the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 May 2022
Performance comparison between electrochemical and semiconductors sensors for the monitoring of O3
Abstract. As part of the Quality of Life and Urban Mobility (MouVIE) Chair, an individual mobile sensor designed as an adaptable and scalable "platform" is being developed within the LATMOS (Atmospheres Space Observations Laboratory). This sensor must contribute to answering problems related to the exposure of individuals to air pollution and their impact on health. In this context, its adaptable and scalable nature will allow the insertion of new consumer measurement components available ("low cost" micro-sensors).
In this paper we present a laboratory evaluation of commercially sensors for the monitoring of ozone (O3). Two type of sensors are tested: electrochemical and semiconductors sensors. Theses sensors are tested at different temperatures, humidity and at ppb level. The voltage response and their dependence on ambiant temperature and humidity are evaluated. The time drift effect on electrochemical sensors was also evaluated during 4 months of use.
- Preprint
(976 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on amt-2022-75', Anonymous Referee #1, 07 Aug 2022
General comments
The objective of this work is to evaluate the performance of commercial -electrochemical and semiconductor type- low-cost sensors (LCS) to quantify tropospheric ozone under laboratory conditions. The experiments were carried out at different experimental conditions, focused on humidity and temperature effects, to evaluate the LCS output signal dependence on these variables. Given the global and local consequences of air pollution, the subject under study is of prime importance. The evolution of the field of low-cost sensors is essential as they could provide information at finer scales than ever before. However, is in the consideration of this reviewer that the present manuscript needs to be reformulated. An important point that should be extensively reviewed is the structure and clarity of the manuscript, which makes it difficult to follow the content and main points. The writing style needs to be revised as well. The literature review is somewhat scarce and needs to be reconsidered as, in this reviewer's opinion, it is impacting the novelty of the manuscript. This is evident in the results, as much of what the authors find has already been discussed in the scientific literature, neglecting potential impacts.
In relation to the abstract, although it mentions the main objective, including a short summary of results and conclusions will benefit the reader. The description of the problems/issues but also the advantages of these sensor types provided in the introduction need to be refreshed. The references here are somewhat outdated, and this in a field as dynamic as that of LCS seems to be impacting the results and conclusions novelty. I also think that the text in general would benefit if the main points made by the authors are clearly stated. This will also make it easier for the reader to identify and follow some important points that this work could potentially offer. As the current manuscript is focused on evaluating the sensor response (voltage signal) to different lab conditions, correction methods and colocation studies review included in the text dilutes the discussion of topics more relevant to this study. The experimental design is partially described, impacting the transparency and reproducibility of the study. A detailed description would also help to clarify the results and conclusions of the study. Regarding the results and the analysis of the data, it is chaotic, which dilutes important information. There is also a lot of duplicate information that could be summarized or outlined in a more efficient way.
Specific comments
Having a continuous numeration of lines will be greatly appreciated.
Abstract
Page 1
Line 14: Perhaps “answering questions” fits better the sentence.
Line 15: “In this context, its adaptable and scalable nature will allow the insertion of new consumer measurement components available ("low cost" micro-sensors)”. Please rephrase, is not clear what the authors want to convey.
Line 18: Replace “Theses” with “These” and “are tested” by “were tested”
Line 19: Replace “and at ppb level” by “and ppb level”, “ambiant" with “ambient"
Introduction
Page 1
Line 31: this paragraph seems to be the continuation of the previous one, which, however, is separated from the first. Also, it is provided with a reference from 2012 -ten years ago- meaning the assertion included in the text represents the technology at that time. Specific studies can be found for each of the named technologies (metal oxide, electrochemical, infrared, photo-ionization sensors) in the last 5-10 years. I strongly suggest diving into more recent literature. Also, what does “target gas” mean in this context?
Line 39: Despite starting this paragraph with “Recently”, a similar issue is found as mentioned in the previous comment. Also, would be interesting to know if the mentioned sensors still in the market and if the price corresponds to 2022.
