Droplet Characteristics of Rotating Packed Bed in H2S Absorption: A Computational Fluid Dynamics Analysis
Akinola Adeniyi
Round 1
Reviewer 1 Report
This manuscript describes CFD results of rotating packing bed in H2S absorption. Although this work is of interest to the Research community, the manuscript needs to be improved significantly to become publishable in Processes.
The authors should review carefully the English and make sure the manuscript is up to standard. Although the manuscript is understandable in general, there are sentences which need reviewing. The authors should also ensure that the manuscript flows better, with good transitions between the different parts.
Regarding the technical content, the CFD is not well explained and there is no mesh sensitivity analysis performed or shown here, this needs to be addressed. In addition, the authors mention experimental data which have apparently been used but they don’t show them, and they don’t compare their results with those.
It is not clear how the droplets are injected into the system. How many droplets? What were the options selected for the particles in the settings?
How is Equation 36 obtained? How are the equations in Table 4 obtained?
In Section 3.3.1, it is written: “the droplet diameter in RPB was fitted by experiment data with correlation [17] (), and it also indicates CFD model introduced in this study predicts reasonably the droplet diameter distribution”. I might have missed something, but I cannot see how the CFD model predicts the droplet diameter distribution here.
In Section 3.3.1 again, the authors mention the mesh and the size of the mesh. It would have been useful to show the mesh in this manuscript and, of course, as mentioned previously, perform a mesh sensitivity analysis.
On Page 2, the sentence: “Shi [7]…CFD simulation [9]” is too long and not clear. The following sentence is not clear either: “..and the liquid flow was trend form droplet…”
On Page 21, “The reason why droplet diameter make the difference… with rotating packing”. This sentence needs to be re-written.
In the Conclusion section, the authors state that droplet characteristics in RPB have been analysed by diagrams and correlations, which are compared with available experimental data from the literature. It is not clear how this has been done in the manuscript. I have not seen any clear comparison of the numerical results with the experimental data and/or correlations.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Dear Author,
Basically, the article uses the CFD approach appropriately. For this analysis case:Droplet Characteristics of Rotating Packed Bed in H2S Absorption: A Computational Fluid Dynamics Analysis, the moving mesh reference approach is the most appropriate.
The article does not mention how the author calibrated his CFD model and does not specify whether validation was done or not.
Please specify the two elements. Regards,
Author Response
Dear reviewer:
Thank you for giving me some comments for my manuscript------Droplet Characteristics of Rotating Packed Bed in H2S Absorption: A Computational Fluid Dynamics Analysis. Now, I will response to your comments and give ourselves explanations point-by-point.
Question 1: The article does not mention how the author calibrated his CFD model.
Answer: We have added a grid refinement of RPB, which is different number of cells were tested in different parts, figure 3(b). After the work completed, we found no significant different simulation results in 4349428 cells and 3136368 cells. Therefore, this way is reasonable to calibrate our CFD model..
Question 2: does not specify whether validation was done or not.
Answer: Due to the research method of the article is CFD simulation, in order to verify the reliability of simulated results, we must compare it with experimental data. Guo (Guo, F.; Zheng, C.; Guo, K.; Feng, Y.; Gardner, N.C. Hydrodynamics and mass transfer in cross-flow rotating packed bed. Chemical Engineering Science 1997, 52, 3853-3859) obtained a formula by fitting relevant experiment data,, which indicated the droplet diameter decreases along the radial direction. Both figure 12 and figure 13 all show that the average droplet diameter is negatively correlated with the radius position, which is consistent with Guo’s research. Therefore, it manifests CFD model introduced in this study agree well with the experiment data and predicts reasonably the droplet diameter distribution
Reviewer 3 Report
The work lacks experimental validation. This is understandable by the nature of the application. However, this is a very well developed and prepared article. It is sound in all ramifications.
The article should be checked for a few typographical errors, such as spacing in section 2.4 [is set to 1 to 5mm.Steady...].
There are several places in the paper where droplet diameter distribution would have been more appropriate. For example "3.3.1 Effect of initial droplet diameter on droplet diameter" is confusing to read. Consider "3.3.1 Effect of initial droplet diameter on droplet diameter distribution" instead. See Fig. 15.
Surface tension was not described in the Navier Stokes equation form used. So the Eulerian phase. It is however discribed in the Taylor Analogy Breakup model. Is this coupled into $F_bar$ in (2)?
Chapter 3.2.2: Can you clarify or justify the difference in Qian[16] and your work a little more. 2 Nanoseconds and your 49-85 milliseconds difference.
3.3: Be more discriptive in the definition of "surface renewal frequency". For example does this mean the rate at which the droplets fill up the surface?
3.3.2 ....The centrifugal force gradient is linear to ....[this is as measured by what?].
