Water Treatment Plant Prototype with pH Control Modeled on Fuzzy Logic for Removing Arsenic Using Fe(VI) and Fe(III)

Round 1

Reviewer 1 Report

The authors used fuzzy logic model for the operation of a pilot plant for arsenic removal from synthetic groundwater using Fe(VI) and pH control. The objectives are interesting, however the manuscript presentation must be improved substantially.   General comments: 1. The readability must be improved, because most readers are not experts in fuzzy logic models. Description and figures/tables must be improved so that many environmental engineers can understand. 2. It is not clear why fuzzy logic model was used instead of classic algorithms for this treatment. What is advantage of fuzzy logic model, and what is the challenge for more classical algorithms?  3. The overall structure of introduction is not straightforward and not easy to comprehend. There are nine paragraphs and the focus goes forward and backward in various ways (e.g. fuzzy logic is mentioned twice in the 1st and 8th paragraphs, treatment by Fe(VI) is mentioned in four paragraphs, 3rd to 6th). Also, water quality problems regarding arsenic in groundwater is not mentioned in introduction, which is confusing. More simple, straightforward structure (e.g up to four paragraphs) is desired for readability. 4. Method section should provide more basic knowledge. e.g. basic chemistry regarding Fe(VI) and arsenic species, the water quality of feed water (pH, ORP, As, Fe)   Specific comments: L31: what PID means, and what "more traditional manners than classical algorithms" means? It reads like fuzzy logic algorithms are more traditional than classical algorithms, which sounds strange a bit to me. L34: what is IoT2040 gateway platforms? Most readers are not familar with those terms, so better explain more. L46: it looks Fe(VI) appears out of nowhere, because the previous discussions are about processors and application for pilot plant operations. The paragraphs need to be reorganized.  L48-58: these are about application of Fe(VI) to wastewater treatment ,which is a bit far from the focus of this study, which is treatment of groundwater for drinking purpose. More relevant introduction of Fe(VI) could be provided. L59-63 is about removing arsenic by Fe(VI), so this part could be more elaborated. L70-71:  "influent characteristics in a continuous system can change quickly" does this hold for groundwater and arsenic? I believe groundwater quality is quite stable over time in general. L92: what is GJ? L92: in introduction, synthetic groundwater was used for raw water, but here, wastewater was the feed water?  L101: how much is the original ORP before adjustment?  L122: explain what is SBGW at first use. Table 1 must include pH, ORP and arsenic.  L187: adding Fe(VI) indrease pH, although Fe(VI) is dissolved in NaOH? Why? More detailed explanation of Fe(VI) solution must be given in Methods section. Figure 1: where the samples for water quality measurement was taken? Why are there many references which looks strange to me, for example, L150 and L171? What are the citations for? Referring methods? results? These citations make me wonder, which is original contribution of this paper and which is from literature. Fig 6 what are input and output data? pH? Figure 8; the vertical axis must have broader range (e.g. from pH 4) so that overall outlet pH change can be seen.

Author Response

1. The readability must be improved, because most readers are not experts in fuzzy logic models. Description and figures/tables must be improved so that many environmental engineers can understand.

 

Thank you for the comment. Explanation of fuzzy logic algorithms have been simplified and now includes less text (lines 73-82), we hope the text will be more understandable from a practical point of view .

 

[Control strategies based on fuzzy logic are not based on mathematical modeling depicting chemical or physical relationships between the components of the system, instead they deal with a series of rules related to the changes observed in the process when a variable also changes, and therefore are more easily implemented. For example, to adjust the pH, a pH value is obtained using the probe and compared with the set point to calculate the error, then the controller using a set of predeterminate rules will send a signal to the pump to increase or decrease the dosing of chemical. By monitoring the rate of change for a given output the controller can use other rules to modify the response accordingly]

 

 

2. It is not clear why fuzzy logic model was used instead of classic algorithms for this treatment. What is advantage of fuzzy logic model, and what is the challenge for more classical algorithms? 

 

Thank you for the comment, an explanation was added in lines 73-75.

 

[Traditional algorithms operate easily on simple linear processes [13], but fuzzy logic ones can operate in complex or non-linear processes such as pH control [14] where non-linearities arise together with time-variable parameters or dead times [15].]