Line 45: Please rephrase/revise the entire paragraph content.
Page 2
Line 19 (2nd page)-Line 6 (3rd page): The first paragraph in these lines describes correction methods, which seem to be a bit outside the scope of the manuscript. Similarly, the next 5 paragraphs are mainly focused on in-field colocation studies, while this is a lab study. I suggest rethinking and rephrasing what of all this is contributing to sustaining your study case and/or helping the reader to understand your work.
Page 3
Line 8: It is said that “We have developed our own sensor”. I strongly suggest replacing sensor by system (or node). Also, it would be enriching if you could expand on the differences (here or elsewhere) between an OEM and a system.
Experimental setup
Line 19: replace “prototype sensor” with “prototype system”
Line 20: A detailed description of the exposure chamber will benefit any reader trying to replicate the experiments presented here. This point is a bit vague and needs to be reframed as it is an essential part of the experiments. For example, how the different variables (temperature, RH, flow, pressure, etc) are controlled? How long was the ramp held each time?
Line 44 (3rd page) - Lin 8 (4th page): Please check the intended meaning of these paragraphs and rephrase.
Page 4
Line 10 – Line 21: these paragraphs offer an introductory description of the sensors type, not relevant to the experiment conditions description. These paragraphs could be summarised and moved to the introduction.
Table 1: probably the time range in hours could be more informative. What is the “Number Experimental test”? And “Number values on range”? Is not clear what is trying to be said. For the temperature and humidity columns, what is the number in between parentheses?
Lines 23-29: Please rephrase this paragraph, I couldn’t understand what is trying to communicate. Also, could you expand on the reasons behind the different experiments and periods? How many sensors of each type were tested at the same time?
Lines 30-36: Please check the content of the paragraph and rephrase.
Results
Page 5
Line 4-12: The text states that the differences between the different series are a consequence of changes in the sensor response (drift). Is it possible that part of the differences found are due to the experimental set-up? One way to provide evidence in this regard would be to have duplicate sensors exposed to the same conditions.
Line 26-33: what is the “room temperature”? And the “higher temperature”? Again, having duplicate senors would help to support the manuscript statements, but it doesn’t seem to be the case.
Line 39: “The voltage response increases with humidity”, but series #9 seems to be an outlier. What would be the explanation for this? Also, the Figure number should be stated.
Line 49: could you expand based on your findings about the observed hysteresis effect on the sensors?
Page 6
Lines 1-5: the manuscript states a good agreement exists between series #5 (black line, Figure 10) and series #7 (blue line). Please revise this, as it doesn’t seem to be the case.
Line 17: how the series were combined to produce the figure 12?
Line 31: it is not clear how the manufacturer curve shown in fig 14 was obtained.
Line 47: “The mean values of the voltage response between the series are close up to 200 mg/m3 but for values above 200 μg/m3 the quality of the measurements does not allow to obtain a reliable result”. Check the meaning of the sentence. Also, why it doesn’t allow to obtain a reliable result? Expand.
Page 7
Line 2: “The figure shows a strong drift between the values obtained during the first month of use and the 3rd and 4th months of use”. Is the positive change in the baseline shown in Fig 18, due only to drift? Could there be other effects?
Conclusions
Line 29: “the measurements precision is enough for air quality studies with European regulation” Could you expand on this? Qualitative or quantitative studies? There will be soon a new European Directive (with lower limit values in order to follow the new WHO guidelines). Do you think the electrochemical sensors will be able to cope with them?
Line 32: The text mentions “the linearity properties…”, but in all the figures the response was treated as non-linear.
Figures
Regarding the figures (5 to 18), in my opinion, they are too many and important information is diluted. I suggest selecting (and rearranging) the most pertinent figures for the main text and the rest could go to the supplementary. Another relevant comment is that the R-square for non-linear models is statistically incorrect.
Citation: https://doi.org/10.5194/amt-2022-75-RC1 -
CC1: 'Reply on RC1', christophe claveau, 20 Aug 2022
Thanks to Referee 1 for agreeing to review our article. We will modify it taking into account as much as possible the comments of Referee 1.
