Classification of Monofloral Honeys by Measuring a Low-Cost Electronic Nose Prototype Based on Resistive Metal Oxide Sensors

Round 1

Reviewer 1 Report

This study presents an exciting topic; however, it lacks sufficient scientific justification for the sensing mechanism and the chemistry underlying the sensor's specificity toward different types of honey. This point should be clearly explained, as without it, I cannot recommend the publication of this paper.

The introduction should also provide backgrounds in VOC detection with metal oxide sensors and compare them with the aim of this paper.

 

The statement "Although the sensors are not specific for detecting volatile organic compounds (VOCs) present in honey, the objective of the experiments was to evaluate if these VOCs can interact with the sensors as an interferent and thus obtain a chemical fingerprint that allows classifying each type of honey" is confusing. It is essential to explain how the sensors are not specific to an analyte but are used for classifying different types of honey. The authors should consider that honey fingerprints could be influenced by pH, moisture, and other parameters, as explained in Table 1. A clear and sound explanation of the sensing mechanism and how a single material can be used to understand multiple analytes, such as SnO2 as the sensing membrane and its role in differentiating honey, is needed. The authors' statement about each sensor being designed for specific chemicals or gases contradicts the abovementioned statement. Here: “Each sensor is designed to detect a type of gas, range of gases, or range of organic chemicals (Table 3). All of these sensors are sensitive to a wide variety of gases and can be configured with different equations to detect different compounds.”

·         A clear explanation of the specificity is required, along with how the sensors detect the unique chemical fingerprint of each honey and how they ensure that other sample parameters like moisture do not influence it.

·         A visual representation of the sensing mechanism, showing the relationship between VOCs and different honey types and which VOCs are involved, would greatly aid in understanding the specificity of the sensors towards each honey type.

·         A visual is required to show the experimental procedure.  

 

-          Figure 1 does not provide enough information to understand the device's working principle (both physically and chemically). The authors should explain the surface chemistry and specificity in more detail.

 

-          In the abstract's use of "MQ-type metal oxide" is not adequately explained, and there is no prior information about MQ. Too early to bring this up, rasie this question what is MQ-type?

 

-          The figure captions need revision to ensure alignment with MDPI guidelines.

 

-          Figure 7 has a low resolution, making the text unclear. Additionally, Figure 3 lacks scaling of the y-axis, resulting in poor resolution.

 

-          Page 7, RAW ?

 

 

-          Please provide references that used metal oxide for VOCs detection; maybe it helps understand the specificity of this work. 

The English need to be revised as I had difficulty understanding the intended meaning of a few paragraphs. 

 

 

 

Author Response

This study presents an exciting topic; however, it lacks sufficient scientific justification for the sensing mechanism and the chemistry underlying the sensor's specificity toward different types of honey. This point should be clearly explained, as without it, I cannot recommend the publication of this paper.

Response 1: Dear reviewer, thanks for your observations. This aspect has been answered in page 5, 2.2. (green and grey). The resistive sensors are very common in enoses.

The introduction should also provide backgrounds in VOC detection with metal oxide sensors and compare them with the aim of this paper.

Response 2: The modifications for this part has been included in page 2 (color greeen and yellow). The sensors are not detecting a specific VOC, they are obtaining a range of VOCs and give a specific response, depending on the type of honey. This device is conceived for a quick response after a previous calibration.

The statement "Although the sensors are not specific for detecting volatile organic compounds (VOCs) present in honey, the objective of the experiments was to evaluate if these VOCs can interact with the sensors as an interferent and thus obtain a chemical fingerprint that allows classifying each type of honey" is confusing. It is essential to explain how the sensors are not specific to an analyte but are used for classifying different types of honey. The authors should consider that honey fingerprints could be influenced by pH, moisture, and other parameters, as explained in Table 1. A clear and sound explanation of the sensing mechanism and how a single material can be used to understand multiple analytes, such as SnO2 as the sensing membrane and its role in differentiating honey, is needed. The authors' statement about each sensor being designed for specific chemicals or gases contradicts the abovementioned statement. Here: “Each sensor is designed to detect a type of gas, range of gases, or range of organic chemicals (Table 3). All of these sensors are sensitive to a wide variety of gases and can be configured with different equations to detect different compounds.”

Response 3: The modifications for this part has been included in page 5 (color grey). This response is in the line of the previous response.

• A clear explanation of the specificity is required, along with how the sensors detect the unique chemical fingerprint of each honey and how they ensure that other sample parameters like moisture do not influence it.

Response 4: This has been included in page 5 (color green and grey). Again, the sensors are not specifics. They only provide a specific signal for the honey variety. The statistical management is developed for the calibration.

• A visual representation of the sensing mechanism, showing the relationship between VOCs and different honey types and which VOCs are involved, would greatly aid in understanding the specificity of the sensors towards each honey type.

Response 5: This has been included in page 6 (figure 1 and figure 4). This response is in the lines of the two previous responses.

• A visual is required to show the experimental procedure.  

Response 6: This has been included in page 6 (figure 1 and figure 4). All the experimental procedure is included on these figures.

-          Figure 1 does not provide enough information to understand the device's working principle (both physically and chemically). The authors should explain the surface chemistry and specificity in more detail.

Response 7: This has been included in page 6 (figure 1 and figure 4). This is in the same line than the previous response.

-          In the abstract's use of "MQ-type metal oxide" is not adequately explained, and there is no prior information about MQ. Too early to bring this up, rasie this question what is MQ-type?

Response 8: This has been included in page 2 (green and yellow).

-          The figure captions need revision to ensure alignment with MDPI guidelines.

Response 9: All the tittles have been changed.

-          Figure 7 has a low resolution, making the text unclear. Additionally, Figure 3 lacks scaling of the y-axis, resulting in poor resolution.

