of the flame retardant composite resin with the comparison of these models. However, it is hard to distinguish the main effect of structure and composite in this model. Whether the structure acted as the dominant factor for this enhancement.

Author Response

The authors begin by addressing our thanks for the availability and interest in reviewing the article proposed for publication. We completely agree that the comments made are necessary and certainly help to increase the scientific level of the presented topic. The text of the article was again analyzed and repaired from a grammatical and linguistic point of view (the score of the text editor is 98%). The answers to your comments are (point by point) as follows:

C: The authors simulated the effect of a composite flame-retardant resin on battery life. This work is meaningful.

1. The temperature of Li-ion cells has been simulated with different configuration. However, the difference between simulated temperature and real temperature should be discussed.

Reply:  The aim of this research is only a comparative analysis by numerical methods of the thermal conductivity performance/efficiency.

2. Authors displayed the superiority of the flame-retardant composite resin with the comparison of these models. However, it is hard to distinguish the main effect of structure and composite in this model. Whether the structure acted as the dominant factor for this enhancement.

Reply:  Polymers typically lack free electrons, relying instead on phonon transport for thermal conduction, generally resulting in low thermal conductivity coefficients. There are two primary methods for enhancing thermal conductivity in polymer composites: synthesizing intrinsic thermally conductive polymers and preparing filled thermally conductive polymers. While intrinsic polymers involve complex and costly synthesis processes, filled polymers offer advantages such as ease of fabrication, lower cost, and suitability for industrial production. By incorporating high-thermal-conductivity fillers into polymer matrices, the thermal conductivity coefficient of composites can be significantly improved. The thermal conductivity of filled polymers is influenced by many factors, such as filler type, structure, distribution, content, interface thermal resistance, and intrinsic conductivity of fillers.

Reviewer 2 Report

The article is dedicated to the effect on battery life span using a composite flame-retardant resin in a thermal management system. The title of the article does not correspond to the content. It may seem from the title that the article used natural experiments to study the life cycle of batteries, but this is only a simulation.

Reviewer comments:

In the humble opinion of the reviewer, the article is written in a poor English especially the introduction part. The sentences are replete with many additions, which makes the article difficult to follow. The article itself needs to be shortened.

The simulation of heating the battery pack was not confirmed by a natural experiment with real batteries. Such an experiment does not seem expensive or difficult. In the opinion of the reviewer, the article demonstrates more the capabilities of the SolidWorks package than research useful for sustainability. It is not clear from the article how the thermal conductivity of special resins, especially with nanocomposites, was simulated. How accurate were the models used?

Figure 1 (c)-(e) Pointer number 3 “resin” is absent.

Some abbreviations used like “4S4P type”, “BTMS” not deciphered in text.

In the humble opinion of the reviewer, the article is written in a poor English especially the introduction part. The sentences are replete with many additions, which makes the article difficult to follow. 

Author Response

The authors begin by addressing our thanks for the availability and interest in reviewing the article proposed for publication. We completely agree that the comments made are necessary and certainly help to increase the scientific level of the presented topic. The text of the article was again analyzed and repaired from a grammatical and linguistic point of view (the score of the text editor is 98%). The answers to your comments are (point by point) as follows in the document.

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors have conducted systematic research on battery thermal management modeling and heat dissipation scheme optimization, which is of great theoretical significance. However,The entire manuscript only consists of numerical simulation analysis and lacks experimental data to verify the accuracy of simulation results. The following issues should be brought to the attention of the author.

1. In the analysis of simulation results, the author provided cloud maps of the battery temperature field under different heat dissipation conditions, but did not quantitatively compare the heat dissipation effects of different schemes.

2. In section 2.2 of the manuscript, the simulation time was relatively short (only 300 seconds), and the thermal effects under long-term operating conditions were not evaluated.

3. When comparing and analyzing the cooling methods of air cooling and liquid cooling, the authors only provided qualitative conclusions and lacked quantitative data support.

4. When studying the liquid cooling system, the author only considered the heat dissipation effect analysis of the cooling liquid (water) and the flow channel layout under a single factor. It is recommended to carry out research on the compound influence of different cooling liquids (such as ethylene glycol solution) and flow channel layouts (in parallel or series form) on heat dissipation performance.

The quality of the English grammar structure and other aspects of the article still need improvement.

