Submit to this Journal Review for this Journal Propose a Special Issue

Article Menu

Share Help Cite Discuss in SciProfiles

Open AccessArticle

Peer-Review Record

Experimental and Numerical Evaluation of Toxic Pool Evaporation

Appl. Sci. 2020, 10(23), 8448; https://doi.org/10.3390/app10238448

by Benjamin Truchot^1,*, André Carrau², Véronique Debuy¹

, Thibauld Penelon¹

and Jean-Pierre Bertrand¹

Reviewer 1: Anonymous

Reviewer 2: Anonymous

Appl. Sci. 2020, 10(23), 8448; https://doi.org/10.3390/app10238448

Submission received: 30 September 2020 / Revised: 20 November 2020 / Accepted: 25 November 2020 / Published: 26 November 2020

(This article belongs to the Section Applied Physics General)

Round 1

Reviewer 1 Report

Introduction - I think you should mention the importance of the concentration gradient as this is an important factor in accurately predicting the evaporation rate.

1.1 has allowed to develop = has allowed several evaporation models to be developed.

final sentence of 1.1: influence of the scale factor - influence of pool size?

last sentence page 3:small and medium scale tests were designed to be as ..

Figure 3 labels right left - replace with upstream downstream

sentence below Fig 4 - not clear if 4% turbulence applies to both experiments.

last sentence of page 4 - dimensions of 'specific system'?

section 2.2 suggest move sentence 'The FTIR sampling probe was placed ...' up to follow sentence ending '... flow rate in the ventilation section.'

Page 6, second sentence' ..6 tests were carried out at small scale and 5 at medium scale.'

Page 6 penultimate paragraph 'represented 29%' - 'manufacturer's 29%'?

Fig 6 caption: Evaporation rate and pool temperature ...

Final sentence of 3.1 'close starting point' is clumsy and needs re-wording.

3.2 I think the time is ~45mins rather than 1 hour.

3.3 This paragraph is confusing and difficult to understand - please re-word.

3.4 second paragraph, 3rd sentence suggest : This was confirmed by experimental measurements at lower concentrations in tests 1 and 2, compare Fig 8 below with Fig 6.

Final paragraph page 8 - I think this is a separate, but important point, i.e. experimental repeatability, which would be clearer under its own sub-section. I think the current text should be broken into shorter, clearer sentences: Finally, it is important to verify the repeatability of the experiments. This was done by comparing data from tests 1 and 2. Figure 8 shows they had the same ...

Caption fig 8 - does not reflect what it is conveying. consistency of labels.

First sentence below fig 8: '..ambient velocity compared to the reference test.

Fig 9 - I think you either need to comment on why you are not showing results for test 5 or show them.

First sentence below Fig 9: structure needs to be improved.

4.1 first sentence: concentration gradient not mentioned before- see later comments.

p10 para below fig 10: flooding's volume = filling of the pool?

p10 final para, suggest move final sentence to after fig 11.

4.3, first para: '.. was measured.' suggest add 'by conducting test 4'

4.4 First sentence , suggest remove and add: Test 2 was conducted to assess the effect of in increasing wind velocity from 1.0 to ..

4.4 3rd sentence, 'above described approach' = proposed model (section 5)

Fig 14 caption - there are only 2!

p12 final sentence: this model (EVAP-Tox)

fig 15 meaning of T is not clear pool/air? there is an arrow missing

p13 para 4 'concentration equation (eq 3)

Fig 16 improve quality?

Para below fig 16 - I think this should largely be in section 2.2, and some mention made in section 3 that the data was analysed to account for the discrepancy in predicted and observed evap rate?

conclusions - 'too conservative' - suggest 'overly', final sentence: 'process in' - process into?

Please note there are many small points of English and typos which need addressing.

Author Response

Dear reviewer,

you will find attached the answer to your valuable comment.

regards,

B. TRUCHOT

Author Response File: Author Response.docx

Reviewer 2 Report

The article is interesting from the point of view of chemical safety and environmental protection. It is written in a clear manner, although the authors did not avoid a few mistakes. It also has some confusion. References selected correctly. However, it needs to be corrected before publication. Detailed Notes:
There is numerical evaluation in the title of the article, but the adopted numerical model is very briefly described. This needs to be completed.
Figure 1 is hardly legible and, above all, badly described. It shows that the highest vapor pressure is obtained at the lowest concentrations (30%) and the lowest temperature (270 K) - this is impossible.
Tables 1 and 2 show ammonia concentrations in the pool - actual and given by provider. What are the provider concentrations given for? By what method was the concentration in the pool determined?
The authors did not provide what is represented by the thick gray line in Figures 6, 8 and further?
In equations 3 and 4 it is necessary to specify how the parameter ωk - the production or consumption term is defined

Author Response

Dear reviewer,

you will find enclosed our answers to your valuable comments that help us to improve the submitted paper.

regards,

B. TRUCHOT

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The authors show that the evaporation rate of ammonia from the solution decreases rapidly from its initial value. They state that this decrease is much larger than that suggested by the reduction in vapor pressure that accompanies the change in surface temperature. The authors ascribe the unexplained reduction to the resistance in the liquid phase to mass transfer of ammonia to the surface. A model , Equation 3, based on this idea is is used to describe the data. The main problem with this paper is that there is not enough detail in the paper to evaluate the assumptions made in deriving the model. Some specific questions and comments are:

1) How was the ammonia vapor pressure computed as a function of temperature and solution concentration?