Christophe Claveau
Citation: https://doi.org/10.5194/amt-2022-75-CC1 -
AC1: 'Reply on RC1', christophe claveau, 11 Oct 2022
We thank again for the interest in our article.
Here are the answers we provide:
Specific comments: numeration of lines has been modified in the revised version of our article
Abstract
Page 1-Line 14: the text has been modified in the revised version of our article
Page 1 -Line 15: We have reworded the paragraph to show that the adaptable nature of our sensor will allow us to test the latest and most
micro sensors and help to answer questions related to pollution exposure.Page 1 -Line 18: the text has been modified in the revised version of our article
Page 1-Line 19::the text has been modified in the revised version of our article
Introduction
Page 1-line 31-45: The most recent and detailed studies on the performance of MOS (metal oxide sensors)
and electrochemical sensors and concerning ozone are between 2012 and 2018. It seemed important to us
to present a summary of the studies during these last decades on the subject. All the paragraph
has been rewritten to meet the referee's request. The mentioned caiclips are still on the market at these prices.
Page 2/page 3 (INTRO)Line 19 (2nd page)-Line 6 (3rd page): the interest of this work is to show in particular that these micro-sensors can be used on the field for precise measurements of the precise measurements of the concentration of the studied gases. This is the reason why
the different studies that have already used these sensors in the field have been presented. The paragraph has been rewritten in order to keep only the studies that allow to better understand the interest of this work.Page 3- line 8: the text has been modified in the revised version of our article
Experimental setup
Page 3-line 19: the text has been modified in the revised version of our article
Page 3-line 20: We have detailed the way in which the experiment was carried out as well as the means of control
of the various parameters (temp, humidity...).Page 3-line 44 - Page 4-line 8: the meaning of the sentences has been checked and the text has been rewritten in the revised version of our article
Page 4- Line 10 – Line 21: the paragraph has been moved to the end of the introduction in the revised version of our article
Table 1: The time range has been specified below the table as well as the number of tests and the number of values for each selected range.
The determined voltage measurement for a given concentration range is the average of all voltage measurements of this range.
The number in parenthesis corresponds to the temperature or humidity differences between the different experiments.
the comments below the table have been added.Page 4- lines 23-29: The periods group together all the experiments that have been performed in a given period of time.
The number of sensors tested depends on each period and is given in Table 1.The duration of each period has been chosen to realize a sufficient number of experiments to study the performance of each of the micro-sensors. The first two measurement periods were performed to compare the performance of sensors of different technologies. The next two periods of measurements were carried out to study the medium and long term drift of the measurements obtained with electrochemical sensors tested during the first two periods.Page 4-Lines 30-36: the text has been verified and modified n the revised version of our article
Results
Page 5-line 4-12:The experimental set-up and conditions remained unchanged between series 1 and 2. Similarly, the conditions are unchanged between series 3 and 4. There is therefore no reason why part of the differences observed should be due to the experimental setup. On the other hand, the drift observed between the measurements obtained between series (1-2) on the one hand and (3-4) on the other hand were expected since the temperature and humidity varied.
Page 5-Line 26-33: the temperature is about 301,5K for the series. Measurements with two alphasense sensors in parallel (Alpha 2 and Alpha 3) were made during the third test period. The same results were obtained for the two sensors, i.e. a very weak drift in the short term (over two weeks of measurements). A new figure has been added and presents the results for the Alpha 2 sensor.
Page 5-line 39: Series 9 was made after series 8 where the humidity was higher. There is no reason to think that series 9 gives aberrant results
since they group the measurements on 2 consecutive days and 3 different experiments.Page 5-line 49: the text has been modified to meet the referree request
Page 6-lines 1-5: The difference between the two series is about 1 to 1,5 %. The text has been modified in the revised version of our article
Page 6-Line 17: the curve is obtained by averaging for each measurement point all the measurements obtained for series 1 and 2 on the one hand (series 1-2 curve) and series 3 and 4 on the other hand (series 3-4 curve).
