Central Composite Experiment Design (CCD)-Response Surface Method (RSM) to Optimize the Sintering Process of Ti-6Al-4V Alloy
Round 1
Reviewer 1 Report
Is the reviewer wondering what's new? The introductory part is not enough to convince readers of the rationale for the research work. Furthermore, if the aim of this work is to find an optimal set of parameters for the sintering conditions it is necessary to apodically explain which conditions are optimized and it is even necessary to explain why these sintering conditions are optimized. Authors should highlight the novelty present in their work by demonstrating that the optimization analysis of the chosen sintering conditions is an original work within the scientific literature showing what it adds to previous technology.
This work addresses a specific technological problem: optimization.
In the introductory paragraph it is not necessary to explain what a CCD is and what it is potential, these aspects are well known. You should argue your choice in using this statistical method and above all you must demonstrate that the CCD is, among the statistical method, the best method for your objectives related to: what you want to optimize and the chosen factors.
Line 109 replaces MINNTAB with MINITAB.
It is necessary to introduce a paragraph "2.1 Experimental strategy" describing the statistical method used, the reason why a CCD was used, show here the table of the experimental configuration and explain the number of central points (replicas of the central points). Prove that the repeatability of the error calculated at the central points is representative of the entire analysis domain.
In the paragraph "Sample preparation" it is necessary to show the final powder distribution also using a table.
As sintered samples show a high surface roughness, the Archimedes method using water should be less accurate than the gas pycnometer. Can you downgrade the high precision of your measurements?
What kind of standard was followed for the compression test?
What kind of standard was followed for the micro hardness test?
Could you explain this sentence better? Line 173-175 "While for the sintering temperature of 1300°C, the fluctuation of the EDS peak pattern at 1200°C was more stable, indicating that the diffusion of Ti, Al and V was more uniform."
Section "Pore analysis"- What kind of analysis was conducted to determine the porosity distribution shown in Fig. 4? What the standard followed, the accuracy of the gauges, etc ... In addition, some images showing pores are needed.An analysis of the shape and dimensions of pores should be required.
Line 223-224:Obviously, the compressive strength and micro-hardness increased signifi-223 cantly with the increase of sintering temperature. Could you explain this sentence better?
Line 226 Can you better explain how the micro-hardness campaign was conducted.
Line 229 Broadly define the term compression ratio.
Fig.5 Just one question about this. Are these fig.5 results for compression resistance or traction resistance?
Line 241 “especially to find out the optimal sintering parameters, an ANOVA … was carried out” This sentence is uncorrect. The ANOVA is only aimed at investigate the influence of sintering parameters.
Line 244 “the value of R-sq was employed to judge the reliability of the model” This sentence is uncorrect R-sq show the correlation level between output and input of the model, not its reliabitity. On the contrary in order to state that the model is reliable you need to verify the model assumptions with residuals analisys that you have to show in this paper.
Line 254 Eq. (1) How is it possible that secondary order factor containing Ht is significant while Ht main factor is not?
Line 267 Eq. (3) How is it possible that secondary order factor containing Ht and Hr are significant while Ht and Hr main factor are not?
Fig.8 there is no surface diagram of the compression ratio. Also, the surface diagram of compressive strength and density is very similar in shape, are you sure they are correct?
Fig.9 this graph wants to demonstrate the optimal parameters but to me it is not clear, you have to demonstrate the optimal set of parameters for density and compression force mathematically.
Modify the conclusions based on the change you will make in response to the reviewer'comments
Author Response
Dear Editor:
Thank you for your exciting letter of “Major Revisions”. According to the reviewers’ comments, we revised the manuscript carefully and examined it many times before submitted. Taking this opportunity, we also wish to thank the reviewers for their constructive comments and suggestions. The following is our detailed response to the reviewers’ comments.Since the picture is not displayed in the box, it is recommended that you directly view the attachment
Metals-1057679
Title: Sintering process optimization of PM Ti-6Al-4V alloy by Central Composite Designs (CCDs) method.
Answers to Reviewer 1:
Thank you very much for your useful comments and suggestions on the whole manuscript. We have checked the manuscript carefully and revised it (highlighted in yellow type in the "Revised Manuscript") according to the comments point by point, and the detailed answers and corrections are listed below:
- Is the reviewer wondering what's new? The introductory part is not enough to convince readers of the rationale for the research work. Furthermore, if the aim of this work is to find an optimal set of parameters for the sintering conditions it is necessary to apodically. explain which conditions are optimized and it is even necessary to explain why these sintering conditions are optimized. Authors should highlight the novelty present in their work by demonstrating that the optimization analysis of the chosen sintering conditions is an original work within the scientific literature showing what it adds to previous technology.
