Development of an o-COSAN Ion-Pair Complex-Modified PVC Membrane for Microconductometric Glyphosate Detection

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article proposes a PVC membrane microconductivity sensor modified with an [o-COSAN]⁻/glyphosate ion-pair complex for the detection of glyphosate. The research approach is novel, as it is the first time that conductometry has been applied to glyphosate detection. The sensor is simple to fabricate, low-cost, and fast-responding, and demonstrates good selectivity and stability. The article has a complete structure, and the data generally support the conclusions. However, there are some shortcomings in data presentation, methodological details, and depth of discussion that require revision and supplementation.
1. In the "3.1. FTIR Analysis" section, there may be an error in the assignment of the wavenumber. The original text states: "The band observed at 3051 cm⁻¹ corresponds to C–H stretching vibrations." However, the preceding sentence had just attributed the same "3051 cm⁻¹" to "N–H and acidic O–H stretching vibrations." Assigning two completely different functional groups to the same wavenumber is contradictory. The authors are requested to re-examine the FTIR spectrum and accurately assign each absorption peak. Typically, ~3050 cm⁻¹ is more likely to correspond to unsaturated C–H stretching vibrations, whereas N–H and O–H stretching vibrations usually appear in the broader 3200–3600 cm⁻¹ region. Please verify and correct accordingly.
2. In the '2.3. Fabrication of the Microconductivity Chip' section, it is mentioned that '5 μL aliquots were deposited twice.' However, it is not specified whether drying is required between the two depositions, nor is the final membrane thickness provided. Membrane thickness has a direct impact on the response time and sensitivity of the sensor. It is recommended to supplement the operational steps between the two depositions (e.g., whether drying is performed after the first deposition and under what conditions), and to provide measured data or estimated values of the membrane thickness if possible.
3. In the 3.2. Analytical Performance' section, the authors observed two linear response regions and attributed this to 'the gradual saturation of recognition sites at high concentrations.' This explanation is overly simplistic. For ion-exchange membranes, a bilinear response may arise from a change in the ion transport mechanism within the membrane (e.g., from diffusion-controlled to migration-controlled), or be related to variations in the Donnan exclusion effect. It is recommended that the authors cite relevant literature to provide a more in-depth mechanistic discussion of the bilinear response phenomenon, thereby enhancing the scientific depth of the article.
4. The paper states that 'the membrane exhibits good stability, maintaining its performance for at least three months at room temperature.' However, no data are provided in the Results and Discussion section to support this claim. It is recommended to supplement the long-term stability test data of the sensor over three months. For example, a plot showing the variation of the sensor's response signal over time (e.g., measured weekly) at a fixed concentration (e.g., 100 μM) could be presented. This is critical data to demonstrate the practical applicability of the sensor.
5. The abstract mentions data such as 'a repeatability of 8% (RSD)' and 'a reproducibility of 3% (RSD),' but the corresponding experimental descriptions and data are not found in the main text
6. The article conducts detection at pH 3.5 and explains in the introduction that the protonation of glyphosate at low pH facilitates ion-pair formation. However, no specific data on the effect of pH on the sensor response are presented. It is recommended to supplement a pH optimization experiment, showing the sensor response to a fixed concentration of glyphosate at different pH values (e.g., pH 2–6), to demonstrate that pH 3.5 is the optimal choice. This would make the experimental design more rigorous.
7. The membrane formulation (PVC, DOS, ion-pair complex) is fixed, but no explanation is provided as to why this specific ratio was chosen. It is recommended to supplement a simple membrane composition optimization experiment, for example, by varying the content of the ion-pair complex and observing its effect on the sensor sensitivity, in order to justify the selected formulation.
8. When conditions permit, it is valuable to supplement morphological data by taking TEM images, as they allow observation of the microscopic morphology, dispersion of the ion-pair complex, and the surface structure of the membrane. This helps in understanding the relationship between the sensor's performance and its structure.

Author Response

Reviewer #1

 The article proposes a PVC membrane microconductivity sensor modified with an [o-COSAN]⁻/glyphosate ion-pair complex for the detection of glyphosate. The research approach is novel, as it is the first time that conductometry has been applied to glyphosate detection. The sensor is simple to fabricate, low-cost, and fast-responding, and demonstrates good selectivity and stability. The article has a complete structure, and the data generally support the conclusions. However, there are some shortcomings in data presentation, methodological details, and depth of discussion that require revision and supplementation.

