Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript presents a systematic optimisation study on the esterification modification of beech wood using combinations of polyethylene glycol (PEG400) with three different carboxylic acids (BTCA, CA, and MA). The work builds logically on the authors' previous screening experiments and provides useful data on the effects of catalyst addition, chemical concentration, and curing conditions on weight percent gain (WPG), bulking, and anti-swelling efficiency (ASE). The topic is relevant to the field of wood modification, and the experimental approach is sound. However, several issues must be addressed before the manuscript is suitable for publication. Below is a detailed review.

 

1. The manuscript does not justify why specific concentration ranges were chosen (e.g., why reduce from CA 38%/PEG400 40% to 19%/20%, or why use 3% SHP specifically). Similarly, the choice of curing times (3, 5, 7, 9 hours) and temperatures (120-180°C) appears somewhat arbitrary without reference to prior kinetic studies or preliminary optimisation.
2. The manuscript states that "ten specimens of one modification" were used for each ASE cycle, and standard deviations are provided. However, the number of replicates for WPG and bulking measurements at cycle 0 is not explicitly stated.
3. The manuscript speculates about the esterification mechanism (anhydride formation, pH effects) but does not provide direct evidence. The pH measurements (Table 2) are useful but limited (only three formulations, only with wood dust).
4. You used PEG400. Would higher molecular weight PEG (e.g., PEG1000, PEG1500) provide better bulking and reduced leaching due to larger molecular size, or would diffusion into the cell wall become limiting? Have you considered this in the context of your results?

Author Response

Reviewer #1

1. The manuscript does not justify why specific concentration ranges were chosen (e.g., why reduce from CA 38%/PEG400 40% to 19%/20%, or why use 3% SHP specifically). Similarly, the choice of curing times (3, 5, 7, 9 hours) and temperatures (120-180°C) appears somewhat arbitrary without reference to prior kinetic studies or preliminary optimisation.

Has been added to the manuscript accordingly.

“SHP has been used as a catalyst in wood modification with CA in many studies and test-ed at various concentrations [4,11,12,22,28,29]. Hasan et al. [4] found that higher SHP concentrations result in a reduction in ASE. He et al. [30] and Guo et al. [31] used only 1.5 % and 1 % SHP, respectively. In the current study, 3 % SHP was used to avoid any potential negative effects caused by excessively high concentrations. Furthermore, it is expensive and not environmentally friendly, meaning its use is generally called into question [12,32].” (p. 3, ll. 87-93)

“For each acid, a variant with half the chemical concentration was also tested. This generally requires fewer chemicals, which can result in a final product with a lower density and, consequently, a reduction in costs.” (p. 3, ll. 103-105)

“Curing temperatures of 120 °C, 140 °C or 160 °C have often been used in the literature, par-ticularly in relation to the modification of wood with CA [4,11,12,22,28]. A curing temper-ature of 180 °C has also been investigated [4,33]. Curing times of 10 hours were predomi-nantly tested [1,4,11,34], but also 8 hours [29] or just 2.5 hours [4]. Against this back-ground, the corresponding curing temperatures were selected for the current study. With regard to duration, the times mentioned above were chosen in order to analyse potential differing effects.” (p. 3, ll. 109-116)

 

2. The manuscript states that "ten specimens of one modification" were used for each ASE cycle, and standard deviations are provided. However, the number of replicates for WPG and bulking measurements at cycle 0 is not explicitly stated.

The section has been generalised and the number of samples has been explained in detail.

“Specimens from beech sapwood with a size of 25x25x10 mm³ were used for the water leaching experiments. After the modification and curing of the specimens, ten cycles of weight-percent-gain (WPG), bulking and ASE have been carried out for each ten specimens of one modification. For one cycle, the samples were placed in demineralised (…)” (p.4, ll. 136-139)

“The process was repeated a total of ten times for WPG, bulking and ASE.” (p.4, ll. 145-146)

 

3. The manuscript speculates about the esterification mechanism (anhydride formation, pH effects) but does not provide direct evidence. The pH measurements (Table 2) are useful but limited (only three formulations, only with wood dust).

Has been added to the manuscript accordingly.