 

3. The overall structure of introduction is not straightforward and not easy to comprehend. There are nine paragraphs and the focus goes forward and backward in various ways (e.g. fuzzy logic is mentioned twice in the 1st and 8th paragraphs, treatment by Fe(VI) is mentioned in four paragraphs, 3rd to 6th). Also, water quality problems regarding arsenic in groundwater is not mentioned in introduction, which is confusing. More simple, straightforward structure (e.g up to four paragraphs) is desired for readability.

 

We appreciate the feedback. The introduction was rewritten almost entirely and now covers the following sections: arsenic, ferrate(VI), water treatment with ferrate(VI) and pH control.

 

4. Method section should provide more basic knowledge. e.g. basic chemistry regarding Fe(VI) and arsenic species, the water quality of feed water (pH, ORP, As, Fe)

 

Comment acknowledged, arsenic species and iron concentration was added to the table #1, pH and ORP was added as a text before the table. Synthetic ground water preparation is detailed in methods section and Ferrate(VI) preparation was polished to emphasize the use of the alkali that will rise the pH during the treatment.

 

5. Specific comments: L31: what PID means, and what "more traditional manners than classical algorithms" means? It reads like fuzzy logic algorithms are more traditional than classical algorithms, which sounds strange a bit to me.

 

We thank you your comment. Text was rephrased to clarify that traditional algorithms work well in linear processes while those based on fuzzy logic work better in non-linear processes like pH adjustment.  Reference to PID control  (proportional, integral, and derivative control) was deleted for clarity.

Lines 78-84.

 

 

6. L34: what is IoT2040 gateway platforms? Most readers are not familar with those terms, so better explain more.

 

We appreciate the comment. A more general explanation was added (lines 84-89).

 

6. L46: it looks Fe(VI) appears out of nowhere, because the previous discussions are about processors and application for pilot plant operations. The paragraphs need to be reorganized. 

 

Thank you for the suggestion.  All the paragraph was modified to improve clarity.

7. L48-58: these are about application of Fe(VI) to wastewater treatment ,which is a bit far from the focus of this study, which is treatment of groundwater for drinking purpose. More relevant introduction of Fe(VI) could be provided.

Suggestion acknowledged.  References regarding wastewater were changed for the ones related to drinking water treatment (lines 52-58). A more detailed section about ferrate(VI) uses was added (lines 43-49 and 59-67)

8. L59-63 is about removing arsenic by Fe(VI), so this part could be more elaborated.

Comment appreciated, arsenic removal by ferrate can be found in lines (35-42).

 

9. L70-71:  "influent characteristics in a continuous system can change quickly" does this hold for groundwater and arsenic? I believe groundwater quality is quite stable over time in general.

It is correct, comment appreciated. Line was deleted.

10. L92: what is GJ?

 

Comment appreciated, GJ is the model of the pump head.

11. L92: in introduction, synthetic groundwater was used for raw water, but here, wastewater was the feed water? 
Correction appreciated, corrected to “raw water”.

12. L101: how much is the original ORP before adjustment?

Thanks for the indication, initial ORP (100 mV) of raw water added to line 127. 

 

13. L122: explain what is SBGW at first use. Table 1 must include pH, ORP and arsenic.

 

Thank you for the correction,  table #1 now  has the arsenic initial concentrations;  pH (line 128) and ORP (line 127) was added to lines to keep the caption in the table (mg/L).

 

 

14. L187: adding Fe(VI) indrease pH, although Fe(VI) is dissolved in NaOH? Why? More detailed explanation of Fe(VI) solution must be given in Methods section.

Commentary appreciated.

 

Ferrate(VI) solution is produced electrochemically in a divided reactor where the anolyte and catolyte is NaOH 20 mol/L solution (lines 107-108). Ferrate(VI) is produced in the anolyte solution and keeps dissolved. Adding the anolyte to water -even in small volumes- increases its pH because the ferrate solution contains an appreciable amount of NaOH.

 

 

15.Figure 1: where the samples for water quality measurement was taken?

 

Indication appreciated, figure 1 have “Sampling Point” caption added.