2) Why did the authors choose to focus on the formulation by Mackay and Matsugu(1973)? There are several other competing formulations. See Mazzarotta B., Bubbico R., 2016, Predicting evaporation rates from pools, Chemical Engineering Transactions, 48,49-54 DOI:10.3303/CET1648009

2) How was transport in the liquid solution modeled? Was molecular diffusion the only mechanism?

3) What were the turbulent velocities and friction velocity? This information is important in evaluating the claim that turbulence did not matter. I would think that surface shear would lead to vertical mixing in a thin layer of fluid.

4) The time variation of evaporation is a well known phenomenon. The authors need to conduct a more extensive review of the literature to support their claim for originality. The reference list only contains seven papers. four of which do not appear in peer reviewed journals. Sutton(1934) and Mackay and Matsugu(1973) are clearly very dated.

5) Why is Froude number scaling relevant? There are no gravity controlled stability effects that affect evaporation in the experiments.

Author Response

The authors show that the evaporation rate of ammonia from the solution decreases rapidly from its initial value. They state that this decrease is much larger than that suggested by the reduction in vapor pressure that accompanies the change in surface temperature. The authors ascribe the unexplained reduction to the resistance in the liquid phase to mass transfer of ammonia to the surface. A model , Equation 3, based on this idea is is used to describe the data. The main problem with this paper is that there is not enough detail in the paper to evaluate the assumptions made in deriving the model.

The organization of the paper was modified. The part concerning the model was extracted to a dedicated chapter (number 5). This new chapter gives additional details regarding the model.

Some specific questions and comments are:

How was the ammonia vapor pressure computed as a function of temperature and solution concentration?

In the model, both concentration and temperature were considered to evaluate ammonia vapor pressure. A sentence was added for clarification in new chapter 5 : Then, the ammonia vapor pressure is computed considering both pool surface concentration and pool surface temperature based on the relation described in section 1.1..

Why did the authors choose to focus on the formulation by Mackay and Matsugu(1973)? There are several other competing formulations. See Mazzarotta B., Bubbico R., 2016, Predicting evaporation rates from pools, Chemical Engineering Transactions, 48,49-54 DOI:10.3303/CET1648009

We focused on the Mackay and Matsugu correlation since this is the most widely used correlation in the field of risk assessment and land use planning, more especially in France but not only having in mind that this correlation was introduced in the Phast software. We are aware that numerous of other formulations exist with, sometimes a very specific field of use (a given product or specific flow conditions, …). Some clarifications were added on the reasons of this choice in section 1.1 and many references were added.

How was transport in the liquid solution modeled? Was molecular diffusion the only mechanism?

A specific paragraph was added to clarify the model. Molecular diffusion is the only mechanism considered but a constant should be added on the diffusion velocity to take account of the convection that could occur in an ammonia pool. Details of the model are provided in the new paragraph 5.1 and 5.2.

What were the turbulent velocities and friction velocity? This information is important in evaluating the claim that turbulence did not matter. I would think that surface shear would lead to vertical mixing in a thin layer of fluid.

Turbulent velocity was estimated using mac-caffrey probes associated with high frequency transducer. The turbulent velocity was found to be about 4% of the average velocity.

The time variation of evaporation is a well-known phenomenon. The authors need to conduct a more extensive review of the literature to support their claim for originality. The reference list only contains seven papers. four of which do not appear in peer reviewed journals. Sutton(1934) and Mackay and Matsugu(1973) are clearly very dated.

Many references were added in the reference. We found evolution along time mentioned on some paper for hydrocarbon mixtures (Okamoto and Fingas typically) but only the paper from Salin regarding toxic mixtures.

Why is Froude number scaling relevant? There are no gravity-controlled stability effects that affect evaporation in the experiments.

We though that the stratification of the vapor above the pool are mainly govern by the density gradient. This is why we have chosen the Froude scaling. This is not relative to the pool behavior but to the flow above the pool one.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper presents a new set of experiments to measure evaporation from a pool of ammonia. The main focus is on demonstrating how evaporation rates change with time and show that traditional numeric approaches to deriving the evaporation rate are only valid at the start. Exactly how “the start” is defined in this experimental work is unclear however. The paper also introduces a new numerical method to better capture the time-evolution of evaporation, but the full description of the model is lacking.