Page 6-Line 31:The manufacturer gives a technical specification where there is a curve of the voltage response as a function of the concentration. Each point of the curve on the figure is the average of the values obtained on the manufacturer's Technical specification.
Page 6-line 47: Above 200ug/m3 the average values obtained are less reliable than for lower values because the values obtained are more dispersed and the uncertainties on the measurements are more important.
Page 7-line 2: Between the first and third month of the study these sensors were tested with NO2 (measurements not reported in this article).
It is possible that this changed the baseline voltage response when the sensors were used again during the third and fourth month. New tests with new sensors will be necessary to verify this.Conclusions
Page 7-Line 29:: the European directive (2008/50/EC) defines an alert threshold of 240ug/m3, an information and recommendation threshold of 180ug/m3 on average per hour and an air objective of 120 ug/m3 average concentration over 8 hours. Calibrating every 40ug/m3 up to 320 ug/m3 is equivalent to having 8 levels of measurements every 40ug/m3 which is sufficient to meet the air quality criteria. Even in the case of a loss of sensitivity, with a calibration every 60ug/m3 that would be sufficient to obtain measurements corresponding to these three thresholds (120, 180 and 240 ug/m3). The air quality objective set by the WHO is 100ug/m3 for the maximum daily value over 8 hours and 60 ug/m3 for the average value over 8 hours. during the last 6 months when the concentration is the highest. Even with these new thresholds it will be possible to use these sensors.
Page 7-line 32: All figures have been treated as non-linear because some sensors give results requiring a non-linear fit.
For the alphasense a linear fit is usually sufficient. To illustrate this result I have added a table with a linear and nonlinear
linear and nonlinear treatment of the 4 alphasense sensors.Figures (5 to 18): the figures have been reorganized and only the most relevant ones have been kept in the revised version of our article.
The others will be put in the supplementar materielsA new revised manuscript with supplementar materiels will be submitted
Christophe Claveau
Citation: https://doi.org/10.5194/amt-2022-75-AC1
-
CC1: 'Reply on RC1', christophe claveau, 20 Aug 2022
-
RC2: 'Comment on amt-2022-75', Maria Dolores Andrés Hernández, 13 Sep 2022
The manuscript by Claveau et al. presents a series of measurements of the voltage response of electrochemical and metal oxide low cost portable sensors which can be of interest for the AMT community.
The analysis of data is however quite descriptive, erratic and contradictory (e.g. linear responses with quadratic regressions). In addition, it is difficult to identify any new aspects revealed by those data respect to the references given. As a consequence, I strongly suggest the rejection of this manuscript for publication in its current form.
Overall, the manuscript would clearly benefit from:
- a better description of the methodology which also includes more details about the principle of the sensors used,
- a deeper statistical analysis and quantitatively interpretation of the data sets and the effect of short-term calibrations of the sensors. The discussion of the repeatedly mentioned NO2 cross sensitivity is also missing,
- the improvement of the figure legends and figure captions to make the figures self-explanatory,
- the description and explanation of the new findings of the paper and potential recommendations, both in the conclusions and in the abstract.
- a revision of the usage of the English language, which makes difficult the understanding of some parts of the text.
I recommend the authors to thoroughly revise the current text and analysis and to submit a new in-depth modified manuscript.
Citation: https://doi.org/10.5194/amt-2022-75-RC2 -
AC2: 'Reply on RC2', christophe claveau, 11 Oct 2022
We thank again for the interest in our article.
Here are the answers we provide:
1- the experimental part has been modified in order to specify the methodology followed and the operating
principle of the sensors used2- the discussion of the results has been modified to take into account the editor's requests. The cross-sensitivity with NO2 was not investigated in this paper.The article only focuses on the performance of ozone sensors. This sensitivity with No2 is only mentioned only to give the limits of the use of these ozone sensors in the field under real conditions.
