Ammonia Nitrogen Removal and MAP Crystal Morphology Affected by Reaction Conditions in High-Concentration Wastewater

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Please find the attached file.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

Thank you for your thoughtful and constructive comments, which have greatly helped us improve the manuscript.

 

1. Is this process can be applied on low-concentration wastewater treatment? Agricultural water, groundwater, or surface water generally have low ammonium concentrations, mostly less than 50mg/L, and low ammonia is more problematic for human health, because it is directly used for drinking. Please add one paragraph on low-concentration wastewater treatment in your paper, discussing whether it can be removed or not. Also, differentiate the different processes for low ammonium concentration removal such as biological, physicochemical processes.

 

Response:
We sincerely thank the reviewer for raising this important point regarding the applicability of the MAP process to wastewater with low ammonium concentrations. We agree that this is a meaningful issue to clarify, and we will keep attention on low ammonium wastewater. But this time we may not add this in manuscript as our topic is high ammonium concentration wastewater.

For low-strength ammonium wastewater (NH₄⁺-N ≤ 50 mg·L⁻¹), such as agricultural runoff, groundwater, or drinking water sources, conventional treatment methods are generally more suitable. Biological processes (e.g., nitrification–denitrification, biofilm reactors) can achieve high removal efficiencies of 90–99% under stable operating conditions [1]. Physicochemical methods (e.g., ion exchange, adsorption, reverse osmosis) also remove 80–95% of ammonium effectively, though they often involve higher costs and secondary waste generation [2].

By contrast, the single MAP process is less efficient at low ammonium concentrations. The limited supersaturation reduces struvite nucleation and growth, resulting in fine particles with poor settleability. Moreover, the need for external phosphate dosing may cause undesired residual phosphorus in the effluent [3]. The MAP method can be combined with processes such as fluidized bed reactors to treat low-concentration ammonia nitrogen wastewater, achieving an ammonia nitrogen removal rate of up to 80% [4].Therefore, MAP is more suitable for high-strength ammonium wastewater or for nutrient recovery when both ammonium and phosphate are present at elevated levels.

References

1. 1. Miao, L., G. Yang, T. Tao and Y. Peng. "Recent advances in nitrogen removal from landfill leachate using biological treatments - a review." J Environ Manage 235 (2019): 178-85. 10.1016/j.jenvman.2019.01.057.
2. 2. Farghali, M., Z. Chen, A. I. Osman, I. M. Ali, D. Hassan, I. Ihara, D. W. Rooney and P.-S. Yap. "Strategies for ammonia recovery from wastewater: A review." Environmental Chemistry Letters 22 (2024): 2699-751.
3. 3. Li, P., L. Chen, Y. Ding, X. Tian, D. Guan, Z. Zhang and J. Li. "Phosphorus recovery from urine using cooling water system effluent as a precipitant." J Environ Manage 244 (2019): 391-98. 10.1016/j.jenvman.2019.05.057.
4. 4. Wang, J., H. Gong, X. Liu, Z. Wei and K. Wang. "Optimizing induced struvite crystallization in a fluidized bed reactor for low-strength ammonium wastewater treatment." Desalination and Water Treatment 201 (2020): 95-104. 10.5004/dwt.2020.26181.

 

 

2. In this paper, you mentioned that the MAP method is more efficient. Is this process also efficient for lower ammonium concentrations (less than or equal to 50mg/L)? Is MAP more effective than other biological and physicochemical processes in terms of ammonium removal? Moreover, is this process more cost-effective compared to other methods? Please prepare a comparison table. There is a lot of data available on ammonium removal through different processes.