“The addition of 3 % [w/v] of SHP to the system of CA (38 % [w/v]):PEG400 (40 % [w/v]) led to a remarkable decrease in ASE in leaching experiments. The underlying mechanism for this behavior remains unclear. Feng et al. [12] emphasise that the mechanism of action and the results of SHP have not yet been fully clarified. Furthermore, the role of SHP is controversial. It has been observed that SHP may react competitively with anhydride, and that the resulting stable acylphosphinates reduce the degree of esterification [12]. Morris et al. [44] point out that the catalytic effect of hypophosphite on the reaction between anhydride and cellulose remains unclear.” (pp. 8-9, ll. 286-293)

“At present, there is no evidence to suggest that cross-linking between CA and the wood components actually occurs [50]. However, Christ et al. [51] found indices of cross-linking between acid, PEG and wood components by investigating the impact bending strength and elongation. The impact bending strength has decreased significantly, which could be attributed to the cross-linked structure that increases the stiffness of the wood and hinders the free movement of the microfibrils [52]. Elongation also decreased, suggesting a cross-linked wood structure.” (pp. 9-10, ll. 334-340)

 

4. You used PEG400. Would higher molecular weight PEG (e.g., PEG1000, PEG1500) provide better bulking and reduced leaching due to larger molecular size, or would diffusion into the cell wall become limiting? Have you considered this in the context of your results?

PEG3000 and PEG8000 were used in previous studies: PEG 400 delivered significantly better results in terms of WPG, bulking and ASE. Furthermore, dimensional stability decreases as molecular weight increases, and higher processing temperatures are required.

Reviewer 2 Report

Comments and Suggestions for Authors

The paper is a highly significant step towards the elaboration of wood modification processes since it provides a comprehensive examination of the effects of different parameters of the process of beech wood esterification with PEG400 and some chosen carboxylic acids. The authors provide a sound scientific concept of the research due to their focus on three main factors (WPG, bulking, ASE) that objectively define the effectiveness of the process of stabilization. It is especially helpful to know the comparative characteristics of BTCA, CA, and MA that significantly vary in their chemical activity and thus give quite different results in the process of wood modification. One can conclude from the data provided that the positive effect of SHP catalyst occurs in BTCA/PEG systems only, which is a highly relevant conclusion for the technology. One of the drawbacks of this study is that it does not offer thorough chemical analysis of the products formed, which prevents complete validation of the suggested cross-linking processes. However, this publication has significant practical significance and can serve as an excellent basis for future studies on environmentally friendly wood modification techniques.

 

Questions:

1. What specific chemical processes lead to the differences in the effectiveness of cross-linking of BTCA, CA, and MA, taking into account their ASE and bulking indices, without providing an analysis of the structures formed as a result of the reaction?
2. How did the researchers manage the decomposition of components of wood during heating and curing, especially at 160-180 °C, when hemicelluloses tend to degrade?
3. What is the reason for the positive impact of the SHP catalyst on BTCA/PEG blends, and why is it not effective in blends of CA and MA, and do you see evidence of the different kinetics of reactions in these blends?
4. To what degree can the results of studies on certain reagents' ratios be applied to other compositions and other kinds of wood, taking into consideration the significant impact of the ratio between PEG and acids on the results?

Author Response

Reviewer #2

1. What specific chemical processes lead to the differences in the effectiveness of cross-linking of BTCA, CA, and MA, taking into account their ASE and bulking indices, without providing an analysis of the structures formed as a result of the reaction?

Has been added to the manuscript accordingly.

“The addition of 3 % [w/v] of SHP to the system of CA (38 % [w/v]):PEG400 (40 % [w/v]) led to a remarkable decrease in ASE in leaching experiments. The underlying mechanism for this behavior remains unclear. Feng et al. [12] emphasise that the mechanism of action and the results of SHP have not yet been fully clarified. Furthermore, the role of SHP is controversial. It has been observed that SHP may react competitively with anhydride, and that the resulting stable acylphosphinates reduce the degree of esterification [12]. Morris et al. [44] point out that the catalytic effect of hypophosphite on the reaction between anhydride and cellulose remains unclear.” (pp. 8-9, ll. 286-293)

