Gas–Water Two-Phase Displacement Mechanism in Coal Fractal Structures Based on a Low-Field Nuclear Magnetic Resonance Experiment
Round 1
Reviewer 1 Report
The ms is well stractured and the experiment study is meaningful. Here are some comments for the authors.
(1) How does the wettability, adsorption dynamics, and relative permeability affect the two-phase flow behavior in the coal? Does the wettability change during the injection process (e.g., as a result of gas adsorption on the pore surfaces of the coal)? Please include come discussions in the ms if possible.
(2) Do the relative permeabilities of gas and water change during the flowing process under different injection pressure conditions? And how? Please include come discussions in the ms if possible.
(3) What are the implications of the results on CO2 storage? The results are likely indicating higher injection pressure is better for storage. But high-pressure injection may cause potential challenges for storage safety in practice. Please include come discussions in the ms if possible.
Author Response
The ms is well stractured and the experiment study is meaningful. Here are some comments for the authors.
Issue 1: How does the wettability, adsorption dynamics, and relative permeability affect the two-phase flow behavior in the coal? Does the wettability change during the injection process (e.g., as a result of gas adsorption on the pore surfaces of the coal)? Please include come discussions in the ms if possible.
Discussion: Thanks to your suggestion, we have added the effects of wettability, gas adsorption and permeability on two-phase flow in Section 3.3 and marked it in red. Please review it.
Issue 2: Do the relative permeabilities of gas and water change during the flowing process under different injection pressure conditions? And how? Please include come discussions in the ms if possible.
Discussion: Thank you for your suggestion, we have added a discussion of the effect of gas injection pressure on gas-water two-phase permeability in Section 3.3 and marked it in red. Please review it.
Issue 3: What are the implications of the results on CO2 storage? The results are likely indicating higher injection pressure is better for storage. But high-pressure injection may cause potential challenges for storage safety in practice. Please include come discussions in the ms if possible.
Discussion: Thank you for your suggestion, we have added the discussion of the implications of this article for CO2 geological storage in Section 4.2 and marked it in red. Please review it.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In this paper, the CO2 gas displacement experiment of saturated water coal body is carried out based on the low field nuclear magnetic resonance device. Through the T2 spectrum obtained from the experiment, the pore size distribution of the coal sample and the gas-water two-phase distribution state in the coal body under different gas injection pressures are obtained. By analyzing the fractal dimension variation law of pore size distribution, combined with the gas-water two-phase distribution law under different gas injection pressures, the water-gas two-phase migration mechanism in the fine microstructure of the coal body is inverted. Overall, this article is worth publishing in the Sustainability, but a few minor issues need to be addressed before this article is published.
（1）In Section 2.2, it is mentioned that 'the relationship between water volume and the nuclear magnetic signal was established'. How is it established?
（2）In Fig.3, after local amplification of the T2 spectrum of the adsorption hole, the comparison between the data is still not clear enough. The local amplification map only needs to select a representative small section to amplify to highlight the change law of the data, not all amplification.
（3）The method of pore classification in Section 3.3 should add references.
（4）A schematic diagram of the concept of ' dominant pathways ' and a schematic diagram of its effect on displacement efficiency should be added in Section 3.3.
（5）Why can the fractal dimension of liquid phase distribution reflect the uniformity of pore composition in dominant pathways? How does it reflect the law of gas-water two-phase displacement?
（6）Some pictures ( such as Fig.1, Fig.2, and Fig.6 ) in the article have poor clarity. Please modify the picture format to improve the clarity of the picture.
In this paper, the CO2 gas displacement experiment of saturated water coal body is carried out based on the low field nuclear magnetic resonance device. Through the T2 spectrum obtained from the experiment, the pore size distribution of the coal sample and the gas-water two-phase distribution state in the coal body under different gas injection pressures are obtained. By analyzing the fractal dimension variation law of pore size distribution, combined with the gas-water two-phase distribution law under different gas injection pressures, the water-gas two-phase migration mechanism in the fine microstructure of the coal body is inverted. Overall, this article is worth publishing in the Sustainability, but a few minor issues need to be addressed before this article is published.
（1）In Section 2.2, it is mentioned that 'the relationship between water volume and the nuclear magnetic signal was established'. How is it established?