Response: Dear reviewers, thank you very much for your reminder. In order to prove that the optimization analysis of the selected sintering conditions was an original work, and analyze the data of other researchers. In the introduction, I added a discussion on the work and shortcomings of other researchers, as well as the necessity of doing this work, and supplemented relevant references. “The process of sintering was the essential step in traditional powder metallurgy. J.-M. Oh. et al. have studied the effect of sintering temperature on the sintering performance of TC4, and the result shown that sintering temperature has a major impact on sintering performance [35]. In addition, Zhang et al. have also studied the influence of the main process factors on the sintering process of titanium hydride powder, the control of variables such as sintering temperature (St), holding time (Ht) and heating rate (Hr) were the main parameters that affect the performance of the sample during the sintering process. However, according to previous studies [31,33,34,35], these studies did not specifically analyze the degree of impact of each parameter and lack of research on the effect of the interaction of three process parameters. Therefore, this study used the experimental design method of combining CCD and RSM to study the influence of the interactions of various factors (St, Ht, Hr) and the specific contribution of each parameter, then determined the optimal sintering parameters.
The references as follow: [31] Zhang, J.M.; Yi, J.H.; Lei, T.; Fang, Z.G.; Dehydrogenation and sintering process of titanium hydride for manufacture titanium and titanium Alloy. Adv. Mater. Res.,2012, Vol. 616-618, 1823-1829. Lenth, R.V.; Response-Surface Methods in R, Using rsm. J. Stat. Softw., 2009, Vol. 30, 7-15. [33] Zhu, Y.L.; Yang, S.L.; Ma, L. Direct sintering of hydrogenated titanium powder and vanadium aluminum alloy powder experimental study of TC4 alloy. Sichuan Metall. 2019, Vol. 41, 28-31. [34] Ivasishin, O.; Moxson, V. Low-cost titanium hydride powder metallurgy. In Titanium powder metallurgy (Science, Technology and Applications); Ma, Q., Ed.; Butterworth-Heinemann.: Oxford, UK, 2015; pp.115-148. [35] Oh, J.M., Heo, K.H.; Kim, W.B.; Sintering Properties of Ti–6Al–4V Alloys Prepared Using Ti/TiH2 Powders. Mater. Trans., 2013, Vol. 54, 119-221.
2.This work addresses a specific technological problem: optimization. In the introductory paragraph it is not necessary to explain what a CCD is and what it is potential, these aspects are well known. You should argue your choice in using this statistical method and above all you must demonstrate that the CCD is, among the statistical method, the best method for your objectives related to: what you want to optimize and the chosen factors.
Response: Dear reviewers, thank you very much for your reminder. After referring to your suggestions, I reviewed the relevant literature to re-examine the factors selected in my experiment and gave the reasons for choosing this method to optimize. Zhang et al. have studied and the influence of the main process factors on the sintering process of titanium hydride powder, the control of variables such as sintering temperature (St), holding time (Ht) and heating rate (Hr) were the main parameters that affect the performance of the sample during the sintering process. These studies did not specifically analyze the degree of impact of each parameter and lack of research on the effect of the interaction of three process parameters [31]. Therefore, this study used the experimental design method of combining CCD and RSM to study the influence of the interactions of various factors (St, Ht, Hr) and the specific contribution of each parameter, then determined the optimal sintering parameters. First, a number of experiments were conducted according to the central composite experimental design (CCD), then response surface methodology (RSM) was adapted to analyze and optimize the data. RSM was a method under optimizing experimental conditions, suitable for solving related problems of nonlinear data processing. [26] The combination of central composite experimental design (CCD) and response surface method (RSM) could involve the cross-effects of variables on the overall response and obtain the specific contribution of each parameter. In addition, the correlation between the influencing factor and the response value could also be represented by a three-dimensional rendering diagram or a two-dimensional contour diagram [27, 32].
3.Line 109 replaces MINNTAB with MINITAB.
Response: Dear reviewers, thank you very much for your reminder. I have revised the mistake in manuscript.
4.It is necessary to introduce a paragraph "2.1 Experimental strategy" describing the statistical method used, the reason why a CCD was used, show here the table of the experimental configuration and explain the number of central points (replicas of the central points). Prove that the repeatability of the error calculated at the central points is representative of the entire analysis domain.