1. In the "3.1. FTIR Analysis" section, there may be an error in the assignment of the wavenumber. The original text states: "The band observed at 3051 cm⁻¹ corresponds to C–H stretching vibrations." However, the preceding sentence had just attributed the same "3051 cm⁻¹" to "N–H and acidic O–H stretching vibrations." Assigning two completely different functional groups to the same wavenumber is contradictory. The authors are requested to re-examine the FTIR spectrum and accurately assign each absorption peak. Typically, ~3050 cm⁻¹ is more likely to correspond to unsaturated C–H stretching vibrations, whereas N–H and O–H stretching vibrations usually appear in the broader 3200–3600 cm⁻¹ Please verify and correct accordingly.

It was a mistake. The sentence “The absorption peak observed at 3051 cm⁻¹ is attributed to the N–H and acidic O–H stretching vibrations, indicating the presence of protonated amino and hydroxyl groups.“ was canceled.

2. In the '2.3. Fabrication of the Microconductivity Chip' section, it is mentioned that '5 μL aliquots were deposited twice.' However, it is not specified whether drying is required between the two depositions, nor is the final membrane thickness provided. Membrane thickness has a direct impact on the response time and sensitivity of the sensor. It is recommended to supplement the operational steps between the two depositions (e.g., whether drying is performed after the first deposition and under what conditions), and to provide measured data or estimated values of the membrane thickness if possible.

A sentence was added lines 161-162:

“after each deposition, an air-drying for one hour was required. The final membrane thickness was 26±0.5 µm (Fig. 2A).”

Figure 2A. Measurements of the membrane thickness on the top of different electrodes

 

3. In the 3.2. Analytical Performance' section, the authors observed two linear response regions and attributed this to 'the gradual saturation of recognition sites at high concentrations.' This explanation is overly simplistic. For ion-exchange membranes, a bilinear response may arise from a change in the ion transport mechanism within the membrane (e.g., from diffusion-controlled to migration-controlled), or be related to variations in the Donnan exclusion effect. It is recommended that the authors cite relevant literature to provide a more in-depth mechanistic discussion of the bilinear response phenomenon, thereby enhancing the scientific depth of the article.

When the concentration exceeds 360 µM, the slope of the calibration curve decreases. This phenomenon is due to the decrease of Donnan exclusion potential, which decreases the perselectivity of the membrane when the concentration of glyphosate increases [26].  The same shape of calibration curve was obtained for the detection of tetracycline with a PVC/ [o-COSAN]⁻/tetracycline membrane [24], the charge-transfer resistance increasing when the concentration of tetracycline increases, and the slope of the calibration curves decreases when the concentration is higher than 0.1 ng/L.

 

4. The paper states that 'the membrane exhibits good stability, maintaining its performance for at least three months at room temperature.' However, no data are provided in the Results and Discussion section to support this claim. It is recommended to supplement the long-term stability test data of the sensor over three months. For example, a plot showing the variation of the sensor's response signal over time (e.g., measured weekly) at a fixed concentration (e.g., 100 μM) could be presented. This is critical data to demonstrate the practical applicability of the sensor.

Figure 7. Variation of the response of the microconductometric sensor for three months (glyphosate concentration: 100 µM)

 

 

5. The abstract mentions data such as 'a repeatability of 8% (RSD)' and 'a reproducibility of 3% (RSD),' but the corresponding experimental descriptions and data are not found in the main text

This point was added in the text (lines 214,216):

The device exhibits excellent analytical performance, with reproducibility of 3% for 5 measurements of the same concentration with the same sensor and repeatability of 8% (RSD) for the same concentration with 5 different sensors.

6. The article conducts detection at pH 3.5 and explains in the introduction that the protonation of glyphosate at low pH facilitates ion-pair formation. However, no specific data on the effect of pH on the sensor response are presented. It is recommended to supplement a pH optimization experiment, showing the sensor response to a fixed concentration of glyphosate at different pH values (e.g., pH 2–6), to demonstrate that pH 3.5 is the optimal choice. This would make the experimental design more rigorous.

The effect of pH on the sensitivity of detection of glyphosate (100 µM) was studied (Fig. 4). For pH values equal and lower than 3.5, the signal is high (5.45 µS); it decreases for pH values higher than 3.5, due to the negative charge of glyphosate.

Figure 4. Effect of pH value on the response in conductivity (glyphosate concentration: 100 µM)

 

 

7. The membrane formulation (PVC, DOS, ion-pair complex) is fixed, but no explanation is provided as to why this specific ratio was chosen. It is recommended to supplement a simple membrane composition optimization experiment, for example, by varying the content of the ion-pair complex and observing its effect on the sensor sensitivity, in order to justify the selected formulation.

The effect of the ion-pair complex content in the membrane was already optimized in previously published papers [24, 25]. The same content was kept.