（2）In Fig.3, after local amplification of the T2 spectrum of the adsorption hole, the comparison between the data is still not clear enough. The local amplification map only needs to select a representative small section to amplify to highlight the change law of the data, not all amplification.
（3）The method of pore classification in Section 3.3 should add references.
（4）A schematic diagram of the concept of ' dominant pathways ' and a schematic diagram of its effect on displacement efficiency should be added in Section 3.3.
（5）Why can the fractal dimension of liquid phase distribution reflect the uniformity of pore composition in dominant pathways? How does it reflect the law of gas-water two-phase displacement?
（6）Some pictures ( such as Fig.1, Fig.2, and Fig.6 ) in the article have poor clarity. Please modify the picture format to improve the clarity of the picture.
Author Response
In this paper, the CO2 gas displacement experiment of saturated water coal body is carried out based on the low field nuclear magnetic resonance device. Through the T2 spectrum obtained from the experiment, the pore size distribution of the coal sample and the gas-water two-phase distribution state in the coal body under different gas injection pressures are obtained. By analyzing the fractal dimension variation law of pore size distribution, combined with the gas-water two-phase distribution law under different gas injection pressures, the water-gas two-phase migration mechanism in the fine microstructure of the coal body is inverted. Overall, this article is worth publishing in the Sustainability, but a few minor issues need to be addressed before this article is published.
Issue 1: In Section 2.2, it is mentioned that 'the relationship between water volume and the nuclear magnetic signal was established'. How is it established?
Discussion: Thank you very much for your question. Because the NMR signal is proportional to the content of water in the sample under the same detection parameters. Therefore, we performed nuclear magnetic resonance experiments on a set of standard samples with known water content to obtain their nuclear magnetic signals at different water contents. The experimental data were fitted to a curve of water content and nuclear magnetic resonance signal, and used as a standard. The amount of water in the sample can be obtained by taking the signal measured by the test sample into the curve equation. Based on this, we established a quantitative relationship between water content and nuclear magnetic signal.
Issue 2: In Fig.3, after local amplification of the T2 spectrum of the adsorption hole, the comparison between the data is still not clear enough. The local amplification map only needs to select a representative small section to amplify to highlight the change law of the data, not all amplification.
Discussion: We are very sorry, due to our negligence, in some pictures, the change rule of data is not prominent enough. We have modified it and marked it in red. Please review it.
Issue 3: The method of pore classification in Section 3.3 should add references.
Discussion: Thank you for pointing this out. We have cited the corresponding references in section 3.3 and marked it in red. Please review it.
Issue 4: A schematic diagram of the concept of ' dominant pathways ' and a schematic diagram of its effect on displacement efficiency should be added in Section 3.3.
Discussion: Thank you for pointing this out. We have added a schematic diagram of the ' dominant pathways ' in Section 3.3 and marked it in red. Please review it.
Issue 5: Why can the fractal dimension of liquid phase distribution reflect the uniformity of pore composition in dominant pathways? How does it reflect the law of gas-water two-phase displacement?
Discussion: Thank you very much for your question. Because the fractal dimension of liquid phase distribution is obtained by linear fitting of cumulative water-bearing porosity, it can reflect the complexity of liquid phase distribution in coal body to a certain extent. When the liquid phase distribution inside the coal body is not uniform, and the liquid phase is mainly concentrated in a certain pore size range, the slope of the fitting process will be larger, resulting in a smaller fractal dimension. Therefore, through the size of the fractal dimension, we can analyze the uniformity of the liquid phase distribution inside the coal body. Similarly, since the gas will preferentially displace the dominant channel with less resistance during the gas-water two-phase displacement process, the liquid phase distribution will change. Therefore, we can reversely obtain the uniformity of the pore size composition of the dominant channel by the change of the fractal dimension of the liquid phase distribution.
In the process of gas-water two-phase displacement, the liquid phase distribution in the pores is constantly changing. By analyzing the fractal dimension of the liquid phase distribution of the coal body, the variation law of the liquid phase distribution in different pore size ranges can be obtained. When the fractal dimension changes greatly, it shows that a large amount of water is displaced by gas in the pore size range. Based on this, we can analyze the influence of gas injection pressure on gas-water two-phase displacement in different pore size ranges.
Issue 6: Some pictures (such as Fig.1, Fig.2, and Fig.6) in the article have poor clarity. Please modify the picture format to improve the clarity of the picture.