Response:

MAP method demonstrates the most significant advantages when precipitated from high-concentration ammonium or phosphorus-rich streams, such as digested liquor, landfill leachate, and specific industrial wastewaters. In such cases, it enables the simultaneous removal and recovery of nitrogen (N) and phosphorus (P), while producing a marketable or reusable crystalline solid. However, in low-concentration wastewater (≤50 mg·L⁻¹ NH₄⁺-N), the driving forces for MAP precipitation diminish. To achieve effective sedimentation under these conditions, additional dosing of phosphate (P) and magnesium (Mg) is typically required, or techniques such as seed crystal addition and agglomeration enhancement are employed. These approaches tend to be economically and operationally more complex, making them generally less favorable compared to biological treatment or ion exchange methods.

In addition, we have included Table S1 (" Comparison of Treatment Methods for Ammoniacal Nitrogen Wastewater ") for a better demonstration.

 

 

 

Table S1. Comparison of treatment methods for ammoniacal nitrogen wastewater

Method	Applicable NH₄⁺-N Conc. (mg/L)	Removal Efficiency (%)	Cost-Effectiveness	Advantages	Disadvantages	References
Air Stripping	500 - 5000	80 - 95%	Moderate to High	Simple process, easy operation Effective for high concentrations Relatively low equipment investment	Requires pH adjustment to 10.5-11.5 Prone to scaling and clogging Poor performance in winter Cause secondary air pollution	[5, 6]
Breakpoint Chlorination	50 - 1000	85 - 99%	Low to Moderate	Stable and reliable treatment effect Short reaction time Unaffected by temperature	High consumption of chlorinating agents High operating costs Generates by-products like chloramines Uneconomical for high concentrations	[7]
Ion Exchange	20 - 500	90 - 98%	Moderate	High-quality effluent Simple operation Allows for ammonia nitrogen recovery	Resin is susceptible to fouling/poisoning Requires frequent regeneration Limited application scope High pretreatment requirements	[8, 9]
Struvite Precipitation (MAP)	200 - 8000	85 - 95%	Moderate to High	Recovers nitrogen and phosphorus resources Low sludge production Effective for high concentrations By-product can be used as fertilizer	Requires addition of Mg²⁺ and PO₄³⁻ Strict pH control is necessary Potential for scaling issues Higher investment costs	[10, 11]
Biological Nitrification-Denitrification	20 - 200	85 - 95%	High	Low operating costs Environmentally friendly Simultaneously remove COD Mature and established technology	Long start-up time Sensitive to high ammonia concentrations Requires an external carbon source Large footprint	[1, 12]
Electrochemical Oxidation	100 - 2000	70 - 90%	Low	Mild reaction conditions No chemical reagents needed Small footprint High degree of automation	High electricity consumption Electrode passivation is common High investment costs Short electrode lifespan	[13]
Membrane Separation	50 - 1000	90 - 99%	Low to Moderate	Excellent separation efficiency Small footprint Stable effluent quality Can achieve ammonia recovery	Severe membrane fouling High operating costs Concentrate stream requires further treatment High pretreatment requirements	[14]
Catalytic Oxidation	100 - 5000	80 - 95%	Moderate	Fast reaction speed High treatment efficiency Strong adaptability Can handle high concentrations	Catalyst is prone to deactivation Relatively high operating costs Complex process Pretreatment is often required	[15, 16]

 

3. Section 2.1, add a table showing the species concentrations such as, Ammonium chloride, disodium hydrogen phosphate, magnesium chloride (MgCl₂), sodium hydroxide (NaOH), analytical-grade reagents, including Seignette salt, etc.

 

Response:

Thank you for your suggestion. We added a reagent/species table (Table S4) listing chemical details.

Although the prepared ammonium chloride solution was used to simulate high-concentration ammonia nitrogen wastewater, in the MAP experiment, the supernatant may contain magnesium ions due to excessive magnesium. During the ammonia nitrogen concentration test, seignette salt needs to be added as a masking agent. The operation should be carried out in accordance with the standard steps of the national standard of the People's Republic of China (HJ 535-2009) to ensure the accuracy of the results [1].