8. When conditions permit, it is valuable to supplement morphological data by taking TEM images, as they allow observation of the microscopic morphology, dispersion of the ion-pair complex, and the surface structure of the membrane. This helps in understanding the relationship between the sensor's performance and its structure.

MEB observation of the PVC membrane is presented in Figure 2B, showing the homogeneity of the PVC membrane.

 

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript describes a new sensor for monitoring glyphosate, a common  herbicide, in the environment and is worthy of publication.

However, it is highly recommended that the authors present the data of real samples.

Minor point: line 151, 100 μm &nd

 

Author Response

Reviewer #2

This manuscript describes a new sensor for monitoring glyphosate, a common  herbicide, in the environment and is worthy of publication.

However, it is highly recommended that the authors present the data of real samples.

Data on spiked real samples are presented in §3.3.

“The practical applicability of the developed microconductometric sensor was assessed through the determination of glyphosate in two representative environmental matrices, namely tap water and river water. Before analysis, the pH of both samples was adjusted to 3.5 to ensure optimal sensing conditions. To evaluate the analytical performance of the sensor, the water samples were spiked with glyphosate at two concentration levels (100 µM and 500 µM), representing both low and relatively high contamination scenarios.

The obtained results, summarized in Table 2, demonstrated excellent analytical performance, with an average recovery of 102%. These recovery values indicate that the sensor provides accurate and reliable quantification of glyphosate in both water matrices, with negligible interference from the sample composition. The close agreement between the added and measured concentrations confirms the robustness of the sensing platform and highlights its capability to operate effectively in real environmental samples without significant matrix effects.

These findings demonstrate the suitability of the proposed microconductometric sensor for practical environmental monitoring. Its satisfactory recovery, combined with its rapid response, simple operation, and low-cost fabrication, makes it a promising analytical tool for routine glyphosate determination in surface and drinking water samples.”

 

Table 2. Application of the developed sensor for glyphosate detection in water samples.

	Added (µM)	Found (µM)	Recovery (%)
Tap water	100	99.9	99 ± 3
	500	510	102 ± 3
River water	100	100.2	102 ± 3
	500	515	103 ± 3

 

Reviewer 3 Report

Comments and Suggestions for Authors

Improve the manuscript according to the attached comments.

 

 

Comments for author File: Comments.pdf

Comments on the Quality of English Language

All in all, English might work well.

Author Response

Reviewer #3

1. In the introduction. The manuscript does not cite the herbicide limits of law to be used as a reference.

This point was cited in the introduction (lines 38-43)

“However, in 2005, the Food and Agriculture Organisation (FAO) reported that glyphosate and its major metabolite, aminomethylphosphonic acid (AMPA), are of potential toxicological concern, mainly as a result of accumulation of residues in the food chain. The FAO further states that the dietary risk of glyphosate and AMPA is unlikely if the maximum daily intake of 1 mg kg⁻¹ body weight (bw) is not exceeded.” 

2. The instrumentation to which the microconductimetric chip was connected should be added to the "Materials and Methods" section.

Instrumentation for conductometric measurements was added to the “Materials and Methods” Section (lines 156-157)

“The VigiZMeter conductometer developed by Covarians (Gif-sur-Yvette, France) was used for conductance measurements.”

 

3. Regarding characterization, only FTIR analysis is presented. An additional characterization technique, such as DSC or TGA, could be included to verify the properties of the PVC membrane. Furthermore, the FT-IR spectrum should be presented more effectively, with better identification of the peaks.

PVC membrane was characterized through FTIR analysis (Fig. 4A)

“Figure 4A presents the FTIR spectrum of PVC that displays the ν(C-H) at 2910-2960 cm^-1; ν(C-H) at 1329-1426 cm⁻¹; ν(C-C) at 1093-1250 cm⁻¹; ν(C-Cl) at 686 cm^-1. “

FT-IR spectrum of [o-COSAN]⁻/glyphosate ion-pair complex was completed with the wavenumber of the different bands.

 

4. Figure 3 shows a calibration curve. How were the concentrations chosen?

The concentrations were obtained following successive additions.

1. Figure 4 should be better presented and discussed.

The calibration curve was discussed as follows:

When the concentration exceeds 360 µM, the slope of the calibration curve decreases. This phenomenon is due to the decrease of Donnan exclusion potential, which decreases the perselectivity of the membrane when the concentration of glyphosate increases [26].  The same shape of calibration curve was obtained for the detection of tetracycline with a PVC/ [o-COSAN]⁻/tetracycline membrane [24], the charge-transfer resistance increasing when the concentration of tetracycline increases, and the slope of the calibration curves decreases when the concentration is higher than 0.1 ng/L.

