Skip to Content
MDPIMDPI

Advanced

by
  • Shuai Quan 1,
  • Zheng Sun 1 and
  • Pingwen Ming 3
  • et al.

Reviewer 1: Anonymous Reviewer 2: Anonymous

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Abstract

  1. It is recommended to include the conductivity value of Nafion/SiO₂-3 membrane at 120°C. Few other important data can be included as well. (line 23,24)

Keywords

  1. The keyword “Fuel cells” can be replaced with “Proton exchange membrane fuel cell”. Moreover, the keyword “Nafion/SiO₂ composite membranes” is rather long. (line 31,32)

Introduction

  1. The authors should explain more about other promising methods for preparing polymer electrolyte membrane (PEM) which is able to be used at high temperature, such as sulfonated-based membranes (e.g., SPEEK) and ionic liquid (IL)-based membranes. Recently, conductive membranes for HT-PEMFC are prepared with polybenzimidazole (PBI). Such membranes can be employed at vey low level of humidity. (line 58-66)
  2. It should be clarified whether the authors used SiO₂ or SiO₂ nanoparticles (NPs) for the membrane preparation. (line 67-80)
  3. Why did the authors choose to work with SiO2 among other well-known nanoparticles, such as TiO2, GO, CuO, and AgNPs? This comment should be addressed. (line 67-80)

Materials and methods

  1. It is recommended to include a table at the end of Section 2.1 Membrane Fabrication to clearly present the membrane compositions. Additionally, a schematic figure could be added to visually illustrate the membrane preparation steps. (line 97-119)
  2. Why did the authors select SiO₂ concentrations of 1, 2, 3, 4, and 5 wt.%? The rationale behind choosing these specific loadings should be clarified. (line 97-119)
  3. The characterization methods could be explained in different subsections separately instead of including them all in one section. (line 121-151)

Results and discussion

  1. Have the authors assessed the presence of SiO2 NPs by scanning electron microscopy (SEM)? In this case, the authors can evaluate the dispersion of NPs on the membrane surface (and in the membrane structure by cross-section mode)? Moreover, the possibility of formation of agglomerations can be checked.
  2. Regarding the AFM analysis, the surface average roughness (Ra), as well as the root mean square roughness (Rq), should be reported for all membranes. (line 196-225)
  3. To obtain accurate information on the thermal stability of the membranes (including their thermal degradation temperatures), TGA analysis should be conducted over the standard temperature range of 25–800 °C. (line 322-344)
  4. Since the membranes are intended for use at elevated temperatures (100–140°C), the TGA analysis should also be performed in isothermal mode. Specifically, maintaining the temperature at 140°C for 24 h would allow evaluation of the membranes’ stability under these operating conditions. (line 322-344)
  5. Have the authors conducted proton conductivity measurements at 0% RH and elevated temperatures (100–140°C) to evaluate whether the membranes can overcome the known limitations of Nafion under these conditions? (line 348-380)
  6. The proton conductivity of the membranes should be measured in isothermal mode–maintaining the temperature at 120 °C for at least 24 h–to verify their long‑term stability. (line 348-380)

 

Author Response

Abstract

1. It is recommended to include the conductivity value of Nafion/SiO₂-3 membrane at 120°C. Few other important data can be included as well. (line 23,24)

Reply: We thank the reviewer for this important comment. The proton conductivity value of 61.9 mS·cm⁻¹ at 120 °C has been added to the Abstract (lines 23–24). In addition, other important data have also been included. Specifically, single-cell tests under MHT-PEMFC conditions (120 °C, 50% RH) demonstrate the practical efficacy of these membrane-level enhancements, with the Nafion/SiO₂-3 membrane exhibiting an open-circuit voltage and peak power density 11.2% and 8.9% higher, respectively, than those of pristine Nafion under identical MEA fabrication and operating conditions (lines 27–29).