3- Legends below the tables, pictures and figures have been added to make the figures and tables more explicit
4- The conclusion has been modified to more explicitly highlight potential recommendations
5- The use of the English language has been revised to make the different parts of the text clearer
A revised manuscript will be submitted with supplementat material
Christophe Claveau
Citation: https://doi.org/10.5194/amt-2022-75-AC2
Status: closed
-
RC1: 'Comment on amt-2022-75', Anonymous Referee #1, 07 Aug 2022
General comments
The objective of this work is to evaluate the performance of commercial -electrochemical and semiconductor type- low-cost sensors (LCS) to quantify tropospheric ozone under laboratory conditions. The experiments were carried out at different experimental conditions, focused on humidity and temperature effects, to evaluate the LCS output signal dependence on these variables. Given the global and local consequences of air pollution, the subject under study is of prime importance. The evolution of the field of low-cost sensors is essential as they could provide information at finer scales than ever before. However, is in the consideration of this reviewer that the present manuscript needs to be reformulated. An important point that should be extensively reviewed is the structure and clarity of the manuscript, which makes it difficult to follow the content and main points. The writing style needs to be revised as well. The literature review is somewhat scarce and needs to be reconsidered as, in this reviewer's opinion, it is impacting the novelty of the manuscript. This is evident in the results, as much of what the authors find has already been discussed in the scientific literature, neglecting potential impacts.
In relation to the abstract, although it mentions the main objective, including a short summary of results and conclusions will benefit the reader. The description of the problems/issues but also the advantages of these sensor types provided in the introduction need to be refreshed. The references here are somewhat outdated, and this in a field as dynamic as that of LCS seems to be impacting the results and conclusions novelty. I also think that the text in general would benefit if the main points made by the authors are clearly stated. This will also make it easier for the reader to identify and follow some important points that this work could potentially offer. As the current manuscript is focused on evaluating the sensor response (voltage signal) to different lab conditions, correction methods and colocation studies review included in the text dilutes the discussion of topics more relevant to this study. The experimental design is partially described, impacting the transparency and reproducibility of the study. A detailed description would also help to clarify the results and conclusions of the study. Regarding the results and the analysis of the data, it is chaotic, which dilutes important information. There is also a lot of duplicate information that could be summarized or outlined in a more efficient way.
Specific comments
Having a continuous numeration of lines will be greatly appreciated.
Abstract
Page 1
Line 14: Perhaps “answering questions” fits better the sentence.
Line 15: “In this context, its adaptable and scalable nature will allow the insertion of new consumer measurement components available ("low cost" micro-sensors)”. Please rephrase, is not clear what the authors want to convey.
Line 18: Replace “Theses” with “These” and “are tested” by “were tested”
Line 19: Replace “and at ppb level” by “and ppb level”, “ambiant" with “ambient"
Introduction
Page 1
Line 31: this paragraph seems to be the continuation of the previous one, which, however, is separated from the first. Also, it is provided with a reference from 2012 -ten years ago- meaning the assertion included in the text represents the technology at that time. Specific studies can be found for each of the named technologies (metal oxide, electrochemical, infrared, photo-ionization sensors) in the last 5-10 years. I strongly suggest diving into more recent literature. Also, what does “target gas” mean in this context?
Line 39: Despite starting this paragraph with “Recently”, a similar issue is found as mentioned in the previous comment. Also, would be interesting to know if the mentioned sensors still in the market and if the price corresponds to 2022.
Line 45: Please rephrase/revise the entire paragraph content.
Page 2
Line 19 (2nd page)-Line 6 (3rd page): The first paragraph in these lines describes correction methods, which seem to be a bit outside the scope of the manuscript. Similarly, the next 5 paragraphs are mainly focused on in-field colocation studies, while this is a lab study. I suggest rethinking and rephrasing what of all this is contributing to sustaining your study case and/or helping the reader to understand your work.
Page 3
Line 8: It is said that “We have developed our own sensor”. I strongly suggest replacing sensor by system (or node). Also, it would be enriching if you could expand on the differences (here or elsewhere) between an OEM and a system.