Author Response

The authors begin by addressing our thanks for the availability and interest in reviewing the article proposed for publication. We completely agree that the comments made are necessary and certainly help to increase the scientific level of the presented theme. The text of the article was again analyzed and repaired from a grammatical and linguistic point of view (the score of the text editor is 98%). The corrections made in the text of the manuscript are highlighted with red fonts. The answers to your comments are (point by point) as follows:

C: The authors have conducted systematic research on battery thermal management modeling and heat dissipation scheme optimization, which is of great theoretical significance. However, the entire manuscript only consists of numerical simulation analysis and lacks experimental data to verify the accuracy of simulation results. The following issues should be brought to the attention of the author.

1. In the analysis of simulation results, the author provided cloud maps of the battery temperature field under different heat dissipation conditions but did not quantitatively compare the heat dissipation effects of different schemes.

Reply: Due to the focus on battery lifespan analysis, the only parameters considered were maximum temperature and temperature gradient. Consequently, heat dissipation effects were not quantified for each case. Therefore, our results are based just on temperature and temperature gradient analysis, without considering heat dissipation effects. Nevertheless, this data provided sufficient information to determine the battery lifespan under various simulated conditions. Future work may include heat dissipation effects for more comprehensive assessment of battery performance.

2. In section 2.2 of the manuscript, the simulation time was relatively short (only 300 seconds), and the thermal effects under long-term operating conditions were not evaluated.

Reply: This study focuses on the critical case of fast charging/discharging, which directly and decisively impacts battery lifespan. For this reason, as specified in the text, 300 seconds correspond to the 5 minutes of fast charge/discharge, leading to a complete charge/discharge cycle.

3. When comparing and analyzing the cooling methods of air cooling and liquid cooling, the authors only provided qualitative conclusions and lacked quantitative data support.

Reply: The scope of the article is focused on air-cooling analysis only, with emphasis on the resin material’s influence on different air-cooling BTMS configurations (presented in chapter 2, page 5).

4. When studying the liquid cooling system, the author only considered the heat dissipation effect analysis of the cooling liquid (water) and the flow channel layout under a single factor. It is recommended to carry out research on the compound influence of different cooling liquids (such as ethylene glycol solution) and flow channel layouts (in parallel or series form) on heat dissipation performance.

Reply: The authors recognize the significance of this perspective; therefore, they are currently incorporating this approach into ongoing and future studies to further explore this research direction.

Author Response File: Author Response.pdf

Reviewer 4 Report

Please find comments in the attached document.

Comments for author File: Comments.pdf

Author Response

The authors begin by addressing our thanks for the availability and interest in reviewing the article proposed for publication. We completely agree that the comments made are necessary and certainly help to increase the scientific level of the presented topic. The text of the article was again analyzed and repaired from a grammatical and linguistic point of view (the score of the text editor is 98%). The answers to your comments are (point by point) as follows:

C: The authors simulated the effect of a composite flame-retardant resin on battery life. This work is meaningful.

1. The temperature of Li-ion cells has been simulated with different configuration. However, the difference between simulated temperature and real temperature should be discussed.

Reply:  The aim of this research is only a comparative analysis by numerical methods of the thermal conductivity performance/efficiency.

2. Authors displayed the superiority of the flame-retardant composite resin with the comparison of these models. However, it is hard to distinguish the main effect of structure and composite in this model. Whether the structure acted as the dominant factor for this enhancement.

Reply:  Polymers typically lack free electrons, relying instead on phonon transport for thermal conduction, generally resulting in low thermal conductivity coefficients. There are two primary methods for enhancing thermal conductivity in polymer composites: synthesizing intrinsic thermally conductive polymers and preparing filled thermally conductive polymers. While intrinsic polymers involve complex and costly synthesis processes, filled polymers offer advantages such as ease of fabrication, lower cost, and suitability for industrial production. By incorporating high-thermal-conductivity fillers into polymer matrices, the thermal conductivity coefficient of composites can be significantly improved. The thermal conductivity of filled polymers is influenced by many factors, such as filler type, structure, distribution, content, interface thermal resistance, and intrinsic conductivity of fillers.

Author Response File: Author Response.pdf

Reviewer 5 Report

This work studied the effect of employing a flame-retardant composite resin or an aluminum block on the thermal transfer process to better dissipate the heat generated during a Li-ion battery cycling, on top of conventional air cooling. The authors referred to a clearly-defined model and simulation methodology (Figure 3) for their analysis in 5 difference cases - they presented battery mode thermal behaviors of each case, as well as maximum temperatures with respect to physical time fo comparison. 