 

Reference：

1.  Ministry of Environmental Protection of the People’s Republic of China. (2009). HJ 535-2009: Water quality—Determination of ammonia nitrogen—Nessler’s reagent spectrophotometry. Beijing, China: China Environmental Science Press.

Table S4. Chemicals and reagents

Chemical Formula	Mass	Volume	Concentration
NH4Cl	1.07 g	500 mL	560 mg/L
MgCl2	Based on experimental conditions	-	-
Na2HPO4	Based on experimental conditions	-	-
NaOH	-	-	5.0 M
KNaC4H4O6·4H2O	50g	100 mL	500g/L

Note: All chemical reagents used were of analytical grade and purchased from Tianjin Damao Reagent Co., Ltd., China.

 

 

 

 

4. Nitrogen Removal Efficiency: is there any accumulation of Nitrite or Nitrate?

Response:
Thank you for your question. Since MAP is a non-biological oxidation process, it does not involve the pathway of oxidizing NH4⁺ to NO₂^-/NO₃^-. And under the short-term (30 min) chemical crystallization conditions of this study, it is theoretically not expected to generate nitrites or nitrates.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

1. The introduction (lines 37–99) provides a general background but could benefit from a deeper contextualization. Explicit definition of the challenge of ammonia nitrogen in various wastewater sectors with recent statistical data or regulatory context. Elaboration on the current gaps in HANW treatment, including practical bottlenecks and technology adoption in real-world scenarios.

 

2. Lines 140–160 lacks sufficient detail regarding the construction and training of the machine learning models. Specify the dataset size, handling of outliers, data splitting strategy, feature selection rationale, model hyperparameters (beyond those already mentioned), and cross-validation procedure.

 

3. The discussion (lines 498 onwards) focuses on crystalline morphology and process optimization but should be strengthened by critical evaluation of the MAP process scalability—how these findings transfer from lab to pilot or full-scale HANW treatment.

Author Response

Dear Reviewer,

Thank you for your thoughtful and constructive comments, which have greatly helped us improve the manuscript.

 

1. The introduction (lines 37–99) provides a general background but could benefit from a deeper contextualization. Explicit definition of the challenge of ammonia nitrogen in various wastewater sectors with recent statistical data or regulatory context. Elaboration on the current gaps in HANW treatment, including practicalbottlenecks and technology adoption in real-world scenarios.

 

Response:
We express our gratitude to the reviewer for their insightful and constructive suggestions. In the revised Introduction (lines 39–46 and 76–99), we have incorporated the following enhancements: (i) a regulatory background to provide context; (ii) a concise summary of the primary challenges in treating high-ammonia-nitrogen wastewater (HANW); and (iii) an overview of the current research on MAP morphology alongside identified research gaps. These additions elucidate the practical significance and technological deficiencies that underpin the motivation for this study.

Manuscript changes：

Lines 39–46：

Ammonia nitrogen (NH₃–N) is a critical pollutant in wastewater and is strictly regulated due to its ecological and health impacts. In China, although nutrient discharges have declined since 2016, legacy nitrogen contamination in rivers and groundwater remains widespread, often exceeding regulatory thresholds [1]. The implementation of Special Discharge Limits (SDL) has further tightened NH₃–N standards, yet compliance frequently demands energy- and chemical-intensive upgrades, posing economic and operational challenges [2]. Similar issues are reported globally, underscoring the urgent need for more efficient and resource-oriented treatment technologies.

 

Lines 76–99：

Previous studies confirm that MAP precipitation can be safely achieved in complex wastewater and that struvite is a marketable slow-release fertilizer [17, 19, 20]. However, research efforts to date have largely focused on two directions: (i) the treatment and nutrient recovery from complex industrial wastewaters, where issues such as ion interference (Ca²⁺, heavy metals, organic matter) and product variability are critical [16]; and (ii) pilot- or full-scale applications, emphasizing nitrogen removal efficiency, energy consumption, and life-cycle assessment of process sustainability. While these are important for practical deployment, they often bypass systematic laboratory investigations of how fundamental reaction parameters control the morphology, size distribution, crystallinity, and purity of struvite crystals.