Keywords

2. The keyword “Fuel cells” can be replaced with “Proton exchange membrane fuel cell”. Moreover, the keyword “Nafion/SiO₂ composite membranes” is rather long. (line 31,32)

Reply: The keyword “Fuel cells” has been replaced with “Proton exchange membrane fuel cell”, and the keyword “Nafion/SiO₂ composite membranes” has been revised accordingly to “Nafion/SiO₂ membranes” (line 34).

Introduction

3. The authors should explain more about other promising methods for preparing polymer electrolyte membrane (PEM) which is able to be used at high temperature, such as sulfonated-based membranes (e.g., SPEEK) and ionic liquid (IL)-based membranes. Recently, conductive membranes for HT-PEMFC are prepared with polybenzimidazole (PBI). Such membranes can be employed at very low level of humidity. (line 58-66)

Reply: We thank the reviewer for this valuable suggestion. In the revised manuscript, additional background information on other promising high-temperature PEM systems, including sulfonated-based membranes (e.g., SPEEK) and ionic liquid (IL)-based membranes, has been added to provide a more comprehensive overview of current HT-PEM development. In addition, the discussion of polybenzimidazole (PBI)-based membranes, which are widely used for HT-PEMFC operation under low-humidity conditions, has been clarified and strengthened in the Introduction. The corresponding comments have been addressed in the revised manuscript (lines 63-79 and 539-543).

4. It should be clarified whether the authors used SiO₂ or SiO₂ nanoparticles (NPs) for the membrane preparation. (line 67-80)

Reply: We thank the reviewer for this important comment. In this work, SiO₂ nanoparticles were used as the inorganic filler for membrane preparation. The term “SiO₂” in the manuscript refers specifically to SiO₂ nanoparticles throughout the text. To avoid ambiguity, the description has been clarified accordingly in the revised manuscript (line 84).

5. Why did the authors choose to work with SiO2 among other well-known nanoparticles, such as TiO2, GO, CuO, and AgNPs? This comment should be addressed. (line 67-80)

Reply: We thank the reviewer for this insightful comment. SiO₂ was selected as the inorganic nanofiller in this study due to its excellent thermal and chemical stability, strong hydrophilicity, and abundant surface silanol groups, which are particularly beneficial for proton transport and water retention under the high-temperature, low-humidity conditions relevant to MHT-PEMFC operation. In addition, SiO₂ exhibits good compatibility with PFSA-based membranes and does not introduce electronic conductivity or undesirable electrochemical side reactions. While other nanofillers such as TiO₂, GO, CuO, and AgNPs have also been reported to enhance membrane performance, they may suffer from limitations such as particle agglomeration, interfacial incompatibility, or potential electronic conductivity, which are less favorable for the present sol–gel-assisted membrane design. Therefore, SiO₂ represents a rational and well-established choice aligned with the objectives of this work. The corresponding comments have been addressed in the revised manuscript (lines 90-96).

Materials and methods

6. It is recommended to include a table at the end of Section 2.1 Membrane Fabrication to clearly present the membrane compositions. Additionally, a schematic figure could be added to visually illustrate the membrane preparation steps. (line 97-119)

Reply: In response to the reviewer’s suggestion, a table summarizing the compositions of all prepared membranes has been added at the end of Section 2.1 (Table 1) (lines 141-148). In addition, a schematic figure illustrating the sol–gel-assisted membrane preparation process has been included to provide a clearer visualization of the fabrication steps (Figure 1).