 

16. Why are there many references which looks strange to me, for example, L150 and L171? What are the citations for? Referring methods? results? These citations make me wonder, which is original contribution of this paper and which is from literature. Fig 6 what are input and output data? pH?

 

Comment appreciated. The section corresponding to the development of the fuzzy logic algorithm was moved to the supplementary information to avoid interruption of the reading flow. Explanation of figure 6 (now figure 10 of supplementary information) was rewritten and variables explained with more clarity.

 

 

17. Figure 8; the vertical axis must have broader range (e.g. from pH 4) so that overall outlet pH change can be seen.

 

Thank you for the suggestion. Vertical axis scaling updated, now it spans pH from 3 to7

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript „Water treatment plant prototype with pH control modeled on fuzzy logic for removing Arsenic using Fe(VI) and Fe(III)“ by Paredes Larroca et. al is an engineering study on the remediation of synthetic wastewater. This is an interesting work which could make a good contribution in Water. However, there are several points to be clarified, thus making the present manuscript not yet publishable.

Hereafter is a list of comments which need to be addressed (without meticulous answers, I would have to maintain my actual recommendation):

• Lines 48-50: to what are corresponding the percentages?
• Line 56 and section 3.1: I do not understand why you are controlling the pH. Indeed, Fe(VI) is stable at alkaline pH and a decrease to circumneutral medium leads to the decomposition of Fe(VI) into Fe(III). Therefore, if the addition of ferrate increases the solution pH, I would not change this parameter in order to have an efficient degradation of the target pollutants by Fe(VI). Therefore, the change of pH is necessary to be done at the end of the treatement. Please, discuss and clarify.
• The previous comment opens another question about the economic viability of such a treatment. Indeed, this is an engineering work and the change of pH requires chemicals. Please, discuss and clarify.
• Line 85: I do not understand why Fe(III) is added. Indeed, Fe(VI) is decomposed into Fe(III) during reaction. Moreover, to acidify pH, other chemical could be used. Furthermore, the addition of Fe(III) accelerate the decomposition of Fe(VI). Please, discuss and clarify.
• Introduction: I advise you to introduce more the ferrate. It could be used as an oxidants but also as coagulant since it decomposes into Fe(III). Many literatures reported this.
• Line 116: what is the purity of the synthesized Fe(VI)?
• Line 121: how was perform the determination of Fe(VI) concentration by UV-VIS spectrophotometry? Was the method using ABTS used?
• Line 187: what was the initial concentration of Fe(VI)?
• Section 3.3 and Fig. 11: You describe the figure and the discussion with initial dosage starting from 10min… To me, initial dosage is at t=0min. Please, clarify.  

Author Response

1. Lines 48-50: to what are corresponding the percentages?

Thank you for the question. For clarity, introduction was rewritten completely and now references emphasize drinking water treatment.  Reference regarding percent removal of variables related to waste water treatment was deleted.

 

2. Line 56 and section 3.1: I do not understand why you are controlling the pH.

Your comment is acknowledged. Ferrate(VI) was electrochemically synthesized in a concentrate (20 mol/L) NaOH solution (lines 107-109), therefore ferrate(VI) is dissolved in a concentrated alkaline matrix. Adding ferrate(VI) solution to treat the raw water adds concentrated NaOH that elevates the pH, that makes mandatory a later pH correction because the study aims to remove arsenic from underground drinking water.

 

3. Indeed, Fe(VI) is stable at alkaline pH and a decrease to circumneutral medium leads to the decomposition of Fe(VI) into Fe(III). Therefore, if the addition of ferrate increases the solution pH, I would not change this parameter in order to have an efficient degradation of the target pollutants by Fe(VI). Therefore, the change of pH is necessary to be done at the end of the treatement. Please, discuss and clarify.

Thank you for the comment, it is appreciated.  It makes us realize that we need to rephrase it completely. 

 

It is correct to say that ferrate is more stable at higher pH. The addition of ferrate is done in the first part of the treatment: ferrate(VI) solution is pumped continuously into the flow of raw water and flows through the pipe with a static mixer (captions added to figure 1), there arsenite is oxidated and some arsenate will react with the Fe(III) ions product of Fe(VI) reduction.