Such experiments are very challenging and time-consuming to prepare and conduct and this is valuable data, but the authors overstep what can realistically be concluded from their results when each variable has only been tested in one experiment. Major modification is needed to all aspects of the paper before it can be considered for acceptance. This includes improvements to the English throughout and the majority of the figures also need improving. I provide further details below.

Specific comments:

The abstract mentions a model called EVAP-Tox, but this model is not presented anywhere in the paper and is only mentioned in the last paragraph of the conclusions. If this paper is about this new model, then a section introducing the model is needed. Including it as part of the results, as currently in section 3.2, is not appropriate as it is not a result of the small-scale experiments. It needs to go in a new methods or discussion section. This needs to include: How is the model coded? What variables does it contain? Is it available? Is it only applicable to ammonia? Ideally the code would be made available along with the paper. One option would be to introduce the model in the methods, then present all of the experimental results, and then compare the experiments to the numerical model in a separate results section.

Throughout the text there are many examples of numbers where the decimal point is given by a comma. Please convert all of these to a full stop “.” and be consistent throughout.

The experimental set-up section is lacking a lot of details about the set-up, for example the dimensions of the gallery and the obstacles and how theses affect the flow field, and other information that you would expect for a wind-tunnel experiment. More information is needed here about the testing platform design prior to the sections on metrology and experimental design. There are many experimental papers that could potential be cited to justify the choices.

The evaporation equation is based on the pool radius, but the experiments utilise a square pool. The authors need to explain this choice and what impact is has on the applicability of the equation, even if it is considered to be negligible. The number used for r in each case needs to be given.

In the metrology section, more details are needed on (i) what the “dedicated mixing system” is, (ii) how the air flow in the section is measured, (iii) what the temporal frequency of all the measurements is.

In the experimental design, details are needed as to how the start of the experiment was initiated and hence the *initial* evaporation rate determined. As this paper is about proving the rate changes with time and therefore is different from the theory, understanding how any prior evaporation has been prevented in these experiments is critical here.

The text in section 2.3 suggests that equivalent experiments have been conducted at the small-scale and medium-scale, but this is not obvious from the tables. It would be helpful if tables 1 and 2 presented the same information in the same order of columns for a start. Also for the purposes of the paper, it would be helpful to order the medium-scale tests so they are numbered the same as the small-scale tests they correspond to. It is currently not at all clear which small-scale tests 1, 2 and 5 in Table 2 correspond to. Finally, and most critically, these tests do not appear to have been carried out under the same conditions with regard to concentration or temperature. The authors need to provide a lot more clarity as to the impact these differences will have. In particular, the difference in pool temperatures (and the differences between air temp and pool temp) is a significant difference between the two experiment sets, which needs explaining.

An uncertainty interval is given in Figure 4, but no information is given as to how this has been determined. Is it based on known measurement uncertainty? This information needs to be provided in the text.

Parts of the results would be better in the discussion, including the whole of section 3.2, and some sentences in the results include critical methodology details that should be in section 2. The structure and balance of the sections needs work. Introducing a Discussion section may help resolve this.

I don’t believe that the results are strong enough to back-up some of the claims in the paper. For example, in lines 235-237 I don’t agree that the plots are clear enough to show this. They demonstrate that the model is able to reproduce the trend (i.e. sharp decrease and then much slower reduction with time), but the modelled values start much higher and are then only within a factor of 2 later on in Figure 6 (right) which is not a “precise” representation. The large difference at the start is one of the reasons I’ve asked for the details to be added on how the experimental start is defined. The model is clearly able to better predict the trend than the Mackay & Matsuga correlation, but it is not perfect and the paper would benefit from discussion as to the potential source of differences between the model and the observations.

Similarly the differences in the first 1200 seconds between the small-scale and large-scale experiments presented in Figure 7 needs explaining.

The authors make claims that their findings on the role of atmospheric turbulence can be extrapolated to other atmospheric conditions (stable, neutral, unstable). I do not think these are justified. Firstly there is no proof given that the turbulent intensity actually did vary in the single experiment conducted and secondly without testing any of these environments (or even just more than one different turbulent intensity), there is no evidence that such a result would hold. I do not believe that the paper should be published with the turbulence section and request that section 4.3.1, lines 304-308 and line 25-26 in the abstract are removed.

Some of the other claims made about the influence of other parameters are also not supported by the experiments because only one alternative value has been tested. This is not a comprehensive sensitivity test. I’d advise that the discussion and conclusion is altered to be more balanced as a consequence.

In section 4.3.2 on the influence of wind velocity, the authors need to specify the scale equivalent velocities that have been tested to help the reader, as it is not possible to judge the range of actual velocities that these four results are comparable to. (Also the caption says 3 velocities, but it appears to only be 2?) When scaled up it is likely that the range of real-world velocities tested is actually quite small, so it is not necessarily surprising that little variation is seen. This result cannot and must not be extrapolated to all velocities however.

There are very few references for this paper. I was surprised to not see Mozer et al (2014) and Corruchaga and Casal (2015) cited, as these both present similar experimental results. I recommend the authors do a more thorough scientific literature search to ensure their work is put properly in context.