“At present, there is no evidence to suggest that cross-linking between CA and the wood components actually occurs [50]. However, Christ et al. [51] found indices of cross-linking between acid, PEG and wood components by investigating the impact bending strength and elongation. The impact bending strength has decreased significantly, which could be attributed to the cross-linked structure that increases the stiffness of the wood and hinders the free movement of the microfibrils [52]. Elongation also decreased, suggesting a cross-linked wood structure.” (pp. 9-10, ll. 334-340)

“The multifunctional nature of BTCA [36,37] likely promotes the formation of a three-dimensional cross-linked structure involving both cellulose hydroxyl groups and PEG chains, leading to increased dimensional stability [27]. However, the observed losses in WPG during leaching suggest that not all incorporated material is covalently bound, indicating a coexistence of chemically anchored and physically retained fractions.” (p. 5, ll. 188-193)

“In general, the lower performance of MA can be attributed to the limited number of carboxyl groups, which restricts the formation of anhydride intermediates and reduces the probability of effective cross-linking [13,54,55]. Conversely, the intermediate behavior of CA reflects its trifunctional nature [56], which enables network formation.” (p. 11, ll. 386-389)

 

2. How did the researchers manage the decomposition of components of wood during heating and curing, especially at 160-180 °C, when hemicelluloses tend to degrade?

This has been explained in the relevant section and supported by sources.

“The optimal temperature of 160 °C likely reflects a balance between increased reaction ki-netics and the onset of thermal degradation of wood constituents. At lower temperatures, insufficient activation limits esterification, whereas at higher temperatures, degradation of hemicelluloses and potential structural damage reduce the effectiveness of the modification [13,42,43].” (p. 6, ll. 219-223)

 

3. What is the reason for the positive impact of the SHP catalyst on BTCA/PEG blends, and why is it not effective in blends of CA and MA, and do you see evidence of the different kinetics of reactions in these blends?

This has been explained in the relevant section and supported by sources.

“The addition of 3 % [w/v] of SHP to the system of CA (38 % [w/v]):PEG400 (40 % [w/v]) led to a remarkable decrease in ASE in leaching experiments. The underlying mechanism for this behavior remains unclear. Feng et al. [12] emphasise that the mechanism of action and the results of SHP have not yet been fully clarified. Furthermore, the role of SHP is controversial. It has been observed that SHP may react competitively with anhydride, and that the resulting stable acylphosphinates reduce the degree of esterifica-tion [12]. Morris et al. [44] point out that the catalytic effect of hypophosphite on the reaction between anhydride and cellulose remains unclear.” (po. 8-9, ll. 286-293)

“While lower pH values favor anhydride formation, excessively acidic conditions may hinder subsequent reactions with cellulose or promote hydrolytic degradation, thereby reducing overall modification efficiency. CA can often create highly acidic conditions, causing the reaction to become atuocatalytic [12].” (p. 9, ll. 298-301)

 

4. To what degree can the results of studies on certain reagents' ratios be applied to other compositions and other kinds of wood, taking into consideration the significant impact of the ratio between PEG and acids on the results?

The results cannot be applied directly, as the literature often refers to different chemicals or concentrations, different curing conditions, etc. Nevertheless, certain correlations and trends can be identified. In our investigations, we tested both beech and pine wood, and both types of wood yielded comparable results. These types of wood are frequently used for modification in the literature.

Reviewer 3 Report

Comments and Suggestions for Authors

Critical Analysis

The manuscript presents a substantial experimental dataset; however, the Results and Discussion section remains predominantly descriptive, which significantly limits its scientific contribution. Throughout the text, the authors systematically report trends in WPG, bulking, and ASE without establishing clear mechanistic interpretations or structure–property relationships. For instance, in the BTCA system, the statement that combinations containing PEG and SHP “achieved the highest values” is merely observational and does not explain the underlying physicochemical mechanisms governing these results. This pattern is repeated across all systems, where improvements in measured parameters are described but not critically analyzed in terms of crosslinking density, chemical fixation, or leaching resistance.