Crystal morphology and purity strongly determine the stability, settling, and reuse value of struvite products. By contrast, phosphorus recovery studies have shown that pH and supersaturation directly influence both removal efficiency and the particle characteristics of MAP [21, 22]. For example, Shaddel used lab-scale batch experiments to vary supersaturation (via pH, Mg and NH₄ concentrations) and quantified how these changes affect crystal size and morphology while also assessing phosphorus recovery and process performance [23]. In another study using digestate from anaerobic digestion, the effects of n(Mg):n(N), n(P):n(N), pH, and temperature on struvite yield and SEM-observed crystal morphology were systematically evaluated via Taguchi experimental design [24]. Similarly, studies in (synthetic or real) urine have shown that particle size, crystal habit, and separation behaviour depend very sensitively on pH, supersaturation, and mixing intensity [25]. Extending such knowledge to nitrogen recovery is essential to provide the mechanistic understanding needed for optimizing product quality and supporting in-dustrial-scale application.

 

2. Lines 140–160 lacks sufficient detail regarding the construction and training of the machine learning models. Specify the dataset size, handling of outliers, data splitting strategy, feature selection rationale, model hyperparameters (beyond those already mentioned), and cross-validation procedure.

Response:
We thank the reviewer for the valuable feedback regarding the lack of detail in lines 140–160 concerning the construction and training of the machine learning models.

We expanded Section 2.4.4. (“Model Development”) to fully describe data curation and training: dataset size, outlier handling, feature engineering/selection, data splits, hyperparameter search, and cross-validation.

Lines 218-258:

Data analysis and computations were conducted using Python (version 3.12.4) in a Jupyter Notebook environment. The following libraries were employed: Pandas (version 2.2.2) for data loading and preprocessing, NumPy (version 1.26.4) for numerical opera-tions, and Scikit-learn (version 1.4.2) for machine learning tasks, including data scaling (StandardScaler), feature transformation (PolynomialFeatures), data splitting (train_test_split), and model training with linear models (Linear Regression, Ridge, Lasso) and ensemble methods (RandomForestRegressor, GradientBoostingRegressor). Model performance was evaluated using Scikit-learn metrics (r2_score, mean_squared_error, mean_absolute_error). Visualizations were generated using Matplotlib (version 3.8.4).

1. Dataset Size

The dataset consists of 90 samples, corresponding to the experimental conditions detailed in Table S3. This dataset was derived from controlled experiments simulating high-concentration ammonia nitrogen wastewater treatment, ensuring a representative range of pH, molar ratios (n(Mg):n(N), n(P):n(N)), stirring speed, time, and temperature for model training and validation.

2. Handling of Outliers

No specific outlier removal was performed, as the experimental data in Table S3 were pre-validated to ensure consistency. Standardization using “StandardScaler” was applied to normalize the feature distributions, mitigating the impact of potential outliers.

3. Data Splitting Strategy

The dataset was split into training and testing according 80/20 ratio, consistent with prior studies [36]. And the dateset sets using “train_test_split” with a fixed random state of 42 to ensure reproducibility.

4. Feature Selection Rationale

Features including pH, n(Mg):n(N), n(P):n(N), stirring speed, time, and temperature were selected based on their known influence on nitrogen removal efficiency, as informed by experimental design and prior literature [37]. No automated feature selection was applied.

5. Model hyperparameters

Linear Regression and Polynomial Regression (degree = 2) used default settings; Ridge Regression with alpha = 1.0, Lasso Regression with alpha = 0.1, Random Forest with n_estimators = 100 (optimized via GridSearchCV with n_estimators = [50, 100, 200], max_depth = [None, 10, 20], min_samples_split = [2, 5]), and Gradient Boosting with n_estimators=100. Default values for other parameters. Hyperparameter tuning for Random Forest was conducted using 5-fold cross-validation via GridSearchCV to mini-mize mean squared error [38].