 

2. In the manuscript, the practical applicability of the developed microconductometric sensor was assessed through the determination of glyphosate in two representative environmental matrices, namely tap water and river water. Prior to analysis, the pH of both samples was adjusted to 3.5 to ensure optimal sensing conditions. So, is a real water sample acidified before being introduced to the sensor? Real samples can contain a wide range of compounds; acidifying the sample alters it and could potentially create substances that interfere with the sensor. Please explain this concept in more detail.

The treatment of water samples was detailed (lines 332-342):

“Before analysis, these samples were filtered through a 0.45 µm filter, and then the pH of both samples was adjusted to 3.5 to ensure optimal sensing conditions for glyphosate. The acidification process could change the speciation of the compounds in these samples, but not for glyphosate; only its charge value could change.” The water samples were then spiked with glyphosate at two concentration levels (100 µM and 500 µM), representing both low and relatively high contamination scenarios.

The obtained results, summarized in Table 2, demonstrated excellent analytical performance, with an average recovery of 102%. These recovery values indicate that the sensor provides accurate and reliable quantification of glyphosate in both water matrices, with negligible interference from the sample composition.”

7. The manuscript refers to the literature, the comparison reveals that nanostructured electrochemical sensors generally achieve detection limits in the pico- or nanomolar range because of their enhanced electrocatalytic activity and large effective surface area. What is the active area of the device?

The active surface area of the microconductometric sensors is equal to the interelectrode area: 14 mm².

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Most of the authors' revisions have been incorporated, but the specific line numbers and figures mentioned in the response letter do not match the descriptions, making the review process quite troublesome. It is hoped that the authors will pay attention to this issue to reduce the reviewers' time. Additionally, in Comment 8, the authors stated that they revised Section 2B, but it was actually Section 3B. The original text described it as "SEM," so why has the figure caption changed to "MEB"? In Comment 5, there is only a description without supporting data, which is not rigorous. It is hoped that the authors will address the above issues so that the manuscript can be accepted for publication.Most of the authors' revisions have been incorporated, but the specific line numbers and figures mentioned in the response letter do not match the descriptions, making the review process quite troublesome. It is hoped that the authors will pay attention to this issue to reduce the reviewers' time. Additionally, in Comment 8, the authors stated that they revised Section 2B, but it was actually Section 3B. The original text described it as "SEM," so why has the figure caption changed to "MEB"? In Comment 5, there is only a description without supporting data, which is not rigorous. It is hoped that the authors will address the above issues so that the manuscript can be accepted for publication.Most of the authors' revisions have been incorporated, but the specific line numbers and figures mentioned in the response letter do not match the descriptions, making the review process quite troublesome. It is hoped that the authors will pay attention to this issue to reduce the reviewers' time. Additionally, in Comment 8, the authors stated that they revised Section 2B, but it was actually Section 3B. The original text described it as "SEM," so why has the figure caption changed to "MEB"? In Comment 5, there is only a description without supporting data, which is not rigorous. It is hoped that the authors will address the above issues so that the manuscript can be accepted for publication.Most of the authors' revisions have been incorporated, but the specific line numbers and figures mentioned in the response letter do not match the descriptions, making the review process quite troublesome. It is hoped that the authors will pay attention to this issue to reduce the reviewers' time. Additionally, in Comment 8, the authors stated that they revised Section 2B, but it was actually Section 3B. The original text described it as "SEM," so why has the figure caption changed to "MEB"? In Comment 5, there is only a description without supporting data, which is not rigorous. It is hoped that the authors will address the above issues so that the manuscript can be accepted for publication.Most of the authors' revisions have been incorporated, but the specific line numbers and figures mentioned in the response letter do not match the descriptions, making the review process quite troublesome. It is hoped that the authors will pay attention to this issue to reduce the reviewers' time. Additionally, in Comment 8, the authors stated that they revised Section 2B, but it was actually Section 3B. The original text described it as "SEM," so why has the figure caption changed to "MEB"? In Comment 5, there is only a description without supporting data, which is not rigorous. It is hoped that the authors will address the above issues so that the manuscript can be accepted for publication.Most of the authors' revisions have been incorporated, but the specific line numbers and figures mentioned in the response letter do not match the descriptions, making the review process quite troublesome. It is hoped that the authors will pay attention to this issue to reduce the reviewers' time. Additionally, in Comment 8, the authors stated that they revised Section 2B, but it was actually Section 3B. The original text described it as "SEM," so why has the figure caption changed to "MEB"? In Comment 5, there is only a description without supporting data, which is not rigorous. It is hoped that the authors will address the above issues so that the manuscript can be accepted for publication.

Comments for author File: Comments.pdf

Author Response

 

 

Author Response File: Author Response.pdf