Experimental setup
Line 19: replace “prototype sensor” with “prototype system”
Line 20: A detailed description of the exposure chamber will benefit any reader trying to replicate the experiments presented here. This point is a bit vague and needs to be reframed as it is an essential part of the experiments. For example, how the different variables (temperature, RH, flow, pressure, etc) are controlled? How long was the ramp held each time?
Line 44 (3rd page) - Lin 8 (4th page): Please check the intended meaning of these paragraphs and rephrase.
Page 4
Line 10 – Line 21: these paragraphs offer an introductory description of the sensors type, not relevant to the experiment conditions description. These paragraphs could be summarised and moved to the introduction.
Table 1: probably the time range in hours could be more informative. What is the “Number Experimental test”? And “Number values on range”? Is not clear what is trying to be said. For the temperature and humidity columns, what is the number in between parentheses?
Lines 23-29: Please rephrase this paragraph, I couldn’t understand what is trying to communicate. Also, could you expand on the reasons behind the different experiments and periods? How many sensors of each type were tested at the same time?
Lines 30-36: Please check the content of the paragraph and rephrase.
Results
Page 5
Line 4-12: The text states that the differences between the different series are a consequence of changes in the sensor response (drift). Is it possible that part of the differences found are due to the experimental set-up? One way to provide evidence in this regard would be to have duplicate sensors exposed to the same conditions.
Line 26-33: what is the “room temperature”? And the “higher temperature”? Again, having duplicate senors would help to support the manuscript statements, but it doesn’t seem to be the case.
Line 39: “The voltage response increases with humidity”, but series #9 seems to be an outlier. What would be the explanation for this? Also, the Figure number should be stated.
Line 49: could you expand based on your findings about the observed hysteresis effect on the sensors?
Page 6
Lines 1-5: the manuscript states a good agreement exists between series #5 (black line, Figure 10) and series #7 (blue line). Please revise this, as it doesn’t seem to be the case.
Line 17: how the series were combined to produce the figure 12?
Line 31: it is not clear how the manufacturer curve shown in fig 14 was obtained.
Line 47: “The mean values of the voltage response between the series are close up to 200 mg/m3 but for values above 200 μg/m3 the quality of the measurements does not allow to obtain a reliable result”. Check the meaning of the sentence. Also, why it doesn’t allow to obtain a reliable result? Expand.
Page 7
Line 2: “The figure shows a strong drift between the values obtained during the first month of use and the 3rd and 4th months of use”. Is the positive change in the baseline shown in Fig 18, due only to drift? Could there be other effects?
Conclusions
Line 29: “the measurements precision is enough for air quality studies with European regulation” Could you expand on this? Qualitative or quantitative studies? There will be soon a new European Directive (with lower limit values in order to follow the new WHO guidelines). Do you think the electrochemical sensors will be able to cope with them?
Line 32: The text mentions “the linearity properties…”, but in all the figures the response was treated as non-linear.
Figures
Regarding the figures (5 to 18), in my opinion, they are too many and important information is diluted. I suggest selecting (and rearranging) the most pertinent figures for the main text and the rest could go to the supplementary. Another relevant comment is that the R-square for non-linear models is statistically incorrect.
Citation: https://doi.org/10.5194/amt-2022-75-RC1 -
CC1: 'Reply on RC1', christophe claveau, 20 Aug 2022
Thanks to Referee 1 for agreeing to review our article. We will modify it taking into account as much as possible the comments of Referee 1.
Christophe Claveau
Citation: https://doi.org/10.5194/amt-2022-75-CC1 -
AC1: 'Reply on RC1', christophe claveau, 11 Oct 2022
We thank again for the interest in our article.
