Improving the Homogenization of the Liquid-Solid Mixture Using a Tandem of Impellers in a Baffled Industrial Reactor
Round 1
Reviewer 1 Report
The article 'Improving the homogenization of the liquid-solid mixture us-
ing a tandem of impellers in a baffled industrial reactor' compares CFD results of a single and a dual impeller mixing tank configuration in terms of solid suspension for a specific case.
While the paper is well written in terms of structure and language, it shows a substantial lack of fundamental scientific rules. And this is not necessarily related to the work itself but to the content of the paper. From a scientific point of view all necessary information to replicate the (numerical) experiment should be presented and this is not the case here.
- Material properties are not fully given (e.g. viscosity).
- Initial conditions are not given (e.g. solid content)
- How post-processing is done is not explained:
- E.g. Fig. 3-9: Are these snapshots in time or time-averaged values? Makes a huge difference in such transient cases.
- Nomenclature/explanation of variables would be beneficial
- Fig 5, 7, 9: hat does tandem impeller down / up mean?
- Message of the images is sometimes unclear (e.g. isosurfaces of 3a). Cross section plots are better.
- Line 147 ff: Where is 'above' and 'below' of the axial plane?
- Line 189, power calculation: How is this done? Based on forces acting on the impeller or based on energy dissipation? And how well agree these two values?
- which equations are solved? Which phase interactions are considered with which model and model parameters? If they are given in one of the two other papers then I would be good to reference them correctly.
Furthermore I would like to raise the following questions:
- line 30: I would try to avoid advertise Ansys products. Numerous solid suspension simulations were done with other codes too, like OpenFOAM, STAR-CCM+, ...
- line 34 and 112, citing multiple references: It would be good to indicate why each of these papers is worth to be cited. Why did you mention them? Specifically in line 112 I do not see the link.
- line 113: why 90 s? Did you then really reach a kind of pseudo-steady state? If so, can this be shown?
- Fig. 2: I believe a cross section visualizing the mesh would give a better idea of the resolution.
- Does the used Eulerian Multiphase model consider particle-particle interaction? if so, how? if not, why not?
- The mesh in general (1M Tets) for a vessel of 6m x 3m seems a bit coarse, e.g. to resolve particle cluster formation. That is a common issue in fluidized beds and requires correction terms.
- Line 79: I do not understand why you are restricted to tetraeder meshes? Other researches used hexahedron and even polyhedron meshes for solid suspension simulations.
- How is the free surface between water and gas modeled? And how is the interaction between solid particles - water and solid particles - gas treated?
- Line 133, distribution of the volume fraction: how is this evaluated? based on the reactor volume or based on the liquid volume? If the latter, does air entrained by the upper impeller (as visible in Fig. 3b) does not affect the results?
- Line 143, gas production by reaction: That arguement holds also for the single impeller.
Comments for author File: 
Comments.docx
Author Response
Dear esteemed Reviewer 1,
Please find in the attached file our response, marked with green, to your kind recommendations for improving au research article. All the changes in the revised manuscript are marked with the colour green.
Thank you very much for all your suggestions!
Author Response File: Author Response.docx
Reviewer 2 Report
Re: Improving the homogenization of the liquid-solid mixture us- 2 ing a tandem of impellers in a baffled industrial reactor
The authors report a numerical study of tandem configurations of three-blade impellers in a stirred tank reactor. It is a good study but there are a few points including major ones that require dressing before the article can be considered publication quality. The authors should refer to the following comments:
- Line 39 - You mentioned that: “This research paper represents the third paper from a series of three..”. References of the first two papers need to be added for information and completeness.
- The literature review is too shallow and brief and does not sufficiently explore the previous work on this topic including experimental studies.
- As this is a CFD study, section 2 should be more appropriately named rather than “Materials and methods”
- Line 151 - “bellow” change to “below”. Please check all throughout the manuscript for typos like this, there are many others, so proofreading is necessary.
- Please “the three axial planes situated above (y+), at the impellers level (y0) and bellow the impellers (y-)” should be depicted in one of the figures.
- Mesh independency study was not done. This is important for the results to be trusted.
- Also for the results to be trusted, and more importantly, validation with experimental data needs to be done. If this can be done even with secondary data from the literature, that will be satisfactory. Otherwise, it will be difficult to recommend this manuscript for acceptance.