 

As the addition of ferrate, in a 20 mol/L NaOH solution causes an important increase in pH (line 61-67), this pH needs to be lowered with an acid.  As the amount of iron added per liter of solution is very low (about 70 micrograms per liter related to ferrate(VI) addition) then, to remove the arsenate, more iron must be added to the water.

 

Ferric chloride solution lowers the pH while adding iron, then pH can be adjusted to setpoint value while removing the remaining arsenic with the extra iron added, all in one-step.

 

 

 

4. The previous comment opens another question about the economic viability of such a treatment. Indeed, this is an engineering work and the change of pH requires chemicals. Please, discuss and clarify.

Comment is appreciated. The section 3.4 was added to estimate the cost to treat 1 cubic meter of raw water, which turned to be very low (US$ 0.022 per cubic meter, lines 245-246)

 

5. Line 85: I do not understand why Fe(III) is added.

Thank you for your comment. Please refer to answer on comment #3.

 

6. Indeed, Fe(VI) is decomposed into Fe(III) during reaction. Moreover, to acidify pH, other chemical could be used. Furthermore, the addition of Fe(III) accelerate the decomposition of Fe(VI). Please, discuss and clarify.

 

Clarification appreciated.  Fe(III) is added to the flow when Ferrate(VI) has already oxidized arsenic (figure 1) therefore Fe(III) added has no noticeable effect. Certainly, another acidifying agent could be used, however the low amount of iron added in the arsenic pre-oxidation step will not suffice for a complete arsenate removal, therefore the chemical of choice is iron(III) chloride.

 

7. Introduction: I advise you to introduce more the ferrate. It could be used as an oxidants but also as coagulant since it decomposes into Fe(III). Many literatures reported this.

 

Suggestion is appreciated.  A more thoroughly description about ferrate(VI) was added to the introduction (lines 43-69).

 

8. Line 116: what is the purity of the synthesized Fe(VI)?
   Thank you for the question, ferrate purity was not determined because it was produced in solution by the anodic dissolution of the iron electrode instead of being produced by dry or wet synthesis.

 

9. Line 121: how was perform the determination of Fe(VI) concentration by UV-VIS spectrophotometry? Was the method using ABTS used?
   Method was missing, correction acknowledged. Spectrophotometric method description added, ABTS was not used.

 

10. Line 187: what was the initial concentration of Fe(VI)?
    Thanks for the question, concentration was missing. Ferrate(VI) concentration (0.26 mol/L) added (line 119).

 

11. Section 3.3 and Fig. 11: You describe the figure and the discussion with initial dosage starting from 10min… To me, initial dosage is at t=0min. Please, clarify.You are right, treatment began on time zero. Thank you.  Due to the major revision fig.11 of the original submission is figure 5. X-axis initial value changed to zero.

Round 2

Reviewer 1 Report

The authors addressed all of my previous comments. The manuscript is now recommended for publication.

Author Response

​​​​​​​We woud like to thank you for your time and attention to details while reviewing our manuscript. Your revision is very much appreciated.

Reviewer 2 Report

The revised manuscript „Water treatment plant prototype with pH control modeled on fuzzy logic for removing Arsenic using Fe(VI) and Fe(III)“ by Paredes Larroca et. al has not the standard to be published in Waters. Although the quality has been slightly improved, authors did not fulfill my expectations through the answers of my comments. The problematic of adding Fe(III) is not thoroughly discussed. Indeed, it is used there as pH controller but Fe(III) can remove arsenic. In addition, Fe(III) is produced by the decomposition of ferrate. Moreover, we do not know what is the purity of ferrate, so the observed results cannot be discussed.  In addition, the revised version is unreadable since the changes were performed in tracking mode! Furthermore, the manuscript was completely modified, thus requiring a new submission.

Therefore, I have to recommend to REJECT the manuscript.

Author Response

Response to Reviewer 2 Comments

 

We are grateful for the time and effort invested in the thoroughly revision of the manuscript.  We would also like to apologize for the inconveniences caused by the uploading of the MS-Word version with the tracking mode enabled.

Author Response File: Author Response.docx