Here are the answers we provide:
Specific comments: numeration of lines has been modified in the revised version of our article
Abstract
Page 1-Line 14: the text has been modified in the revised version of our article
Page 1 -Line 15: We have reworded the paragraph to show that the adaptable nature of our sensor will allow us to test the latest and most
micro sensors and help to answer questions related to pollution exposure.Page 1 -Line 18: the text has been modified in the revised version of our article
Page 1-Line 19::the text has been modified in the revised version of our article
Introduction
Page 1-line 31-45: The most recent and detailed studies on the performance of MOS (metal oxide sensors)
and electrochemical sensors and concerning ozone are between 2012 and 2018. It seemed important to us
to present a summary of the studies during these last decades on the subject. All the paragraph
has been rewritten to meet the referee's request. The mentioned caiclips are still on the market at these prices.
Page 2/page 3 (INTRO)Line 19 (2nd page)-Line 6 (3rd page): the interest of this work is to show in particular that these micro-sensors can be used on the field for precise measurements of the precise measurements of the concentration of the studied gases. This is the reason why
the different studies that have already used these sensors in the field have been presented. The paragraph has been rewritten in order to keep only the studies that allow to better understand the interest of this work.Page 3- line 8: the text has been modified in the revised version of our article
Experimental setup
Page 3-line 19: the text has been modified in the revised version of our article
Page 3-line 20: We have detailed the way in which the experiment was carried out as well as the means of control
of the various parameters (temp, humidity...).Page 3-line 44 - Page 4-line 8: the meaning of the sentences has been checked and the text has been rewritten in the revised version of our article
Page 4- Line 10 – Line 21: the paragraph has been moved to the end of the introduction in the revised version of our article
Table 1: The time range has been specified below the table as well as the number of tests and the number of values for each selected range.
The determined voltage measurement for a given concentration range is the average of all voltage measurements of this range.
The number in parenthesis corresponds to the temperature or humidity differences between the different experiments.
the comments below the table have been added.Page 4- lines 23-29: The periods group together all the experiments that have been performed in a given period of time.
The number of sensors tested depends on each period and is given in Table 1.The duration of each period has been chosen to realize a sufficient number of experiments to study the performance of each of the micro-sensors. The first two measurement periods were performed to compare the performance of sensors of different technologies. The next two periods of measurements were carried out to study the medium and long term drift of the measurements obtained with electrochemical sensors tested during the first two periods.Page 4-Lines 30-36: the text has been verified and modified n the revised version of our article
Results
Page 5-line 4-12:The experimental set-up and conditions remained unchanged between series 1 and 2. Similarly, the conditions are unchanged between series 3 and 4. There is therefore no reason why part of the differences observed should be due to the experimental setup. On the other hand, the drift observed between the measurements obtained between series (1-2) on the one hand and (3-4) on the other hand were expected since the temperature and humidity varied.
Page 5-Line 26-33: the temperature is about 301,5K for the series. Measurements with two alphasense sensors in parallel (Alpha 2 and Alpha 3) were made during the third test period. The same results were obtained for the two sensors, i.e. a very weak drift in the short term (over two weeks of measurements). A new figure has been added and presents the results for the Alpha 2 sensor.
Page 5-line 39: Series 9 was made after series 8 where the humidity was higher. There is no reason to think that series 9 gives aberrant results
since they group the measurements on 2 consecutive days and 3 different experiments.Page 5-line 49: the text has been modified to meet the referree request
Page 6-lines 1-5: The difference between the two series is about 1 to 1,5 %. The text has been modified in the revised version of our article
Page 6-Line 17: the curve is obtained by averaging for each measurement point all the measurements obtained for series 1 and 2 on the one hand (series 1-2 curve) and series 3 and 4 on the other hand (series 3-4 curve).
Page 6-Line 31:The manufacturer gives a technical specification where there is a curve of the voltage response as a function of the concentration. Each point of the curve on the figure is the average of the values obtained on the manufacturer's Technical specification.
Page 6-line 47: Above 200ug/m3 the average values obtained are less reliable than for lower values because the values obtained are more dispersed and the uncertainties on the measurements are more important.
Page 7-line 2: Between the first and third month of the study these sensors were tested with NO2 (measurements not reported in this article).