They were able to then appeal to Arrhenius Equation and extracted rate constants for Cells#1 and #5 (Figure 15), which is a good frame for understanding the impact of additive materials on the battery kinetics, and thus potential degradation mechanisms changes with different materials add-ons. It was observed from their data that the constructive solution in terms of maximizing battery lifetime should be adding a module component of aluminum block (Case #4). Can the authors in this scenario elaborate on their definition of "total number of chemical reactions"? Like what's the rationale behind coupling #chemical reactions to battery lifetime?

Another thing that can be improved: I would like to have the authors lay out specific experimental details for them to obtain Figures 6, 8, 10, 12 and 14 in the main text for better clarity in their ways of evaluating battery lifetimes, echoing the title of the article.

Also some English language need moderate revising, please see comments in the next section.

1) In the last paragraph of the article, did the authors mean to say "this leads to a future research directive to optimize the dimensioning of the heat exchangers relative to the thickness, shape and placement of the fins to achieve a (quasi) laminar air flow..."? "(cvasi)" reads like a typo.. please correct.

2) In the first paragraph of the Conclusion part, "Battery modules with low local temperatures but high thermal gradients are a sign of an inefficient allocation of resources, since all the energy consumed for the overcooling of some cells while others are undercooled is wasted, and moreover, it leads to electrical imbalance and lifespan reduction." I think the sentence in Bold is not very understandable, thus I suggest a rewrite.

Author Response

Response to Reviewer #5 

The authors begin by addressing our thanks for the availability and interest in reviewing the article proposed for publication. We completely agree that the comments made are necessary and certainly help to increase the scientific level of the presented theme. The text of the article was again analyzed and repaired from a grammatical and linguistic point of view (the score of the text editor is 98%). The corrections made in the text of the manuscript are highlighted with red fonts. The answers to your comments are (point by point) as follows:

C: This work studied the effect of employing a flame-retardant composite resin or an aluminum block on the thermal transfer process to better dissipate the heat generated during a Li-ion battery cycling, on top of conventional air cooling. The authors referred to a clearly-defined model and simulation methodology (Figure 3) for their analysis in 5 difference cases - they presented battery mode thermal behaviors of each case, as well as maximum temperatures with respect to physical time to comparison. 

1. They were able to then appeal to Arrhenius Equation and extracted rate constants for Cells#1 and #5 (Figure 15), which is a good frame for understanding the impact of additive materials on the battery kinetics, and thus potential degradation mechanisms changes with different materials add-ons. It was observed from their data that the constructive solution in terms of maximizing battery lifetime should be adding a module component of aluminum block (Case #4). Can the authors in this scenario elaborate on their definition of "total number of chemical reactions"? Like what's the rationale behind coupling #chemical reactions to battery lifetime?

Reply: Operating Li-ion batteries at high temperatures increases, following the Arrhenius equation, the number of chemical reactions which in consequence can quicken chemical changes, such as the growth of solid electrolyte interphase in cells, loss of active material or electrolytic corrosion. This leads to the reduction in the electrodes’ available surface area for electrochemical reactions and therefore results in a capacity loss, which means that the battery gets closer to the capacity threshold at which it is considered at the end of life (lifespan reduction).

2. Another thing that can be improved: I would like to have the authors lay out specific experimental details for them to obtain Figures 6, 8, 10, 12 and 14 in the main text for better clarity in their ways of evaluating battery lifetimes, echoing the title of the article.

Reply:   Thank you for this suggestion. According to the comment the following paragraph was introduced in the manuscript text (page 9): “The application of the Arrhenius equation to calculate the temperature’s influence on cell longevity requires a fundamental understanding of the impact that temperature changes have on the cells’ performance and durability. Figures 6, 8, 10, 12, and 14 illustrate the temperature variations that were obtained as a result of the simulations for each of the cases.”

3. Also some English language need moderate revising, please see comments in the next section. Comments on the Quality of English Language In the last paragraph of the article, did the authors mean to say "this leads to a future research directive to optimize the dimensioning of the heat exchangers relative to the thickness, shape and placement of the fins to achieve a (quasi) laminar air flow..."? "(cvasi)" reads like a typo.. please correct.

Reply: Thank you for this suggestion. The mentioned mistake has been corrected.

4. In the first paragraph of the Conclusion part, "Battery modules with low local temperatures but high thermal gradients are a sign of an inefficient allocation of resources, since all the energy consumed for the overcooling of some cells while others are undercooled is wasted, and moreover, it leads to electrical imbalance and lifespan reduction." I think the sentence in Bold is not very understandable, thus I suggest a rewrite.