6. Cross-Validation Procedure:

A 5-fold cross-validation procedure was implemented during hyperparameter tuning with GridSearchCV to assess model performance robustly across different data subsets. This choice of 5 folds balances computational efficiency and reliability, aligning with common practices in machine learning studies [39].

 

3. The discussion (lines 498 onwards) focuses on crystalline morphology and process optimization but should be strengthened by critical evaluation of the MAP process scalability—how these findings transfer from lab to pilot or full-scale HANW treatment.

Response:
Thank you for your insightful and valuable feedback, particularly the suggestion to strengthen the discussion by including a critical evaluation of the MAP process scalability for high ammonia nitrogen wastewater (HANW) treatment. We fully agree that addressing the transferability of laboratory findings to pilot or full-scale applications is crucial for enhancing the practical relevance of this study.

To address your comment, we have revised the manuscript by adding a new subsection (Section 4.2, “Scalability Considerations for MAP Process in HANW Treatment”). This section provides a detailed analysis of process parameter adaptability, engineering challenges, and economic feasibility.

Lines 609-633

4.2 Scalability Considerations for MAP Process in HANW Treatment

While the laboratory-scale experiments demonstrated effective nitrogen removal and high-purity struvite precipitation under optimized conditions, scaling the MAP process to pilot or full-scale applications for high ammonia nitrogen wastewater (HANW) treatment presents several challenges and opportunities. The optimized parameters may require adjustments to accommodate the complex matrix of HANW, which often contains high salinity, organic matter, or heavy metals [65]. These components could interfere with crystal nucleation and growth, potentially reducing struvite purity and yield [16]. Pre-treatment steps, such as filtration or chemical conditioning, may be necessary to mitigate these effects, increasing operational complexity and costs.

From an engineering perspective, reactor design and operational conditions must be tailored for larger scales. For instance, achieving uniform mixing in large reactors to avoid concentration polarization, as observed with insufficient agitation in lab studies, requires advanced impeller designs or continuous flow systems. Excessive shear stress from high stirring speeds, which disrupts crystal integrity in lab settings, could be amplified in industrial reactors, necessitating precise control systems. Additionally, thermal management becomes critical at scale, as temperature fluctuations could exacerbate morphological degradation, as noted in Section 3.1.

Economic feasibility is another critical factor. While lab-scale MAP processes demonstrate high nitrogen removal efficiency, the costs of reagents (e.g., Mg and P sources) and energy for mixing and heating must be evaluated for full-scale implementation. Pilot studies integrating cost-benefit analyses and real-world HANW com-positions are essential to validate scalability. Future research should focus on optimizing reactor configurations and exploring low-cost reagent alternatives to enhance the practical applicability of the MAP process.

 

Additionally, the conclusion (Section 5) has also been updated to highlight the importance of future pilot-scale studies for practical implementation.

Lines 645-647

Future studies should prioritize pilot-scale experiments to validate the scalability of the MAP process for HANW treatment, addressing engineering and economic challenges to ensure practical implementation.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Please find below my comments for the improvement of your manuscriptȘ

The introduction has to develop more the originality of the manuscript

Line 58-59 needs bibliography; the same phrase 59-61 needs references

Eq (1) is not balanced regarding the charges

The software used to perform the calculations is not specified

The name, country, city of the instrument used to perform SEM is not specified

It is not clear if all the experimental data was recorded in triplicate experiments

The Discussions section lacks comparison with already published researches

Why there is no empirical regression equation presented, but only the factors allowing to decide which model fits best experimental data? The mathematical expressions of the fits should be included in the manuscript

Author Response

Dear Reviewer,

Thank you for your thoughtful and constructive comments, which have greatly helped us improve the manuscript.