Corrections needed:

Line 10: use of “toxic distances” and “long distances” is unclear. This sentence needs improving.

Line 14: specify what the “velocity profile” is of

Line 19: remove “especially”

Line 25-26: remove sentence (see comments above)

Line 32: Sentence is unclear

Line 35: It would be helpful to explain why ammonia is important and relevant for this study, i.e. provide some more context

Line 36: remove “of”

Line 42: Remove “it should be first reminded that”. This type of language is not necessary. There are similar instances throughout the paper that should all be deleted, including line 59/60, 64, …

Line 47: please specify whether the “Given temperature” is for the liquid or the air

Line 49: spelling of weight is incorrect

Line 53: what does the star * in the equation mean?

Figure 1: The axes font size needs to be increased to make it more legible, also the second horizontal axis title cannot be read as it is overlaid by the plot

Figure 1 caption: Please specify whether this is for ammonia or any liquid

Line 64: the authors say “all these approaches”, but this is not clear. Please be more specific and provide the citations.

Line 66: “a gradient” of what?

Line 68-69: Explain what the two scales are and what you mean by scale factor. Without more detail this is not clear at this point in the text.

Line 82: Figure 3 is referred to before Figure 2, so these two figures need to be swapped around.

Figure 3: it would be preferable for these images to all be aligned.

Line 86: “sale” should be scale

Line 88-90: The units here are confusing. I think it should be 0.316m x 0.316m and 1m x 1m. Also suggest changing 3cm to 0.03m so that consistent units are used.

Line 91-92: replace “process consists in” with “is”

Line 94: “an atmospheric profile”

Line 94-95: change to “for the medium scale test are shown in Figure 2”

Line 97: Please provide details as to why this velocity profile has been chosen. Is 3.2m/s representative of something specific? The context of what this relates to in the real world is needed.

Figure 2: The legend and axes labels need to be in English

Line 109: should “technics” be “techniques”?

Line 110-111: “to evaluate the atmospheric concentration downstream of the pool”

Line 114: replace “managed” with “made”

Line 126: pool “depth” would be more consistent here

Line 144-145: I don’t think that 11 tests is too many to cover in this paper.

Figure 4: (i) It is very difficult to see the temperature curve, so I suggest altering the right axis to move the temperature curve up so it doesn’t overlap the other results, (ii) it would be more useful to have the x axis in minutes or hours rather than seconds, (iii) the figure captions needs to explain what the grey shading is and/or include this in the legend

Line 163: I’d argue that this isn’t “obvious” to an inexperienced reader, so the authors need to explain why this is important

Line 169: Figure 3 should be Figure 4

Line 176: If three temperatures were taken, then what is the purple line in Figure 4 showing? Is it one of these or is it the mean? This needs to be explained.

Lines 177-182: This paragraph is unclear. What is the theory (and where does this come from) versus what has been measured in these experiments? How has the “estimated pool concentration” been estimated?

Line 178: Replace NH3 with ammonia

Line 202: yes, it “could” be, but is this what you have done? What is the time-step used?

Line205-207: This detail needs to be explained and described in the methods section rather than here

Figure 5: Specify whether the left-hand plot is modelled or measured data. I would suggest that it would be better to show both the modelled and the measured data in this left plot to enable comparison

Lines 213-220: The first half of this paragraph belongs in the discussion (213-217), whereas the second half should be in the methods (217-220).

Line 216: I’d suggest that “safe” is not the right word here. “cautious” or “conservative” would be better.

Lines 227-232: This is poorly worded and lacking details about what parameters were varied. Where are the results for test 2?

Figure 6: It’s not clear what experiment the left plot is for. This needs to be explained in the caption. The figure text and line size would be easier to see if they were larger.

Line 244: Please explain how the ERPUA has been measured or derived, this isn’t clear.

Section 4.2: it is important to state that these results only hold true when the pool main axis is parallel to the ambient wind. I am not convinced that this would be the case if the wind was at an angle.

Figure 8: do you have the uncertainty bounds for these curves. They would be interesting to see, as the plot in its current form suggests that there is increased evaporation from the larger pool, which is contrary to the text.

Figure 9: what is the right hand plot? Please explain what the features are in the left-hand plot – this image would probably be better in the methods.

Line 274: Did you measure the turbulence intensity? If not, then how can you be sure that there was any significant change/increase? Are there references from previous experiments in these chambers that can be cited to validate this claim?

Line 299-300: Unclear what this sentence is saying. Where are these results?

Line 311: replace “simply confirm” with “confirms”

Author Response

Dear reviewer,

answers are given

regards

Benjamin

Specific comments:

A dedicated chapter was introduced to present the model. It was chosen to put it after experiment since, in practical, it was built using small scale experiment (to fit the unique constant) and was then ‘validated’ or evaluated using medium scale tests.