A central issue is the absence of differentiation between physical and chemical contributions to wood modification. Bulking is frequently interpreted as a positive outcome, yet no distinction is made between dimensional changes caused by physically retained PEG and those resulting from covalent network formation. This becomes particularly problematic when discussing WPG and ASE, as high WPG values are implicitly treated as indicators of successful modification, despite clear evidence of significant losses during leaching cycles. The manuscript fails to address the well-established concept that mass gain alone does not reflect modification efficiency, especially in systems involving water-soluble components such as PEG.

The discussion of catalytic effects further exemplifies the lack of mechanistic depth. In the CA/PEG system, the authors explicitly state that the negative effect of SHP is “not clear” , which is insufficient for a study positioned as an optimization work. Although pH effects and literature references are briefly mentioned, these are not integrated into a coherent explanation. The manuscript does not adequately explore how SHP influences the balance between anhydride formation, esterification kinetics, and potential hydrolytic degradation. As a result, the catalytic behavior appears empirical rather than chemically justified.

Similarly, the curing optimization lacks a reaction-based interpretation. The selection of 160 °C as the optimal temperature is presented as an experimental outcome, but the discussion does not address the competing processes involved, such as enhanced esterification at elevated temperatures versus thermal degradation of hemicelluloses. This omission prevents the reader from understanding whether the observed trends are governed by reaction kinetics, diffusion limitations, or structural degradation of the substrate.

Another major limitation is the lack of integration between the different acid systems (BTCA, CA, and MA). These systems are discussed sequentially rather than comparatively, despite representing a clear gradient in functionality and expected crosslinking capacity. The manuscript misses the opportunity to establish a unifying framework linking molecular structure (number of carboxyl groups) to macroscopic performance. Consequently, the discussion remains fragmented and does not extract generalizable insights from the dataset.

In addition, the manuscript does not provide direct chemical evidence of esterification or crosslinking, relying entirely on indirect indicators such as WPG and ASE. Without spectroscopic or compositional validation, the conclusions regarding chemical anchoring remain speculative. This weakens the overall argument, particularly in a journal that expects mechanistic support for material modification claims.

Finally, although standard deviations are reported, no statistical analysis is performed to assess the significance of differences between formulations. This further reinforces the descriptive nature of the study and reduces the robustness of the conclusions.

 

Proposed Rewriting

The Results and Discussion section should be restructured to move beyond descriptive reporting and toward a mechanistically grounded interpretation of the data. Instead of presenting WPG, bulking, and ASE as independent outcomes, the discussion should explicitly differentiate between physical swelling effects and chemical crosslinking contributions. In this context, bulking should be interpreted as a combined result of polymer penetration and structural expansion, while ASE retention after leaching should be used as a more reliable indicator of effective chemical fixation. This distinction allows for a more rigorous evaluation of modification efficiency, particularly in systems containing water-soluble PEG.

For the BTCA-based systems, the improved performance observed in the presence of PEG and SHP should be interpreted in terms of enhanced network formation via esterification reactions. The multifunctional nature of BTCA likely promotes the formation of a three-dimensional crosslinked structure involving both cellulose hydroxyl groups and PEG chains, leading to increased dimensional stability. However, the observed losses in WPG during leaching suggest that not all incorporated material is covalently bound, indicating a coexistence of chemically anchored and physically retained fractions. This dual contribution should be explicitly discussed to clarify the relationship between initial mass gain and long-term stability.

In the case of the CA and MA systems, the discussion should emphasize the role of acid functionality in determining crosslinking efficiency. The lower performance of MA-based formulations can be attributed to its limited number of carboxyl groups, which restricts the formation of anhydride intermediates and reduces the probability of effective crosslinking. Conversely, the intermediate behavior of CA reflects its trifunctional nature, enabling network formation but with lower density compared to BTCA. Framing the results within this structure–function relationship would significantly strengthen the scientific interpretation and provide a unifying perspective across all systems.

The catalytic effect of SHP should be reinterpreted through its influence on reaction conditions rather than treated as an empirical variable. The addition of SHP modifies the pH and ionic environment of the system, which in turn affects both anhydride formation and esterification reactions. While lower pH values favor anhydride formation, excessively acidic conditions may hinder subsequent reactions with cellulose or promote hydrolytic degradation, thereby reducing overall modification efficiency. This balance should be explicitly discussed to explain why SHP enhances performance in BTCA systems but negatively affects CA and MA formulations.