Response 10: These figures has been changed in figure 6 and 10. This is related to the statistical calculation.

-          Page 7, RAW ?

Response 11: The raw data are the not processed data. This is clariffied in page 8..

•          Please provide references that used metal oxide for VOCs detection; maybe it helps understand the specificity of this work.

Response 12: This has been included in page 2. 

Thanks again for your collaboration.

Reviewer 2 Report

The article is interesting, however it needs some further polishing:

1. Explain in more details the improvements of electronic detection system compared to previous design. In addition; define some specific parameters of the electronic system like sensotovoty etc.. 

2. For completeness  please provide some further information of the sensors, like for example change of resistance at the presence of specific gas etc

3. The use of PVC in electronic nose is problematic due to outgasing. This is valid for the connection tubes as well as the chanmber where the sensors are. Have you considered using different material and test the snesor to see the influence of the outgasing?

4. In the sentence (Page 5): The device has a 0.4 lpm air pump, The units of lpm are not clear (are thel liter per minute: l/min?)

5.  Null gas measurements ? What does this mean ? Wouldnt it be better to  use nitrogen?

6.  Shortly explain and add references for PCA and SVM for completness for the readers who are not experts in the field.

7. Formatting suggestion: subparagraph title "3.1. Results of the sesond experiment " should be put in a new line.

8. Page 12 below Table 4 there is a text that in a list explains: 1. Precision: , Confusion matrix, ... etc. The whole list is somehow out of context. Please add some explanation in corresponding paragraph and where this list belongs.

9. 

English language is fine; no comments.

Author Response

The article is interesting, however it needs some further polishing:

1. Explain in more details the improvements of electronic detection system compared to previous design. In addition; define some specific parameters of the electronic system like sensotovoty etc.. 

Response 1: This has been clarified in page 7. Moreover, figure 3 has been included in order to clarify the electronic improvements.

2. For completeness  please provide some further information of the sensors, like for example change of resistance at the presence of specific gas etc

Response 2: This has been clarified in page 5. The sensors are not specifics for one VOC. The combination among several sensors are giving a specific signal for each honey.

3. The use of PVC in electronic nose is problematic due to outgasing. This is valid for the connection tubes as well as the chanmber where the sensors are. Have you considered using different material and test the snesor to see the influence of the outgasing?

Response 3: This has been clarified in page 7. This is not vey prioritary because of the low flow and the minimal quantity analysed. Nevertheless, for a commercial purpose and new developments, new materials hve to be included.

4. In the sentence (Page 5): The device has a 0.4 lpm air pump, The units of lpm are not clear (are thel liter per minute: l/min?)

Response 4: This has been changed.

5.  Null gas measurements ? What does this mean ? Wouldnt it be better to  use nitrogen?

Response 5: This has been changed in page 7. The idea is that the reference value is with fresh air and no substance.

6.  Shortly explain and add references for PCA and SVM for completness for the readers who are not experts in the field.

Response 6: This has been changed in the text. Several references have been included.

7. Formatting suggestion: subparagraph title "3.1. Results of the sesond experiment " should be put in a new line.

Response 7: This has been changed.

8. Page 12 below Table 4 there is a text that in a list explains: 1. Precision: , Confusion matrix, ... etc. The whole list is somehow out of context. Please add some explanation in corresponding paragraph and where this list belongs.

Response 8: This has been changed in page 14. Several references have been included.

Thanks for your collaboration.

Reviewer 3 Report

In this study an electronic nose prototype made up of eight MQ resistive metal oxide sensors has been used to analyze and botanically characterize different types of honey based on their chemical fingerprint. The topic and results of this manuscript are quite interesting, however a revision is recommended before the final acceptance of the manuscript. 

- The introduction section should provide a proper explanation of the choice of resistive sensors. It should also emphasize the novelty of the manuscript, as the authors utilized commercially available resistive sensors and conducted data analysis afterwards, showcasing their unique approach.

- Special efforts should be invested in improving the presentation of the manuscript to enhance its clarity and readability.

- For readers interested in this field, a detailed description of the detection mechanism is recommended.

- In the conclusion, the authors should highlight the significant contributions of their work, technology, or knowledge. They should also discuss the drawbacks of the prototype and provide insights into future prospects and directions for further research.

 

- Grammatical errors throughout the manuscript should be refined.

Editing of English language is required. 

Author Response

In this study an electronic nose prototype made up of eight MQ resistive metal oxide sensors has been used to analyze and botanically characterize different types of honey based on their chemical fingerprint. The topic and results of this manuscript are quite interesting, however a revision is recommended before the final acceptance of the manuscript. 

- The introduction section should provide a proper explanation of the choice of resistive sensors. It should also emphasize the novelty of the manuscript, as the authors utilized commercially available resistive sensors and conducted data analysis afterwards, showcasing their unique approach.

Response 1: This has been included in page 2. The novelty of the manuscript is the aplication of these sensors in a enose for honey classification and the changes in the device that are shown in figure 3.

- Special efforts should be invested in improving the presentation of the manuscript to enhance its clarity and readability.

Response 2: This has been changed.

- For readers interested in this field, a detailed description of the detection mechanism is recommended.

Response 3: This has been included in page 5.

- In the conclusion, the authors should highlight the significant contributions of their work, technology, or knowledge. They should also discuss the drawbacks of the prototype and provide insights into future prospects and directions for further research.

Response 4: This has been changed in Conclusions.

• Grammatical errors throughout the manuscript should be refined.

Response 5: This has been refined.

Thanks for your collaboration.

Round 2

Reviewer 1 Report

Perfect. Only one comment; please check the captions of the figure to comply with the MDPI standards.