Throughout the text there are many examples of numbers where the decimal point is given by a comma. Please convert all of these to a full stop “.” and be consistent throughout.

Done, all numbers modified in that way.

Dimensions of the experimental installation were added in the beginning of 2.1: The sections are respectively 0,465 m width and 0.165 m height for the small-scale tests and approximatively 3 m x 3 m for the medium scale one. .

Regarding obstacles, information was added priori to figure 4: Small-scale obstacles were 55 mm long, 21 mm width and height. There were distributed over 700 mm upstream the pool. For large scale test, obstacles were approximatively 600 mm in each direction and were distributed over 15 m upstream the pool.

Velocity profiles were measured with and without obstacles (and with some other distribution). The objective was to get something representing an atmospheric profile having in mind that the used installations are not exactly wind-tunnel and is submitted to disturbance. The installation was designed to make the wind profile as relevant as possible.

More information is needed here about the testing platform design prior to the sections on metrology and experimental design. There are many experimental papers that could potential be cited to justify the choices.

Some clarifications were added regarding the installation. Some references were also added more particularly regarding velocity and its fluctuations measurement (new Ref 20).

Right, r is the ‘exposed length’ of the pool. We have chosen keeping the notation r since it is widely used in the field of risk management even when the pool is not a circle. Clarification was added in the notation.

Clarification added in the paper. (i): The mixing system consist in a section reduction to increase velocity and turbulence and then make the mixture homogeneous. Homogeneity of the mixture was then verified through the lack of fluctuation when measuring concentration with the FTIR (ii) Mean velocity and velocity fluctuations were measured using MacCaffrey probes [19] associated with pressure transducer. The velocity was measured closed to the pool, in the middle of the duct height for all test and on several point along the height, same longitudinal position, to determine the velocity profile and establish the link between the middle height velocity and the profile. Following [20], these probes also enable estimating the turbulent velocity, about 4% of the average one. (iii) This information was added, 10 hz for the velocity, 1 hz for the FTIR.

Clarification added on that aspect: “During the experiment, the pool was filled with the ammonia solution. The evaporation was considered starting when filling the pool was over.“

Finally, and most critically, these tests do not appear to have been carried out under the same conditions with regard to concentration or temperature.

The difference in terms of concentration is due to the fact that we have chosen given the real measured concentration of the pool instead of the ‘commercial’ value (this value was added in the table). Consequently, some differences are observed in the initial concentration. While this parameter strongly influences the evaporation rate, this difference is not so important regarding the comparison made to evaluate the influence of other parameters. We also add missing quantities for small-scale tests compared to medium-scale ones.

The authors need to provide a lot more clarity as to the impact these differences will have.

In particular, the difference in pool temperatures (and the differences between air temp and pool temp) is a significant difference between the two experiment sets, which needs explaining.

When dealing with large scale test, we cannot manage the temperature and ‘use’ the ambient value as it is. This induce a variation on this parameter. Some elements were introduced in the paper focused on that, typically a comparison between the temperature evolution for two comparable cases with a different initial temperature.

Some information added but the whole uncertainties evaluation is too complex and not directly linked with the topic of the paper. We typically used the uncertainties on each measurement system (velocity probe, temperature, FTIR) and combined then using the well-known uncertainties equation to get the global value.

3.2 was modified to chapter 5.

Agree. The text was changed a little bit regarding these. While many source of errors exist, authors believe that the most important is the translation of the diffusion process in the ambient through a velocity with a given exponent in current correlation. As discussed now in the paper we have choosen the MacKak and Matsugu correlation since it is largely used in the field of risk assessement but we have in mind this aspect and will continue working on. A specific discussion was added in the conclusion to clarify this: “Furthermore, it should be kept in mind that such an evaporation correlation is based on a certain effect of the velocity of external flow and a strong improvement may consist in adding a new module that consists in evaluating the real diffusion process in the surrounding air.”

Similarly the differences in the first 1200 seconds between the small-scale and large-scale experiments presented in Figure 7 needs explaining.

Difference are stronger during the first 250 s, this is due to the fact that filling the pool is longer and produce more disturbance at medium scale. This was added prior to new fig 10 (previously figure 7) : “It should be notice that the difference observed during the 250 first s are induced by the flooding. This flooding is longer and produce more disturbance for medium scale test.”

The authors make claims that their findings on the role of atmospheric turbulence can be extrapolated to other atmospheric conditions (stable, neutral, unstable). I do not think these are justified. Firstly, there is no proof given that the turbulent intensity actually did vary in the single experiment conducted and secondly without testing any of these environments (or even just more than one different turbulent intensity), there is no evidence that such a result would hold. I do not believe that the paper should be published with the turbulence section and request that section 4.3.1, lines 304-308 and line 25-26 in the abstract are removed.

The test presented is not strongly demonstrating the low influence of atmospheric turbulence. Text in the abstract modified to the following: The surrounding turbulence effect was also estimated, and it was shown that this seems having a low influence on the evaporation rate.