 

1. The introduction has to develop more the originality of the manuscript

 

Response:

Thank you for your advice. We have added some lines to improve the manuscript.

Lines 76–99：

Previous studies confirm that MAP precipitation can be safely achieved in complex wastewater and that struvite is a marketable slow-release fertilizer [17, 19, 20]. However, research efforts to date have largely focused on two directions: (i) the treatment and nutrient recovery from complex industrial wastewaters, where issues such as ion interference (Ca²⁺, heavy metals, organic matter) and product variability are critical [16]; and (ii) pilot- or full-scale applications, emphasizing nitrogen removal efficiency, energy consumption, and life-cycle assessment of process sustainability. While these are important for practical deployment, they often bypass systematic laboratory investigations of how fundamental reaction parameters control the morphology, size distribution, crystallinity, and purity of struvite crystals.

Crystal morphology and purity strongly determine the stability, settling, and reuse value of struvite products. By contrast, phosphorus recovery studies have shown that pH and supersaturation directly influence both removal efficiency and the particle characteristics of MAP [21, 22]. For example, Shaddel used lab-scale batch experiments to vary supersaturation (via pH, Mg and NH₄ concentrations) and quantified how these changes affect crystal size and morphology while also assessing phosphorus recovery and process performance [23]. In another study using digestate from anaerobic digestion, the effects of n(Mg):n(N), n(P):n(N), pH, and temperature on struvite yield and SEM-observed crystal morphology were systematically evaluated via Taguchi experimental design [24]. Similarly, studies in (synthetic or real) urine have shown that particle size, crystal habit, and separation behaviour depend very sensitively on pH, supersaturation, and mixing intensity [25]. Extending such knowledge to nitrogen recovery is essential to provide the mechanistic understanding needed for optimizing product quality and supporting in-dustrial-scale application.

 

2. Line 58-59 needs bibliography; the same phrase 59-61 needs references.

 

Response:

Thank you for your suggestion. We have added relevant literature and optimized the expression of these sentences.

Line 66-70：

In alkalization evaporation treatment, high ammonia nitrogen concentrations may cause equipment corrosion or scaling, impairing operational stability, while low concentrations increase operating costs due to the high steam demand for heating [14]. The MAP crystallization method is suitable for medium- to high-concentration ammonia nitrogen wastewater (500–3000 mg/L) [15, 16].

 

3. Eq (1) is not balanced regarding the charges.

 

Response:

Thank you for your suggestions for modification. The chemical reaction Eq (1) has been corrected in the text.

(1)

 

4. The software used to perform the calculations is not specified.

 

Response:

 

We appreciate the reviewer's comment. The data analysis and calculations were performed using the Python programming language (version 3.12.4) within a Jupyter Notebook environment.

The specific libraries used for data processing, model training, and performance evaluation are listed below, along with their respective roles:

Pandas (version 2.2.2) was used for data loading and preprocessing, while NumPy (version 1.26.4) handled numerical operations. The Scikit-learn (version 1.4.2) library was extensively used for all machine learning tasks, including: Data scaling and feature transformation (StandardScaler, PolynomialFeatures), Data splitting (train_test_split), Model training using linear models (LinearRegression, Ridge, Lasso) and ensemble methods (RandomForestRegressor, GradientBoostingRegressor).

Performance Evaluation: Model performance was assessed using metrics from Scikit-learn (r2_score, mean_squared_error, mean_absolute_error). Visualization was handled with Matplotlib (version 3.8.4).

These details have been added to the "2.4.4. Model Development" section of the manuscript to ensure full transparency and reproducibility of the analysis.

Lines 218-226：

Data analysis and computations were conducted using Python (version 3.12.4) in a Jupyter Notebook environment. The following libraries were employed: Pandas (version 2.2.2) for data loading and preprocessing, NumPy (version 1.26.4) for numerical opera-tions, and Scikit-learn (version 1.4.2) for machine learning tasks, including data scaling (StandardScaler), feature transformation (PolynomialFeatures), data splitting (train_test_split), and model training with linear models (Linear Regression, Ridge, Lasso) and ensemble methods (RandomForestRegressor, GradientBoostingRegressor). Model performance was evaluated using Scikit-learn metrics (r2_score, mean_squared_error, mean_absolute_error). Visualizations were generated using Matplotlib (version 3.8.4).