7. Why did the authors select SiO₂ concentrations of 1, 2, 3, 4, and 5 wt.%? The rationale behind choosing these specific loadings should be clarified. (line 97-119)

Reply：We thank the reviewer for this important question. The SiO₂ loadings of 1–5 wt.% were selected to systematically investigate the influence of inorganic filler content on membrane structure and performance while maintaining good processability and membrane integrity. Low SiO₂ contents (1–2 wt.%) were chosen to evaluate the initial effect of nanoscale fillers on proton transport and water retention without significantly disturbing the polymer matrix. Intermediate loadings (around 3 wt.%) were included because this range is commonly reported to promote the formation of effective proton-conducting pathways in organic–inorganic composite membranes. Higher loadings (4–5 wt.%) were examined to explore the upper limit of SiO₂ incorporation, where excessive filler content may lead to particle aggregation, increased membrane thickness, or compromised mechanical and electrochemical performance. This compositional range therefore enables a comprehensive assessment of both beneficial and limiting effects of SiO₂ incorporation.

The corresponding clarification has been added to the revised manuscript (Section 2.1, lines 133-136).

8. The characterization methods could be explained in different subsections separately instead of including them all in one section. (line 121-151)

Reply: We thank the reviewer for this helpful suggestion. In the revised manuscript, the description of the characterization methods has been carefully reorganized to improve clarity and readability. Although separate subsections were not introduced due to the concise nature of this section, each characterization technique is now described more clearly and sequentially, with explicit transitions to distinguish different methods. This revision ensures that the characterization procedures are easy to follow, without adding excessive structural complexity (lines 152-155, 157, 164-167, and176-182).

Results and discussion

9. Have the authors assessed the presence of SiO2 NPs by scanning electron microscopy (SEM)? In this case, the authors can evaluate the dispersion of NPs on the membrane surface (and in the membrane structure by cross-section mode)? Moreover, the possibility of formation of agglomerations can be checked.

Reply: We thank the reviewer for this valuable comment. The dispersion and possible agglomeration of SiO₂ nanoparticles were evaluated by large-area AFM height imaging, with scan sizes up to 10 μm × 10 μm, which are comparable to typical SEM observation scales (Figure S2). Representative AFM images for all SiO₂ loadings were provided in the revised version of Supporting Information, and additional large-area images have been included in the Supporting Information to further clarify nanoparticle dispersion at micrometer length scales (Figure S1).

Due to the low atomic number contrast between SiO₂ nanoparticles and the PFSA matrix, as well as the embedding of SiO₂ within the polymer bulk, SEM did not provide sufficient contrast to reliably assess nanoparticle dispersion or aggregation in this system. AFM was therefore selected as the primary technique. The AFM results indicate a relatively uniform distribution of SiO₂-related surface features without pronounced agglomeration, even at the highest SiO₂ loading. The corresponding explanation has been added to the revised manuscript (lines 237-239).

10. Regarding the AFM analysis, the surface average roughness (Ra), as well as the root mean square roughness (Rq), should be reported for all membranes. (line 196-225)

Reply: We thank the reviewer for this helpful suggestion. In response, the surface roughness parameters, including the average roughness (Ra) and the root mean square roughness (Rq), have now been quantitatively evaluated from the AFM height images for all membranes. The corresponding Ra and Rq values are summarized in the Supporting Information (Table S1).

The roughness analysis was performed using the AFM analysis software on multiple representative scan areas for each membrane (five different locations per sample). The results show that the incorporation of SiO₂ nanoparticles leads to moderate variations in surface roughness within the investigated loading range. This quantitative analysis provides additional support to the AFM height images discussed in the manuscript (lines 251-255).

11. To obtain accurate information on the thermal stability of the membranes (including their thermal degradation temperatures), TGA analysis should be conducted over the standard temperature range of 25–800 °C. (line 322-344)

Reply: We thank the reviewer for this comment. In this study, thermogravimetric analysis (TGA) was performed up to 270 °C, which was intentionally selected to sufficiently evaluate the thermal stability of the Nafion-based membranes within and well above the intended operating temperature range for MHT-PEMFC applications (up to 140 °C).