The last sentence of 4.3.1 was removed, this is in good accordance with the comment on the limitation on the model and the role of the velocity and its power in the evaporation correlation that could be improved. The equivalent sentence wes also removed from the conclusion.

Information about the velocity scaling are given in 2.1. Since the scaling is Froude based with ratio of 30 and 10 for the length, this means ratio of 3.1 and 5.5 for the velocity. The range of ‘real’ velocity then covered is 1.55 m/s to 8.2 m/s that is quite the whole range of considered velocity for atmospheric dispersion consideration in risk analysis.

The mentioned paper were considered and added to the references. They clearly confirm the observed tendency presented in this paper.

Corrections needed:

Line 10: use of “toxic distances” and “long distances” is unclear. This sentence needs improving.

Sentence modified to the following : Until now, while safety distances resulting from a toxic pool evaporation may lead to hundreds of meters

Line 14: specify what the “velocity profile” is of

Sentence modified to the following : For both scales, the experimental vertical velocity profile was built to reproduce an atmospheric profile after applying the Froude scaling procedure

Line 19: remove “especially”

Done

Line 25-26: remove sentence (see comments above)

Line 32: Sentence is unclear

Modified to the following: In the context of land use planning, toxic dispersion modelling resulting from pool evaporation should reached hundreds of meters in terms of human possible consequences.

Line 35: It would be helpful to explain why ammonia is important and relevant for this study, i.e. provide some more context

Following sentence added : This product was considered since it is largely used in the industry with various concentration, typically for water treatment in nuclear power station.

Line 36: remove “of”

Done

Line 47: please specify whether the “Given temperature” is for the liquid or the air

Calrification added. The sentence was modified to the following :

Pv vapor pressure at the pool temperature and concentration (Pa),

Line 49: spelling of weight is incorrect

Modified

Line 53: what does the star * in the equation mean?

This * is used for the multiplication sign. It appears clearer with number to use it while it is implicit with letters.

Figure 1: The axes font size needs to be increased to make it more legible, also the second horizontal axis title cannot be read as it is overlaid by the plot

Figure 1 was modified in that way.

Figure 1 caption: Please specify whether this is for ammonia or any liquid

It concerns only ammonia, caption modified

Line 64: the authors say “all these approaches”, but this is not clear. Please be more specific and provide the citations.

Sentence modified to the following : Most of evaporation models used in the literature are based on a constant vapor pressure along the evaporation process.

Line 66: “a gradient” of what?

A gradient of concentration, precision added : that a concentration gradient should occur in the pool and lead to an unsteady phenomenon.

Line 68-69: Explain what the two scales are and what you mean by scale factor. Without more detail this is not clear at this point in the text.

More elements regarding the scale factor were introduced after figure 2, line 100 and following.

Line 82: Figure 3 is referred to before Figure 2, so these two figures need to be swapped around.

Figures were swapped. This figure was also split into two figures quoted at different positions.

Figure 3: it would be preferable for these images to all be aligned.

Modified

Line 86: “sale” should be scale

Modified

Line 88-90: The units here are confusing. I think it should be 0.316m x 0.316m and 1m x 1m. Also suggest changing 3cm to 0.03m so that consistent units are used.

Text modified to the following : When dealing with small scale test, the pool is 0,1 m², 0.316 m x 0.316 m with 0,03 m depth. In this situation, s equals 31.6. For medium scale test, the reference size is 1 m², 1 m x1 m, s equals 10.

Line 91-92: replace “process consists in” with “is”

Modified

Line 94: “an atmospheric profile”

Word ‘an’ added

Line 94-95: change to “for the medium scale test are shown in Figure 2”

Modified.

Line 97: Please provide details as to why this velocity profile has been chosen. Is 3.2m/s representative of something specific? The context of what this relates to in the real world is needed.

The velocity profile was chosen to reproduce the worst case for dispersion as defined in the French regulation but also valid elsewhere, i.e. a stable atmosphere with a velocity close to 3 m/s.

Figure 2: The legend and axes labels need to be in English

Modified

Line 109: should “technics” be “techniques”?

Modified

Line 110-111: “to evaluate the atmospheric concentration downstream of the pool”

Modified, word atmospheric added

Line 114: replace “managed” with “made”

Done

Line 126: pool “depth” would be more consistent here

Modified

Line 144-145: I don’t think that 11 tests is too many to cover in this paper.

New results were added.

Following modifications made: (i) temperature scale was modified to make the curve visibility better. (ii) – Since all rates are given by s, we prefer to plot the graph in second, even this lead large numbers. (iii) – This is explain in the text before, it was added to the caption: this represents the uncertainties.

Line 163: I’d argue that this isn’t “obvious” to an inexperienced reader, so the authors need to explain why this is important

Clarification added: Introducing such a difference in dispersion model, i.e. a division by 6 of the evaporation rate means a strong reduction of the gas quantity emitted to the atmosphere and, consequently a shorter distance for dilution. Consequence of this is a reduction of the distance where toxic threshold is reached.