- The reviewer feels that the discussion of results is very shallow. Given that a sliding mesh was used rather than the multiple reference frame method, a deeper analysis of the transient behaviour should be done. For example how the homogeneity or the phase fraction distribution of the solid changes as the mixing progressed from start to finish for each impeller.
- As a result of the last two reasons, the reviewer feels that a major revision is needed to make the paper more meaningful and useful to the reader.
Author Response
Dear esteemed Reviewer 2,
Please find in the attached file our response, marked with green, to your kind recommendations for improving au research article. All the changes in the revised manuscript are marked with the colour green.
Thank you very much for all your suggestions!
Author Response File: Author Response.docx
Reviewer 3 Report
This manuscript presents a CFD study on the mixing of liquid-solid mixture with impeller. A single impeller and two tandem impellers were tested. It is found the tandem impellers can have better mixing performance in terms of better distribution of solid. The flow fields were also discussed. The results and discussion are generally clear. But there are some major issues that need to be addressed or clarified:
- The numerical model needs more detailed description. There should also be some kind of validation. In the manuscript, the author stated, "This value might be even lower because the numerical simulation of the flow did not take in consideration the lift effect of the gas bubbles developed...". This looks like the model has some deficiencies, but authors didn't discuss.
- Line 103-105, how is the time step calculated?
- The discussion and conclusion are too narrow, only include the study on the two kinds of impellers. As a research article, the discussion should be linked with a broader research scope and be compared to related studied.
- The authors discussed that they have three papers on this topic. From the reference list, two other papers are also under review. Since the results of this manuscript are too thin, the authors can consider combining those two manuscripts.
- Fig.3, the final state is analysed, but unsteady flow was simulated. Then the average results should be analysed.
- The discussion can be better organized. When the authors mentioned "To compare the performance of the stirring mechanism.." (Line 147), and then discussed the velocity distribution, it's more like the following part is to analyse the flow field to understand the reasons of the performance, but not to compare the performance.
Author Response
Dear esteemed Reviewer 3,
Please find in the attached file our response, marked with green, to your kind recommendations for improving au research article. All the changes in the revised manuscript are marked with the colour green.
Thank you very much for all your suggestions!
Author Response File: Author Response.docx
Reviewer 4 Report
The authors of “Improving the homogenization of the liquid-solid mixture using a tandem of impellers in a baffled industrial reactor” present the results of simulations of the mixing of suspended solids in a stirred tank reactor. The primary topic is chemical or process engineering, which is not listed as one of the topics of Applied Sciences, though the application of the modelling techniques can be linked to physics and mechanical engineering via fluid mechanics.
Please describe in a little more detail how you model the liquid phase, the solid phase and the head gas. What assumptions did you make in applying the models? Which drag force did you apply for the granular model. How many dispersed phases did you consider? How many continuous phases did you consider? How did you deal with the gas liquid interface? Did you apply surface compression? Did you include the effect of surface tension?
The authors compare the impact of using two impellers operating in tandem have on the suspension of the solids compared to a single impeller. Both the single and double impeller are operated in the down pumping mode. This means that flow exits the impeller is mainly axial and predominantly with negative values. The presence of the second down pumping impeller enhances the velocities experienced by the solids and increases the amount of solids suspended in the reactor.
It is difficult to qualitatively compare the solids fractions distribution presented in Figure as only a proportion of the stirred tank with a single impeller is plotted (i.e. without the bottom region or the liquid interface region). Also the level of the liquid is not indicated in the figures. A quantitative distribution of the solids distribution at difference radial positions would also be helpful in demonstrating the improved suspension of the solids.
Plotting the velocity vectors or streamlines in an axisymmetric plane would also help the reader understand the impact that using two down-pumping impellers has on the vertical structures formed i.e. the second impeller suppresses the formation of counter-rotating vortices above the lower impeller and the vortices should extend to around the height of the second impeller. This is cause of the lower distribution of solids in the upper region of the reactor.
Please use Figure 1 to indicate the locations of the radial profiles given in Figures 5, 7 and 9. Also please plot a solid line on the 0 axis to help the reader understand which parts of the velocity profile are negative and positive.
Further suggestions for use of language and references are given in the attached pdf of the manuscript.
Author Response
Dear esteemed Reviewer 4,
Please find in the attached file our response, marked with green, to your kind recommendations for improving au research article. All the changes in the revised manuscript are marked with the colour green.