Discussion: We are very sorry that due to our negligence, some of the pictures in the manuscript are not clear enough. We have replaced them in the manuscript and marked it in red. Please review it.
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper develops machine learning models to predict the deteriorated compressive strength of concrete after freeze-thaw exposure. The authors have compiled an extensive dataset from past literature and used it to train artificial neural network, random forest, and support vector machine models. The key novelty is the use of environmental factors like temperature extremes along with material properties as predictors of frost resistance, providing more insight than just mix proportions. Overall, this is a well-written paper that is suitable for the journal after minor revisions.
Minor suggestions:
1. In the abstract, rephrase "outperforming RF and SVM models" as "exhibiting higher prediction accuracy than RF and SVM models".
2. In Section 2.1, improve the explanation of how training optimizes the artificial neural network model.
3. The current Figure 1 caption "Input and output parameters" is somewhat misleading, as the figure presents distributions of the experimental variables used in model development rather than internal model parameters. Changing the caption would enhance clarity for the reader on what is presented in Figure 1.
4. In Figure 7, the y-axis is currently labeled 'Compressive strength (MPa)'. However, based on the context of the paper, it seems this axis represents the deteriorated compressive strength. To make the meaning of this data clearer to readers, consider revising the y-axis label to 'Deteriorated compressive strength (MPa)' or something similar.
5. Consider expanding the conclusion to discuss on practical implications of the developed models, such as how the ability to accurately predict freeze-thaw resistance can guide selection of optimal structural designs for cold climates.
6. Some references are messy, some page no is missing. Some references are old, they should be updated,such as Ref.[3,4,31 and 32].
none
Author Response
Comments：
Issue 1: The size of confining pressure during the experiment was not specified, which should be supplemented in the paper.
Discussion: We are very sorry that due to our negligence, the specific value of the confining pressure during the nuclear magnetic experiment was not explained in detail in the manuscript. We have revised it in section 2.2 of the manuscript and marked it in red. Please review it.
Issue 2: Why is it necessary to measure the porosity of coal samples by weighing method and nuclear magnetic method respectively?
Discussion: Thank you very much for your question. In this paper, two methods are used to test the porosity of coal in order to verify the accuracy of NMR experimental results. Because the nuclear magnetic resonance test is by converting the T2 spectrum signal to water content, the water content of the coal body is the total pore volume of the coal sample, and the conversion process is by comparing the signal amount of the sample to be tested with the standard sample. The ratio of the signal amount is the ratio of the porosity, so as to obtain the porosity of the sample to be tested, but there is a certain error in the porosity obtained by the nuclear magnetic resonance test. The weighing method directly determines the water content of the coal body by calculating the mass change of the coal sample before and after saturation. The error of this method is smaller, so the accuracy of the nuclear magnetic test can be verified by the test results of the weighing method.
Issue 3: A set of piecewise linear fitting graphs of T2 spectrum without gas injection should be added in Section 4.1 to compare the fractal dimension changes of liquid phase distribution before and after gas injection.
Discussion: Thank you for your suggestion, we have added the T2 spectrum linear fitting diagram without gas injection in section 4.1 and modified the analysis. The modification has been marked in red. Please review it.
Issue 4: Section 3.4 says, "As can be seen from the figure, the displacement efficiency of the XLZ coal sample shows a uniform growth trend as the gas injection pressure increases ". This is not accurate, because the increase rate of displacement efficiency of XLZ coal sample is also gradually decreasing, rather than increasing uniformly, and should be modified in the paper.
Discussion: We are very sorry. We have modified the corresponding content in the manuscript and marked it in red. Please review it.
Issue 5: Fig. 9 only shows the principle of the dominant channel and the formation of residual water, but does not show the difference between the gas-water two-phase distribution state in the adsorption hole and the seepage hole with the change of gas injection pressure. This part should be supplemented in Fig. 9.
Discussion: Thank you for your suggestion, we have supplemented the variation of gas-water two-phase distribution in adsorption pores and seepage pores with gas injection pressure in Fig.9. Please review it.
Issue 6: Individual chart titles lack punctuation.
Discussion: We are very sorry. We have supplemented the punctuation marks in the title and marked it in red. Please review it.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Thanks for the authors' work. I have no further comments.