Line 169: Figure 3 should be Figure 4

Modified

Line 176: If three temperatures were taken, then what is the purple line in Figure 4

showing? Is it one of these or is it the mean? This needs to be explained.

The explanation was added together with the curve that presents the three temperature.

The paragraph was modified to the following for clarification : As mentioned in paragraph 1.1, vapor pressure, a key factor for evaporation, is also a function of the pool concentration. Considering the total emission of ammonia along the test, final concentration of the solution, assuming an homogeneous concentration, should be 24%. Estimating the evaporation rate using this concentration and the measured temperature indicates an evaporation rate reduction of about 55% of the initial value. While, during the experiment, the measured evaporation rate is about 20% of the initial value, an additional mechanism should be investigated.

Line 178: Replace NH3 with ammonia

Done

Line 202: yes, it “could” be, but is this what you have done? What is the time-step used?

Yes, it is. Clarification added in chapter 5.

Line205-207: This detail needs to be explained and described in the methods section rather than here

Clarification added with the sampling description and results, see after new figure 16.

This was modified. The figure was split and distributed in two different places to clarify this. The left one was coming from the model.

Lines 213-220: The first half of this paragraph belongs in the discussion (213-217), whereas the second half should be in the methods (217-220).

See previous comment. Figure and associated comments were split in two different paragraph.

Line 216: I’d suggest that “safe” is not the right word here. “cautious” or “conservative” would be better.

Modifed. (line 392 in the new text)

Lines 227-232: This is poorly worded and lacking details about what parameters were varied. Where are the results for test 2?

Test 2 is mainly for demonstrating the test reproducibility. Its curve was added when dealing with the effect of the initial pool concentration. Paragraph modified to the following :

Two main parameters influence was tested during the small-scale campaign. The first tested parameter was the initial pool concentration. Reducing the initial concentration, test n°1, should reduce the initial evaporation rate and consequently the gradient formation kinetic. This is confirmed by experimental measurement, Figure 5. The experimental curve on this figure indicates a smoothest variation of the evaporation rate in the 20 first minutes. This curve also shows that the proposed model is in quite good agreement with the experiment even it over-estimated the evaporation rate in the first 5 minutes.

Finally, it is important to note the reproducibility of the evaporation test while comparing test 1 and test 2 that are quasi-similar and give the same evaporation rate value and evolution tendency.

Figure 5: Evolution of the evaporation rate for a lower initial ammonia pool concentration.

The second tested parameter is the ambient velocity. Figure 6 shows the evaporation rate measurement for test n°4, that correspond to a 1.5 m/s ambient velocity. This confirm that increasing the ambient velocity lead to increase the evaporation rate. This test also shows that the equilibrium reached after 20 min is quite independent of the velocity. This curve also shows that the model tends to over-estimate the evaporation rate.

Figure 6: Evaporation rate over time, small scale evaluation of the pool concentration influence (left) and velocity influence (right).

Figure 6: It’s not clear what experiment the left plot is for. This needs to be explained in the caption. The figure text and line size would be easier to see if they were larger.

Modified, see previous comment

Line 244: Please explain how the ERPUA has been measured or derived, this isn’t clear.

ERPUA is the measured evaporation rate divided by the pool surface. Clarification added : Considering the reference test cases, the first quantity to compare is the evaporation rate per unit area (ERPUA) for both scales, obtained by divided the measured evaporation rate by the pool surface

Authors fully agree, the title was modified to suit with the content, the evaporation rate is considering along the ‘exposed length, not the surface as initially mentioned in the title. That’s typically why we have chosen modifying the length keeping the width identical.

Uncertainties bounds was added to all curves.

Figure 9: what is the right hand plot? Please explain what the features are in the left-hand plot – this image would probably be better in the methods.

Left hand picture moved to chapter 2 and right hand one enlarged. Figure 9 caption modified to the following for clarification : Influence of local upstream turbulence on the evaporation rate

This was just to test the influence of this parameters. Some clarifications were added in the description and the results presentation. The associated text was also modified to limit the conclusion to a modification of local turbulence instead of generalization to the influence of atmospheric turbulence.

Line 299-300: Unclear what this sentence is saying. Where are these results?

A graph was added with temperature at different height to show that, see new figure 7.