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Dear authors, thanks for addressing my suggestions. However, most of my initial questions have still not been answered fully, please see my new comments under your answers.
Question 1: There is no mesh sensitivity analysis performed or showed in manuscript.
Answer: We have added a grid refinement of RPB, which is different number of cells were tested in different parts, figure 3(b). After the work completed, we found no significant different simulation results in 4349428 cells and 3136368 cells. Therefore, it is reasonable that the grid refinement replace the mesh sensitivity analysis.
The mesh sensitivity analysis should show the results in the paper, and show that the simulation results obtained with the two meshes (4,349,428 cells and 3,136,368 cells) are identical.
Question 2: The authors mention experimental data which have apparently been used but they don’t show them, and they don’t compare their results with those and in Section 3.3.1, it is written: “the droplet diameter in RPB was fitted by experiment data with correlation, and it also indicates CFD model introduced in this study predicts reasonably the droplet diameter distribution”. I might have missed something, but I cannot see how the CFD model predicts the droplet diameter distribution here.
Answer: Due to the research method of the article is CFD simulation, in order to verify the reliability of simulated results, we must compare it with experimental data. Guo (Guo, F.; Zheng, C.; Guo, K.; Feng, Y.; Gardner, N.C. Hydrodynamics and mass transfer in cross-flow rotating packed bed. Chemical Engineering Science 1997, 52, 3853-3859) obtained a formula by fitting relevant experiment data,? = 0.7284(?/ ??2 ?)0.5, which indicated the droplet diameter decreases along the radial direction.
Both figure 12 and figure 13 all show that the average droplet diameter is negatively correlated with the radius position, which is consistent with Guo’s research. Therefore, it manifests CFD model introduced in this study agree well with the experiment data and predicts reasonably the droplet diameter distribution.
OK, the authors have shown that the droplet diameter decreases in the radial direction, which agree with Guo et al. Can the CFD results be compared to experimental data somehow?
Question 3: It is not clear how the droplets are injected into the system. How many droplets?
Answer: the droplets are injected into RPB from 4 liquid inlets, figure 3(a). As for how many droplets injected into RPB, it relies on different conditions and what results we want to research. Such as, due to the area of droplet inlet and the initial droplet diameter is constant, when we research the droplet velocity and average residence time distribution in RPB, the number of droplets injected into the system depends only on the initial droplet velocity.
This answer is not clear. The authors should include a clear explanation in the manuscript regarding this point. How many droplets did they inject in the system and which approach did they use? What conditions did they apply on the droplets? Not sure I understand that the “number of droplets injected in the system depends only on the initial droplet velocity”. Can the authors explain this?
Question 4: How is Equation 36 obtained? How are the equations in Table 4 obtained?
Answer: Equation 36 is an empirical correlation by fitting the relation of rotating speed, radial position, fluid density and droplet diameter. The equations in Table 4 are derived from the Stokes Law.
The equations listed in Table 4 should be explained. How are they derived from the Stokes law? Where did the authors get those equations from? This needs to be explained clearly in the manuscript.
Question 4: In Section 3.3.1 again, the authors mention the mesh and the size of the mesh. It would have been useful to show the mesh in this manuscript and, of course, as mentioned previously, perform a mesh sensitivity analysis.
Answer: In the former manuscript, we thought that the reason why the droplets of 1mm diameter were rarely broken up is that the mesh size is too large to capture the 1mm droplets. After our detailed analysis, we found this explanation was wrong. The true is that the packing size is too large than the droplets of 1mm diameter and the structure of RPB introduced in the manuscript is not suitable for researching the droplets of 1mm diameter. Therefore, we have modified the explanation and given another reasonable interpretations in the revised version.
The mesh sensitivity analysis is still missing.
Question 5: On Page 2, the sentence: “Shi [7]…CFD simulation [9]” is too long and not clear. The following sentence is not clear either: “..and the liquid flow was trend form droplet…”
Answer: We have re-written these sentences and the more accurate expressions are as follows: Shi [7] verified the flow patterns was various under different rotating speed in CFD simulation. Ouyang emphasized that higher viscosity of liquid led a liquid line predominantly [8]. In another simulation article, Ouyang et al pointed out droplets diameter increased with decreasing of rotating speed, liquid initial velocity and layers number in a Rotor-Stator reactor(RSR) [9]. Conclusively, Xie [10] fingered out the collision and merging of droplets occurred between droplets and packing, and the liquid flow was the trend to form droplet under the influence of surface tension (larger contact angles) and high rotating speeds (1000-1500rpm).
Please check the English. I am not sure you want to write: “Xie [10] fingered out..”, you probably mean: “Xie [10] figured out..”
Author Response
Please see the attachment
Author Response File: Author Response.pdf