Reply: The highlighted phrase was reformulated in the text: “Battery modules with low local temperatures but high thermal gradients are a sign of an inefficient allocation of resources, since all the energy consumed for the overcooling of some cells can be considered as wasted. Instead, a more efficient energy distribution would mean the better cooling of the undercooled cells, avoiding the formation of thermal gradients that lead to electrical imbalance and lifespan reduction.”

Round 2

Reviewer 2 Report

The article is dedicated to the effect on battery life span using a composite flame-retardant resin in a thermal management system. The title of the article still does not correspond to the content. It may seem from the title that the article used natural experiments to study the life cycle of batteries, but this is only a simulation. Until the word “simulation” appears in the title, the article cannot be published in its current form.

The article is based on unpublished results of a study of the thermal conductivity of resin composites, therefore the results presented in the article are a priori questioned. Authors must present unpublished data in the current article (ESI) or first publish unpublished study results in a separate publication and then cite this one in the presented article. The article that the authors refer to in the discussion with the reviewer [Dadarlat, D.; Tripon, C.; White, I.R.; Korte, D. Photopyroelectric Spectroscopy and Calorimetry. J. Appl. Phys. 2022, 132, doi:10.1063/5.0085594] does not contain data on measuring the thermal conductivity of resins; it is just a tutorial. It is unknown whether any measurements of the thermal conductivity of composite resins have been made at all. Please note that the main conclusions in the article are based on this, which is unacceptable.

Reviewer comments:

The simulation of heating the battery pack was not confirmed by a natural experiment with real batteries. Such an experiment does not seem expensive or difficult. In the opinion of the reviewer, the article demonstrates more the capabilities of the SolidWorks package than research useful for sustainability. It is not clear from the article how the thermal conductivity of special resins, especially with nanocomposites, was simulated. The main question that cannot be avoided: How accurate were the models used?

How does 3% nanocomposite affect the thermal conductivity if it is still 97% resin?

Conclusion: Most questions can be lifted if authors implement the word “simulation” into the title. Accept after major revision.

Author Response

Reply to reviewer #2

Reviewer comments:

The simulation of heating the battery pack was not confirmed by a natural experiment with real batteries. Such an experiment does not seem expensive or difficult. In the opinion of the reviewer, the article demonstrates more the capabilities of the SolidWorks package than research useful for sustainability. It is not clear from the article how the thermal conductivity of special resins, especially with nanocomposites, was simulated. The main question that cannot be avoided: How accurate were the models used?

How does 3% nanocomposite affect the thermal conductivity if it is still 97% resin?

Conclusion: Most questions can be lifted if authors implement the word “simulation” into the title. Accept after major revision.

Reply:

• The validation of the numerical model was performed based on the experimental results previously performed by the authors (please see references [25] and [26]). New text was inserted in the manuscript:

" All the five-case simulation was run on a system with an Intel(R) Xeon(R) Gold 6134 CPU @ 3.20GHz, 130693 MB of RAM, and Windows 10 (Version 10.0.19045), analyzing both laminar and turbulent fluid flow with heat transfer and forced convection. For Case #1, the mesh created for the battery module included 1,223,557 discretized cells. The mesh utilized a total of 234,274 fluid cells to accurately model fluid dynamics, 989,283 solid cells to represent the battery structure, and 156,813 partial cells. The meshing process reached convergence after 300 iterations, lasting an overall of 1867 seconds. In Case #2, the same technique was used to create a discretized mesh for the battery module, which resulted in a total of 1,375,409 cells, including 119,452 fluid cells to represent fluid flow, 1,255,957 solid cells to depict the battery structure, and 66,018 partial cells. Mesh convergence was reached following 300 iterations, required a computational time of 1281 seconds. Significantly, there were no cells removed during the trimming procedure. Case #3 and Case #4 exhibited a meshing approach consistent with Case #2. Case #5 adopted a refined mesh strategy for the battery module, resulting in a total of 1,595,780 cells. This mesh comprised 190,554 fluid cells to capture intricate fluid behavior, 1,405,226 solid cells for detailed representation of the battery structure, and 106,466 partial cells. Notably, no cells were removed during the trimming process, indicating a well-defined mesh.