The analysis of curing conditions should also incorporate a reaction-based perspective. The optimal temperature of 160 °C likely reflects a balance between increased reaction kinetics and the onset of thermal degradation of wood components. At lower temperatures, insufficient activation limits esterification, whereas at higher temperatures, degradation of hemicelluloses and potential structural damage reduce the effectiveness of the modification. This interpretation provides a more comprehensive understanding of the observed trends and aligns the discussion with established principles of thermal chemistry in lignocellulosic systems.

Finally, the manuscript should integrate all results into a comparative framework, highlighting the progressive influence of acid functionality and processing parameters on performance. By correlating WPG, bulking, and ASE retention across different systems, the authors can demonstrate that high initial mass gain does not necessarily translate into improved durability, and that effective crosslinking—rather than total uptake—is the key factor governing dimensional stability.

 

Comments on the Quality of English Language

The English language used in the manuscript is generally understandable and technically adequate; however, it requires improvement to meet the standards of a high-impact scientific publication. The writing is often repetitive and predominantly descriptive, which reduces clarity and weakens the scientific argumentation, particularly in the Results and Discussion section. Several sentences rely on generic expressions such as “the results showed that” or “it can be seen that,” which should be replaced by more direct and precise formulations. For example, instead of writing “the results showed that the addition of PEG had a great influence on the values,” a clearer and more concise alternative would be “the addition of PEG significantly influenced the measured parameters.”

There is also frequent use of vague or imprecise language that should be strengthened. Statements such as “the values were higher when PEG was added” should be reformulated more specifically, for instance: “the incorporation of PEG resulted in higher WPG and bulking values.” Redundancies are also present and should be avoided; expressions like “as already mentioned before” can be simplified to “the values were consistently lower in the absence of PEG.” Additionally, informal phrasing such as “at the moment, it is not clear why this effect occurred” should be replaced with a more appropriate scientific tone, for example: “the underlying mechanism for this behavior remains unclear.”

Many sentences are overly long and loosely structured, which affects readability. For instance, a sentence like “it was observed that when the concentration was reduced, the values also decreased and the losses were higher, which could possibly be explained by the lower amount of chemicals present in the system” can be improved by writing “reducing the concentration led to lower values and higher losses, likely due to insufficient chemical content for effective network formation.” Ambiguous pronouns are also frequently used and should be clarified; instead of “this shows that it is important for the results,” a more precise formulation would be “this indicates that PEG plays a critical role in bulking and WPG.”

The manuscript would further benefit from stronger logical connections and more precise scientific wording. For example, rather than stating “however, the values were lower. This is because of the concentration,” it is preferable to write “however, the lower values can be attributed to the reduced concentration of reactive species.” Similarly, qualitative expressions such as “the values were quite similar” should be replaced with more rigorous statements, such as “the values were comparable within the experimental standard deviation.” Finally, the excessive use of passive voice should be reduced where possible; for example, “it was observed that the samples were affected by the catalyst” can be improved to “the catalyst affected the samples.”

Author Response

Reviewer #3

1. For the BTCA-based systems, the improved performance observed in the presence of PEG and SHP should be interpreted in terms of enhanced network formation via esterification reactions. The multifunctional nature of BTCA likely promotes the formation of a three-dimensional crosslinked structure involving both cellulose hydroxyl groups and PEG chains, leading to increased dimensional stability. However, the observed losses in WPG during leaching suggest that not all incorporated material is covalently bound, indicating a coexistence of chemically anchored and physically retained fractions. This dual contribution should be explicitly discussed to clarify the relationship between initial mass gain and long-term stability.

This has been explained in the relevant section and supported by sources.