For perfluorosulfonic acid (PFSA) materials, thermal degradation at significantly higher temperatures has been extensively reported in the literature and occurs far beyond the practical operating window. Extending TGA measurements to extremely high temperatures (e.g., up to 800 °C) is therefore not necessary for assessing membrane stability under realistic fuel cell conditions. Moreover, degradation of PFSA at very high temperatures may generate fluorinated volatile species, which can contaminate the instrument and adversely affect measurement reliability.

Accordingly, the selected TGA temperature range up to 270 °C is sufficient to provide accurate and application-relevant information on the thermal stability of the membranes investigated in this work.

12. Since the membranes are intended for use at elevated temperatures (100–140°C), the TGA analysis should also be performed in isothermal mode. Specifically, maintaining the temperature at 140°C for 24 h would allow evaluation of the membranes’ stability under these operating conditions. (line 322-344)

Reply: We thank the reviewer for this insightful suggestion. Isothermal TGA at elevated temperatures (e.g., 140 °C for extended durations) can indeed provide valuable information on long-term thermal and chemical stability. However, for PFSA-based membranes, such prolonged isothermal exposure is known to induce not only mass loss but also complex physical and chemical processes, including hydrophilic domain relaxation, polymer chain degradation. These phenomena are typically associated with aging and durability studies rather than with initial thermal stability evaluation.

In fact, in our previous work on pristine PFSA membranes, we have systematically investigated the effects of prolonged high-temperature isothermal treatment and observed significant morphological evolution and chemical degradation after extended exposure [1]. Given that these processes fall beyond the scope of the present study, which focuses on the structure–property relationships of SiO₂-filled composite membranes under relevant operating temperatures, isothermal TGA measurements were not included here.

Instead, the thermal stability assessment in this work was designed to ensure that the membranes remain stable within and sufficiently above the intended operating temperature range, which is adequately addressed by the selected non-isothermal TGA measurements.

[1]   Guo, Q. S.; Feng, C.; Ming, P. W.; Zhang, C. M. Microstructural changes and proton transport behavior of perfluorosulfonic acid membranes for medium-to-high temperature fuel cells. Fuel 2025, 393, 135071.

 

13. Have the authors conducted proton conductivity measurements at 0% RH and elevated temperatures (100–140°C) to evaluate whether the membranes can overcome the known limitations of Nafion under these conditions? (line 348-380)

Reply: We thank the reviewer for this comment. Proton conductivity measurements at 0% RH were not conducted because proton transport in PFSA-based membranes remains fundamentally water-mediated, even with SiO₂ incorporation. The objective of this work is to extend the effective operating window of Nafion-based membranes toward higher temperatures and lower humidity, rather than to achieve anhydrous proton conduction. Therefore, conductivity measurements performed at controlled low-humidity conditions (50% RH) and elevated temperatures are more relevant for evaluating the performance of the present composite membranes.

14. The proton conductivity of the membranes should be measured in isothermal mode–maintaining the temperature at 120 °C for at least 24 h–to verify their long‑term stability. (line 348-380)

Reply: We thank the reviewer for this valuable suggestion. Isothermal proton conductivity measurements at elevated temperatures (e.g., maintaining 120 °C for extended durations) can indeed provide insights into long-term durability. However, for PFSA-based membranes, prolonged isothermal exposure at high temperatures is known to induce complex physical and chemical processes, including hydrophilic domain relaxation, chemical degradation of polymer chain, which directly affect proton conductivity. These effects are typically associated with aging and durability studies rather than with the evaluation of intrinsic membrane performance.

In our previous work on pristine PFSA membranes, we have systematically investigated the impact of long-term high-temperature isothermal treatment and observed pronounced morphological evolution and conductivity degradation. Given that such aging-related phenomena fall beyond the scope of the present study, which focuses on the structure–property relationships and performance enhancement of SiO₂-filled composite membranes under relevant operating conditions, isothermal conductivity measurements over extended durations were not included.

Instead, proton conductivity in this work was evaluated under controlled temperature and humidity conditions representative of practical operation to assess the immediate performance benefits of the composite membrane design.