It is possible that this changed the baseline voltage response when the sensors were used again during the third and fourth month. New tests with new sensors will be necessary to verify this.Conclusions
Page 7-Line 29:: the European directive (2008/50/EC) defines an alert threshold of 240ug/m3, an information and recommendation threshold of 180ug/m3 on average per hour and an air objective of 120 ug/m3 average concentration over 8 hours. Calibrating every 40ug/m3 up to 320 ug/m3 is equivalent to having 8 levels of measurements every 40ug/m3 which is sufficient to meet the air quality criteria. Even in the case of a loss of sensitivity, with a calibration every 60ug/m3 that would be sufficient to obtain measurements corresponding to these three thresholds (120, 180 and 240 ug/m3). The air quality objective set by the WHO is 100ug/m3 for the maximum daily value over 8 hours and 60 ug/m3 for the average value over 8 hours. during the last 6 months when the concentration is the highest. Even with these new thresholds it will be possible to use these sensors.
Page 7-line 32: All figures have been treated as non-linear because some sensors give results requiring a non-linear fit.
For the alphasense a linear fit is usually sufficient. To illustrate this result I have added a table with a linear and nonlinear
linear and nonlinear treatment of the 4 alphasense sensors.Figures (5 to 18): the figures have been reorganized and only the most relevant ones have been kept in the revised version of our article.
The others will be put in the supplementar materielsA new revised manuscript with supplementar materiels will be submitted
Christophe Claveau
Citation: https://doi.org/10.5194/amt-2022-75-AC1
-
CC1: 'Reply on RC1', christophe claveau, 20 Aug 2022
-
RC2: 'Comment on amt-2022-75', Maria Dolores Andrés Hernández, 13 Sep 2022
The manuscript by Claveau et al. presents a series of measurements of the voltage response of electrochemical and metal oxide low cost portable sensors which can be of interest for the AMT community.
The analysis of data is however quite descriptive, erratic and contradictory (e.g. linear responses with quadratic regressions). In addition, it is difficult to identify any new aspects revealed by those data respect to the references given. As a consequence, I strongly suggest the rejection of this manuscript for publication in its current form.
Overall, the manuscript would clearly benefit from:
- a better description of the methodology which also includes more details about the principle of the sensors used,
- a deeper statistical analysis and quantitatively interpretation of the data sets and the effect of short-term calibrations of the sensors. The discussion of the repeatedly mentioned NO2 cross sensitivity is also missing,
- the improvement of the figure legends and figure captions to make the figures self-explanatory,
- the description and explanation of the new findings of the paper and potential recommendations, both in the conclusions and in the abstract.
- a revision of the usage of the English language, which makes difficult the understanding of some parts of the text.
I recommend the authors to thoroughly revise the current text and analysis and to submit a new in-depth modified manuscript.
Citation: https://doi.org/10.5194/amt-2022-75-RC2 -
AC2: 'Reply on RC2', christophe claveau, 11 Oct 2022
We thank again for the interest in our article.
Here are the answers we provide:
1- the experimental part has been modified in order to specify the methodology followed and the operating
principle of the sensors used2- the discussion of the results has been modified to take into account the editor's requests. The cross-sensitivity with NO2 was not investigated in this paper.The article only focuses on the performance of ozone sensors. This sensitivity with No2 is only mentioned only to give the limits of the use of these ozone sensors in the field under real conditions.
3- Legends below the tables, pictures and figures have been added to make the figures and tables more explicit
4- The conclusion has been modified to more explicitly highlight potential recommendations
5- The use of the English language has been revised to make the different parts of the text clearer
A revised manuscript will be submitted with supplementat material
Christophe Claveau
Citation: https://doi.org/10.5194/amt-2022-75-AC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 656 | 304 | 43 | 1,003 | 35 | 40 |
- HTML: 656
- PDF: 304
- XML: 43
- Total: 1,003
- BibTeX: 35
- EndNote: 40
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