“The multifunctional nature of BTCA [36,37] likely promotes the formation of a three-dimensional cross-linked structure involving both cellulose hydroxyl groups and PEG chains, leading to increased dimensional stability [27]. However, the observed losses in WPG during leaching suggest that not all incorporated material is covalently bound, indicating a coexistence of chemically anchored and physically retained fractions.” (p. 5, ll. 188-193)

 

2. In the case of the CA and MA systems, the discussion should emphasize the role of acid functionality in determining crosslinking efficiency. The lower performance of MA-based formulations can be attributed to its limited number of carboxyl groups, which restricts the formation of anhydride intermediates and reduces the probability of effective crosslinking. Conversely, the intermediate behavior of CA reflects its trifunctional nature, enabling network formation but with lower density compared to BTCA. Framing the results within this structure–function relationship would significantly strengthen the scientific interpretation and provide a unifying perspective across all systems.

This has been explained in the relevant section and supported by sources. Density has been deliberately omitted, as the product data sheets for the BTCA used do not provide any information on density.

“In general, the lower performance of MA can be attributed to the limited number of carboxyl groups, which restricts the formation of anhydride intermediates and reduces the probability of effective cross-linking [13,54,55]. Conversely, the intermediate behavior of CA reflects its trifunctional nature [56], which enables network formation.” (p. 11, ll. 386-389)

 

3. The catalytic effect of SHP should be reinterpreted through its influence on reaction conditions rather than treated as an empirical variable. The addition of SHP modifies the pH and ionic environment of the system, which in turn affects both anhydride formation and esterification reactions. While lower pH values favor anhydride formation, excessively acidic conditions may hinder subsequent reactions with cellulose or promote hydrolytic degradation, thereby reducing overall modification efficiency. This balance should be explicitly discussed to explain why SHP enhances performance in BTCA systems but negatively affects CA and MA formulations.

This has been explained in the relevant section and supported by sources.

“The addition of 3 % [w/v] of SHP to the system of CA (38 % [w/v]):PEG400 (40 % [w/v]) led to a remarkable decrease in ASE in leaching experiments. The underlying mechanism for this behavior remains unclear. Feng et al. [12] emphasise that the mechanism of action and the results of SHP have not yet been fully clarified. Furthermore, the role of SHP is controversial. It has been observed that SHP may react competitively with anhydride, and that the resulting stable acylphosphinates reduce the degree of esterification [12]. Morris et al. [44] point out that the catalytic effect of hypophosphite on the reaction between anhydride and cellulose remains unclear.” (po. 8-9, ll. 286-293)

“While lower pH values favor anhydride formation, excessively acidic conditions may hinder subsequent reactions with cellulose or promote hydrolytic degradation, thereby reducing overall modification efficiency. CA can often create highly acidic conditions, causing the reaction to become atuocatalytic [12].” (p. 9, ll. 298-301)

 

4. The analysis of curing conditions should also incorporate a reaction-based perspective. The optimal temperature of 160 °C likely reflects a balance between increased reaction kinetics and the onset of thermal degradation of wood components. At lower temperatures, insufficient activation limits esterification, whereas at higher temperatures, degradation of hemicelluloses and potential structural damage reduce the effectiveness of the modification. This interpretation provides a more comprehensive understanding of the observed trends and aligns the discussion with established principles of thermal chemistry in lignocellulosic systems.

This has been explained in the relevant section and supported by sources.

“The optimal temperature of 160 °C likely reflects a balance between increased reaction ki-netics and the onset of thermal degradation of wood constituents. At lower temperatures, insufficient activation limits esterification, whereas at higher temperatures, degradation of hemicelluloses and potential structural damage reduce the effectiveness of the modification [13,42,43].” (p. 6, ll. 219-223)

 

5. Finally, the manuscript should integrate all results into a comparative framework, highlighting the progressive influence of acid functionality and processing parameters on performance. By correlating WPG, bulking, and ASE retention across different systems, the authors can demonstrate that high initial mass gain does not necessarily translate into improved durability, and that effective crosslinking—rather than total uptake—is the key factor governing dimensional stability.

Has been integrated accordingly

“Finally, the correlation between WPG, bulking and ASE across different systems has demonstrated that a high initial mass gain does not necessarily translate into improved stability, and that effective cross-linking – rather than total uptake – is the key factor gov-erning dimensional stability.” (p. 13, ll. 433-436)

The text has been adapted at the relevant marked points.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors made the suggested corrections. The paper can be accepted for publication.