 

Reviewer 2 Report

Comments and Suggestions for Authors

In this manuscript, the authors prepared Nafion/SiO₂ composite membranes with systematically varied filler contents. The authors investigated the prepared membranes with many techniques such as XOS, XRD, FTIR, AFM, conductivity, tensile test, and fuel cell performance. This study elucidates a clear structure–property–transport relationship in SiO₂-reinforced polymer electrolyte membranes, The interpretations of the results were deeply discussed. The quantity and quality of the figures are appropriate. We believe that this manuscript is very important for studying the way to control inorganic incorporation is a robust strategy. This will be useful for extending the operational temperature of the proton exchange membranes for fuel cell applications.

 

Summary: I recommend the publication of this manuscript after considering my comments as:

 

  • Line 192, the shift of the XPS peaks with increasing the SiO2 concentration should be discussed on the XPS data.
  • Line 282, the XRD trend for the sample with 3% concentration is different compared with the other samples, the authors should discuss this point? The main peak at 17O can be deconvolution into two peaks similar to the references (Journal of Solid State Electrochemistry (2020) 24:1217–1229 & J Solid State Electrochem (2019) 23:2639-2656.
  • Line 345, it is better to present TGA data as derivative (dW/DT) verves temperature and discuss the graph.

 

Author Response

1. Summary: I recommend the publication of this manuscript after considering my comments as:

Reply: We sincerely thank the reviewer for this positive and encouraging comment.

2. Line 192, the shift of the XPS peaks with increasing the SiO2 concentration should be discussed on the XPS data.

Reply: We thank the reviewer for this comment. In the revised manuscript, the binding energy shifts observed in the Si 2p and S 2p XPS spectra with increasing SiO₂ content have been explicitly discussed. The gradual increase in binding energy up to intermediate SiO₂ loadings is attributed to enhanced interfacial interactions between SiO₂ nanoparticles and the PFSA matrix, while the slight decrease at higher loading is associated with saturation and partial disruption of these interactions. This discussion has been added (lines 226-234).

3. Line 282, the XRD trend for the sample with 3% concentration is different compared with the other samples, the authors should discuss this point? The main peak at 17O can be deconvolution into two peaks similar to the references (Journal of Solid State Electrochemistry (2020) 24:1217–1229 & J Solid State Electrochem (2019) 23:2639-2656.

Reply: We thank the reviewer for this insightful comment. The distinct XRD trend observed for the Nafion/SiO₂–3 membrane indeed indicates a unique structural evolution at intermediate SiO₂ loading. The broad diffraction peak around 17° in PFSA-based membranes is commonly associated with the semi-crystalline organization of the polymer backbone and has been reported to consist of contributions from different structural domains.

At an intermediate SiO₂ content (3 wt%), the polymer–filler interfacial interactions are optimized, which can promote local chain rearrangement and microphase organization. This results in a modified peak shape in the 17° region compared with lower or higher SiO₂ loadings. Similar observations and interpretations have been reported in the literature (J. Solid State Electrochem., 2019, 23, 2639–2656; J. Solid State Electrochem., 2020, 24, 1217–1229).

Although peak deconvolution can provide further quantitative insight, the present discussion focuses on the overall structural trend and its correlation with membrane performance. The corresponding discussion and reference have been added to the revised manuscript (lines 313-317, 577-578).

4. Line 345, it is better to present TGA data as derivative (dW/DT) verves temperature and discuss the graph.

Reply: We thank the reviewer for this helpful suggestion. The TGA results have been complemented with derivative thermogravimetric (dW/dT) curves for all membranes, which are provided in the Supporting Information (Figure S4). The DTG curves allow clearer identification of weight-loss events and confirm that no significant thermal degradation occurs within the investigated temperature range relevant to membrane operation. The corresponding explanation has been added to the revised manuscript (lines 393-395).

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I would like to express my appreciation to the authors for this revision.