 

5. The name, country, city of the instrument used to perform SEM is not specified.

 

Response:

Thank you for your attention. The model of the scanning electron microscope (SEM) is Zeiss Gemini 300, manufactured by Carl Zeiss AG in Germany. The experiment was conducted in Shenyang City, Liaoning Province, China. Relevant content has been added to the “2.4.3. Morphological Analysis Methods” of the manuscript.

Lines 213-216：

The morphology and structure of the sample were studied using a Gemini SEM 300 scanning electron microscope (SEM-EDS, ZEISS Corp., Germany) at the experimental facility in Shenyang City, Liaoning Province, China.

 

6. It is not clear if all the experimental data was recorded in triplicate experiments.

 

Response:

Thank you for your question. The experimental conditions for the sequence numbers in Table S2 are repeated three times, and the resulting data are presented in Table S3.

 

7. The Discussions section lacks comparison with already published researches.

 

Response:

Thank you for your suggestion. We have already added relevant literature to the manuscript.

Lines 313-315：

As the pH increases to 9.0 and 9.5, the crystals become more uniform and elongated, exhibiting the typical rod- or needle-like morphology of well-formed struvite, similar to the figures in references [23, 45].

Lines 459-460：

The shape changes as the variation of supersaturation and reaction time [23, 55].

Lines 498-499：

When the temperature is increased by 5 ℃ to 20–25 ℃, the particle size decreases significantly, with wedge-shaped crystals mainly being formed, similar with previous study [25].

 

8. Why there is no empirical regression equation presented, but only the factors allowing to decide which model fits best experimental data? The mathematical expressions of the fits should be included in the manuscript.

 

Response:

Thank you for your valuable suggestions. The relevant empirical regression equation has been added to the manuscript in 3.2. Regression Model Equations.

Lines 507-524：

3.2. Regression Model Equations

In this study, the dependent variable is the product yield (y). The independent variables include: pH (x₁), the molar ratio of magnesium to nitrogen (x₂, n(Mg):n(N)), the molar ratio of phosphorus to nitrogen (x₃, n(P):n(N)), stirring speed (x₄, rpm), reaction time (x₅, min), and temperature (x₆, °C).  The empirical regression equation for the Linear models are expressed as follows:

1. Linear model(Linear Regression, Ridge, Lasso)

Table 1. Empirical regression equations for linear model

Model	Empirical regression equation
Linear Regression	y=39.7778+4.4016x1+8.8262x2+4.9330x3−0.0312x4+0.0118x5+0.0939x6
Ridge Regression	y=50.8501+4.1982x1+3.7786x2+1.2674x3−0.0341x4+0.0033x5+0.1108x6
Lasso Regression	y=97.7371+0.0000x1+0.0000x2+0.0000x3−0.0284x4+0.0000x5+0.0103x6

Note: The coefficients for pH, n(Mg):n(N), n(P):n(N), and Time were regularized to zero by the Lasso model, indicating their minimal contribution to the predictive model.

 

2. Complex model(RandomForest Regressor, Gradient Boosting Regressor)

Unlike the linear regression models, ensemble methods such as Random Forest Regressor and Gradient Boosting Regressor do not have a simple empirical regression equation. These models operate on a series of complex decision rules derived from the training data rather than a fixed set of coefficients. This non-parametric nature allows them to capture intricate, nonlinear relationships in the data more effectively, as evidenced by their superior performance metrics.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you so much for providing the response. I do not have any questions now, and the decision on the manuscript is up to the editor.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors modified the manuscript according to the suggestions received.