Line 311: replace “simply confirm” with “confirms”

Done

Author Response File: Author Response.docx

Reviewer 3 Report

I think the results of your work are significant to a large number of people, and may have a significant impact on the evaluation of down-wind hazards from liquid pools. Your paper is concise and clear and there are just few places in which the English needs a bit of refinement to make the meaning clear. My main concerns with the paper are as follows:

You present comparisons between observations and model predictions but do not present any details of the model (which I believe from your conclusions is called EVAP-Tox?) or how the concentration sampling and temperature data acquired in the experiments are used in the model. I think you either need to present the details of the model or provide a reference to a paper which describes it.
Although you provide details of 5 small scale tests and 6 medium scale tests you do not present all the data. This reduces the confidence which can be placed in your conclusions. E.g in Figure 6 (which I don't think is correct according to the descriptions) why do you not also present the data for the 3 concentrations and 3 windspeeds?
I like that you have presented the uncertainties in evaporation rate in Figure 4, but why do you not then continue this in the other figures? This would be very helpful in assessing how good the model predictions are and the strength of your conclusions.
While I commend you for only citing those references you feel are necessary, I think a few more are required to place your work and model in context, references to mass transfer theory and work by Kunsch (1998) and Brighton (1985) for example

Specific points:

I think Figure 3 should be split and labelled arrows added to identify the various instruments e.g. FTIR probe

You should provide further details of where the various measurements were made and how many there were (e.g velocities)

Figure 1 – needs adjusting as the ‘liquid concentration’ label is covering the scale which the reader needs to see later.

Your definition of the Froude number (line 84) is incorrect

You say you wanted an atmospheric profile – was your target a log one for neutral conditions?

Figure 2 the labelling needs to be put into English, also two profiles are labelled the same, but that is not explained in the text.

You talk about creating realistic turbulence intensity, but need to describe what the characteristics of the elements and the resulting roughness values – were similar non-dimensional profiles achieved in the small and medium scale tests?

Lines 125-128, These are difficult to follow and should re-worded, several=3?

129 remove question mark

Table 1 ‘Test conditions: small scale’, ambient not ambient

Given that the evaporation rate is sensitive to temperatures and temperature differences, I think you need to comment on the fact that the medium scale tests were performed at temperature 10 degrees higher than the small scale ones. Was this because they were in different facilities?

Lines 197-198 – I think you need to either describe your model here or provide a reference to a document which describes it in detail. At present I do not know if it is an equation or a 1-D, 2-D or 3D finite difference model etc

Line 208-209 It would be helpful if you provided a plot which to illustrate the level of agreement

Figure 5 I would split this in two to help the reader.

Line 205 I think you need to say how the sampling was done and at what heights.

Line 208-209 ‘quite good agreement’ – can you insert a plot to show this?

Line 219-220 –Please can you provide some further explanation of why the diffusion coefficient was multiplied by 50? – perhaps also of how wind velocity and temperature are input into the model?

Line 236 – I do not agree that the predictions are ‘precise’

Line 241- I think you should also discuss the temperature profile – later I think you say that it is not important, but do not present any data to substantiate that.

Line 248 Perhaps: .. dependence of the evaporation rate on source area as low, and confirms the Froude number scaling approach.

Line 256-257 experimentally – in fact the measured ERPUA was higher …..increasing the length - because the concentration of ammonia above the pool would increase with distance? Line 258 should be figure 8?

Suggest delete 263-265 266 suggest 4.3

Line 270 – on what basis was this chosen? What was the expected effect? Can you quantify the 2 turbulence conditions?

Figure 9 – suggest split into 2 to assist the reader.

280 suggest 4.4 Line 294 – you say 1 hour earlier.

299-300 – you have not previously discussed the temperature gradient results previously

309-310 –the English is clumsy here – perhaps: these experiments have demonstrated that the rate of evaporation of liquid pools containing ammonia is dependent on the Froude number and not their areas.

Line 315 – EVAP-Tox not mentioned previously 318-319 evolution of pool concentration ….but also as a function of pool depth …

Author Response

Dear reviewers,

answers are given in the attached file.

regards

B. Truchot

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

Thank you for addressing my previous comments, however, I think you have rushed the process. There are a number of obvious errors, for example the caption for Figure 9 is incorrect and equation 4 appears twice and there is still am 'Ambiant' in Table 2. More particularly you have not critically reviewed the English throughout the text. Although the English can be understood there are a large number of grammatical errors: inconsistencies in tense and singular/plural forms which need correcting. I suggest that you try and get a native English speaker to proof read the text. I would be happy to review your paper again when you have thoroughly proof read and critically reviewed the text.

There are a couple of technical details which I think are still missing: the area density of the roughness elements used in the experimental facilities, the depth of the pool in the medium scale experiment and I am rather confused by your description of the 'section reduction' which doesn't seem to be shown in the figures. I am not sure what or where it is. You also say that it increases velocity and turbulence - however, it is my understanding that a flow contraction will increase velocity and reduce turbulence, unless the contraction includes a mesh for example?

Author Response

Dear reviewer,

thanks for your second round of comment. Modifications were done Following your advice.

Caption of fig9 change to:

Figure 9: Evaporation rate over time, small scale evaluation of the velocity influence, tesn°4 corresponds to 1.5 m/s ambient Velocity.

And the second Eq4 change to Eq5.

Strong modification also made by a native speaker for English improvement.

Regards,

Benjamin TRUCHOT

Article Menu

Experimental and Numerical Evaluation of Toxic Pool Evaporation

Further Information

Guidelines

MDPI Initiatives

Follow MDPI