Thank you very much for all your suggestions!
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Your comments mostly satisfied my questions (assuming all missing material/method descriptions are done in the first 2 parts of this series). Also with the new figures it has cleary improved.
From your answers I think you should cleary mention that this is an industrial work, not a fundamental research. E.g. from my point of view a snapshot of the results after 90s is not very accurate, time-averaged values would be better. But I agree that you can still draw some conclusions with a certain uncertainty based on snapshot. Same holds true for the mesh size: And I think it is important to discuss this, that the absolute value of your results might have a large error, but it might be sufficient to compare the two different designs relative to each other.
And I still do not understand how the free surface is treated. Looking forward to read the two other papers.
Author Response
Dear esteemed Reviewer,
Please see the attachment with our response to your observations.
Author Response File: Author Response.docx
Reviewer 2 Report
The authors have swept the major comments from the first review round under the carpet. They have not modified the manuscript accordingly, which for CFD studies, the last two are mandatory. Specifically:
- the literature review remains shallow and has not been improved
- The author's comment that the mesh "may be assumed fine enough" is not acceptable for a research paper especially in a peer reviewed journal of Energies' standing. Readers will question the quality of the results and the paper.
- In addition to the above and even more important is validation. The reader has not been informed based on the manuscript about the currently operating mixer. As you have built the mixer, then that is good, compare its mixing characteristics with your CFD. It may be that the CFD model/fabricated mixer are not optimal and an iterative approach is required. Until you do that, the study is not complete and cannot be accepted.
- results analysis remains shallow. The simulation is transient and time analysis of the results was not done e.g. how well the mixing process progresses comparing both configurations is a minimum expected analysis that has to be shown.
- following up from comment 5 above, the plots and images from figs 3-10 appear to be instantaneous snapshots and not the ensemble mean of the flow, which is more representative of the mixing behaviour. That way the profiles of the velocity distributions will be more symmetrical (i.e., less uneven as presented) and more meaningful.
The authors must address these, otherwise, the reviewer cannot recommend publishing the article in Applied Sciences.
Author Response
Dear esteemed Reviewer,
Please see the attachment with our response to your observations.
Author Response File: Author Response.docx
Reviewer 3 Report
Most previous comments have been satisfactorily addressed. But for the analysis of final state only, it is likely the solid distribution will fluctuate with time. It is better to check the average values, unless authors can show the fluctuation is small.
Author Response
Dear esteemed Reviewer,
Please see the attachment with our response to your observations.
Author Response File: Author Response.docx
Reviewer 4 Report
Just two minor corrections:
1) When quoting the use of an ANSYS software please use the correct convention around registered marks. This can be found here https://www.ansys.com/en-gb/academic/terms-and-conditions.
2) The flow structures you present in Figure 10 need a fuller description to show the understanding behind why two impellers are better than one. This is primarily down to the flow at the side of the tank being stronger and extending beyond the height of the lower impeller. This flow at the side of the tank is significantly stronger than when a single impeller is used. This vertical flow pushes the solid particles further up into the tank than the case with the single impeller. The solids can then potentially be drawn into the region around the upper impeller and remain suspended.
Author Response
Dear esteemed Reviewer,
Please see the attachment with our response to your observations.
Author Response File: Author Response.docx
Round 3
Reviewer 2 Report
Improving the homogenization of the liquid-solid mixture using a tandem of impellers in a baffled industrial reactor
The authors have clarified points, included statements and revised the document to let the reader make judgements on the quality of the work carried out. Hence the reviewer recommends that the paper be accepted after the following minor corrections.
1. Given figure 2, the both structured and unstructured mesh elements were used. Hence, Line 87 may be rephrased as follows: As regards the mesh quality, we use a hybrid tetrahedral/hexahedral mesh cells as recommended for such complex geometries, [2], [3].
2. Also, rephrase Line 90 as follows: Although the mesh influence upon numerical results was not addressed and a larger mesh element number may generate more accurate results, the size of the mesh employed in this study may be assumed to be appropriate to compare the performances of the single and tandem impellers
3. Mention that figures 3-10 are instantaneous results.
Author Response
Dear esteemed Reviewer 2,
Please find in the attachment our response to your recommendations.
Author Response File: Author Response.docx