• To validate the numerical model, the experimental data with the corresponding 10W average heat generation rate obtained in the previous experiments of the authors were used [25, 26], which were compared to the data obtained by running the numerical model in the case of the use of fire-retardant composite resin. The comparative results together with the errors obtained for the considered battery discharge process are presented in Figure 1. It can be stated that due to the heat absorption capacity of the surrounding material, the errors induced by the variation of heat generation rate in function of Depth-of-Discharge (DoD) are mitigated, with an average temperature error of 1.5% and a maximum error of 1.9%. The final temperature values (60.8°C during the experiment and 60.3°C from the simulation) correspond, the error being of 0.9%, results that validate the numerical simulation approach proposed in this paper.".
• The Figure 1 (new inserted) shows the temperature measurements for the case of using flame retardant resin and the errors/differences obtained.
• To simulate the behavior of simulating the effect of flame retardant resin (with and without the addition of nanomaterials), data from the authors' previous research were used. This was highlighted by the introduction of a new bibliographic reference [25]. The composite material was considered in the simulations as having a uniform heat transfer characteristic throughout the volume of the material, in the case of fire retardant resin with composite (please see Table 3, 0.74 w/mK), with the simplifying hypothesis that the volume of nanomaterial is dispersed equally in the volume of resin.
• Two new references [25] and [26] have been introduced and the title was changed.

Reviewer 3 Report

Our suggestion is  to receive a positive response from you, such as supplementing simulation and experimental analysis, and providing more convincing data. Unfortunately, we have not received a direct positive response  from you.

Author Response

Reply to reviewer #3

Reviewer comments:

Our suggestion is  to receive a positive response from you, such as supplementing simulation and experimental analysis, and providing more convincing data. Unfortunately, we have not received a direct positive response  from you.

Reply:  Unfortunately, we did not understand exactly what exactly we did not answer to your requirements/comments. In the past reply to you, we thought that the things reported regarding the purpose of the article were clarified. We are sending you the additions we have made to the article in the hope that some of them touch (and answer) the problems identified by you.

• The validation of the numerical model was performed based on the experimental results previously performed by the authors (please see references [25] and [26]). New text was inserted in the manuscript:

" All the five-case simulation was run on a system with an Intel(R) Xeon(R) Gold 6134 CPU @ 3.20GHz, 130693 MB of RAM, and Windows 10 (Version 10.0.19045), analyzing both laminar and turbulent fluid flow with heat transfer and forced convection. For Case #1, the mesh created for the battery module included 1,223,557 discretized cells. The mesh utilized a total of 234,274 fluid cells to accurately model fluid dynamics, 989,283 solid cells to represent the battery structure, and 156,813 partial cells. The meshing process reached convergence after 300 iterations, lasting an overall of 1867 seconds. In Case #2, the same technique was used to create a discretized mesh for the battery module, which resulted in a total of 1,375,409 cells, including 119,452 fluid cells to represent fluid flow, 1,255,957 solid cells to depict the battery structure, and 66,018 partial cells. Mesh convergence was reached following 300 iterations, required a computational time of 1281 seconds. Significantly, there were no cells removed during the trimming procedure. Case #3 and Case #4 exhibited a meshing approach consistent with Case #2. Case #5 adopted a refined mesh strategy for the battery module, resulting in a total of 1,595,780 cells. This mesh comprised 190,554 fluid cells to capture intricate fluid behavior, 1,405,226 solid cells for detailed representation of the battery structure, and 106,466 partial cells. Notably, no cells were removed during the trimming process, indicating a well-defined mesh.

To validate the numerical model, the experimental data with the corresponding 10W average heat generation rate obtained in the previous experiments of the authors were used [25, 26], which were compared to the data obtained by running the numerical model in the case of the use of fire-retardant composite resin. The comparative results together with the errors obtained for the considered battery discharge process are presented in Figure 1. It can be stated that due to the heat absorption capacity of the surrounding material, the errors induced by the variation of heat generation rate in function of Depth-of-Discharge (DoD) are mitigated, with an average temperature error of 1.5% and a maximum error of 1.9%. The final temperature values (60.8°C during the experiment and 60.3°C from the simulation) correspond, the error being of 0.9%, results that validate the numerical simulation approach proposed in this paper.".

• The Figure 1 (new inserted) shows the temperature measurements for the case of using flame retardant resin and the errors/differences obtained.
• To simulate the behavior of simulating the effect of flame retardant resin (with and without the addition of nanomaterials), data from the authors' previous research were used. This was highlighted by the introduction of a new bibliographic reference [25]. The composite material was considered in the simulations as having a uniform heat transfer characteristic throughout the volume of the material, in the case of fire retardant resin with composite (please see Table 3, 0.74 w/mK), with the simplifying hypothesis that the volume of nanomaterial is dispersed equally in the volume of resin.
• Two new references [25] and [26] have been introduced and the title of article was changed.

Round 3

Reviewer 2 Report

Good job done by the authors. No issues detected.

Conclusion: Accept in present form.

Reviewer 3 Report

Accept as it is