Response: Dear reviewers, thank you very much for your reminder. In the experimental design part of the revised manuscript, the reasons and advantage for using CCD were added, and the experimental configuration table was supplemented. Central experiment design (CCD) added an axial point and a center point to the two-level factorial design point to improve the simulation of the response surface. The entire experimental model could be seen as a cube, the axial point was set at the center of each surface, the center point was set at the center of the cube. The center point was used to provide constant accuracy and pure error estimation. The spherical model was sensitive to the center point, but the was stable to the center point, the center point of cube model was added only for pure estimation.
Table 1 sintering parameter and their levels
|
Sintering parameters |
Level 1 |
Level 2 |
Level 3 |
|
|
-1 |
0 |
+1 |
|
Sintering Temperature |
1100℃ |
1200℃ |
1300℃ |
|
Holding Time |
120min |
180min |
240min |
|
Heating Rate |
5℃/ min |
7.5℃/ min |
10℃/ min |
5.In the paragraph "Sample preparation" it is necessary to show the final powder distribution also using a table.
Response: Dear reviewers, thank you very much for your suggestion. I have supplemented the particle size distribution diagram of the raw material powder in the manuscript according to your suggestions.
Fig. 1 the particle size distribution diagram of TH2 powder
6.As sintered samples show a high surface roughness, the Archimedes method using water should be less accurate than the gas pycnometer. Can you downgrade the high precision of your measurements?
Response: Dear reviewers, thank you very much for your comment. In order to avoid the measurement error of Archimedes drainage method, I used seven parallel samples for measurement and took the average to ensure that the error was within the acceptable range.
7.What kind of standard was followed for the compression test?
Response: Dear reviewers, thank you very much for your comment. This test was carried out under national standard GB/T 7314-87.
8.What kind of standard was followed for the micro hardness test?
Response: Dear reviewers, thank you very much for your comment. This test was carried out under national standard GB/T 4342-1991.
9.Could you explain this sentence better? Line 173-175 "While for the sintering temperature of 1300°C, the fluctuation of the EDS peak pattern at 1200°C was more stable, indicating that the diffusion of Ti, Al and V was more uniform."
Response: Dear reviewers, thank you very much for your comment. I re-described this sentence and revised it in the manuscript. The revised version as follows: “The EDS spectrum in Fig. 1 represented the fluctuation of element content. The stronger the peak intensity, the higher the element content in this region. When the sintering temperature was 1100℃, the EDS spectrum fluctuates significantly, and the rising peaks (1, 3) of titanium could be observed, corresponding to the aggregation areas (AB and CD) of titanium. In this area where the titanium element was concentrated, Al and V have not yet been fully diffused, and the α+β structure has not yet been formed. Therefore, the red dashed area in Fig.1(a2) could be divided into the α phase area, and the red outer dashed area was (α+β) Phase area. When the sintering temperature was 1200℃, the EDS spectra of the three elements have no obvious fluctuation, indicating that the diffusion of Ti, Al and V was almost completed at this temperature.”
10.Section "Pore analysis"- What kind of analysis was conducted to determine the porosity distribution shown in Fig. 4? What the standard followed, the accuracy of the gauges, etc ... In addition, some images showing pores are needed. An analysis of the shape and dimensions of pores should be required.
Response: Dear reviewers, thank you very much for your comment. I made corrections and additions based on your suggestions. The way I conduct pore analysis is as follows. Used “count/size” in image pro-plus to perform porosity and size statistics on metallographic images which showed in Fig. 4, combined pixel conversion to calculate its porosity and maximum pore size. Table 1 clearly showed the statistics of porosity and maximum pore size in various sinter conditions. As seen in Fig. 4, with the increase of sintering temperature, the number of pores was significantly reduced, and the shape of the pores changed from irregular continuous pores to spherical pores.
Fig.5 Pore morphology and distribution of these samples when sintered at various conditions
11.Line 223-224: Obviously, the compressive strength and micro-hardness increased significantly with the increase of sintering temperature. Could you explain this sentence better?
Response: Dear reviewers, thank you very much for your suggestion. According to your suggestion, I rewrite the sentence. Changed version: “Apparently, the compressive strength increased significantly with the rise of sintering temperature. The hardness value was determined by the initial and continuous plastic deformation resistance. While strength of the material increases with the increasing resistance to plastic deformation, hence a correlation between strength and hardness could be established.”
12.Line 226 Can you better explain how the micro-hardness campaign was conducted.
Response: Dear reviewers, thank you very much for your suggestion. I have a specific description of the micro-hardness campaign in the part of experimental tests: “The microhardness of these samples was measured by HVS-1000 microhardness tester (Shanghai special precision instrument) with the loading weight of 500g and the holding time of 10s, which conformed the “metallic materials – Vickers microhardness test” (GBT4342-1991). In addition, at least 20 scattered points of each sample were measured and then averaged to ensure the reliability of the results.”
13.Line 229 Broadly define the term compression ratio.Fig.5 Just one question about this. Are these fig.5 results for compression resistance or traction resistance?
Response: Dear reviewers, thank you very much for your reminder. The compression ratio was defined as the ratio of the deformation of the material after yielding to the initial length. It was compression resistance not traction resistance.
14.Line 241 “especially to find out the optimal sintering parameters, an ANOVA … was carried out” This sentence is uncorrect. The ANOVA is only aimed at investigate the influence of sintering parameters.
Response: Dear reviewers, thank you very much for your reminder and revised my mistake. The mistake sentence has been changed to “In order to find out the influence of sintering parameters on the properties of PM TC4 alloy, an ANOVA (analyze of variance) was carried out.”
15.Line 244 “the value of R-sq was employed to judge the reliability of the model” This sentence is uncorrect R-sq show the correlation level between output and input of the model, not its reliabitity. On the contrary in order to state that the model is reliable you need to verify the model assumptions with residuals analisys that you have to show in this paper.
Response: Dear reviewers, thank you very much for your reminder and corrected the questions. I added residual analysis and normal probability plot: “According the normal probability plot for process parameters in Fig. 7, the experiments data in the plot (Fig. 7) able to generated straight line, which mean the residuals were normally distributed. Therefore, the plot proved that the assumptions of the model were correct and the data was reliable.”
Fig.8 Normal Probability Plot for process parameters
16.Line 254 Eq. (1) How is it possible that secondary order factor containing Ht is significant while is not?
Response: Dear reviewers, thank you very much for your comment. In Equation 1, secondary order factor containing Ht and main factor also containing Ht.
17.Line 267 Eq. (3) How is it possible that secondary order factor containing Ht and Hr are significant while Ht and Hr main factor are not?
Response: Dear reviewers, thank you very much for your reminder. For the compressive strength, the influence of the sintering temperature is higher than 97%. Therefore, the influence of the other two factors is very small and can be basically ignored, but their combined influence may have some effect.
18.Fig.8 there is no surface diagram of the compression ratio. Also, the surface diagram of compressive strength and density is very similar in shape, are you sure they are correct?
Response: Dear reviewers, thanks for your comments. According to your suggestion, I have added a surface diagram of the compression ratio and described in the revised manuscript. And the trends of the two surface graphs are somewhat similar, but the details are different, I am sure the surface diagram of compressive strength and density were correct.
- Fig.9 this graph wants to demonstrate the optimal parameters but to me it is not clear, you have to demonstrate the optimal set of parameters for density and compression force mathematically.
Response: Dear reviewers, thanks for your advice. In order to clearly show the optimization results (1300℃, 144min, and 10℃/min), as well as the predicted values and verification experiment results, I have added specific tables (Table 5: Analysis of the verification experiments) in the revised manuscript.
Table 5: Analysis of the verification experiments
|
Compressive Strength |
Compression Ratio |
Density |
||||||
|
Predict |
Actual |
Error |
Predict |
Actual |
Error |
Predict |
Actual |
Error |
|
1382 MPa |
1398 MPa |
1.1% |
0.323 |
0.300 |
7.1% |
4332 kg.m-3 |
4333 kg.m-3 |
0.02% |
We hope that our revised version will be satisfactory for publishing in Metals. Anyway, more suggestions and comments are welcome. Great thanks to you and the reviewers for the time and effort you expend on this paper.
Best regards.
Yungui Chen ([email protected]) Line 109 replaces MINNTAB with MINITAB.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This work investigates the sintering parameters in Ti-6Al-4V. I strongly recommend considering the following points before publication:
The grammar has serious faults and must be revised along the manuscript.
“Sintering process optimization of PM Ti-6Al-4V alloy by Cen-2tral Composite Designs (CCDs) method” – process can be avoided.
“In addition, 19verification experiments were carried out under the optimum conditions,” --- to explain how these optimal conditions are reached
“AM took the high-energy beam (laser beam/electron beam/electric arc, etc.) as the heat source, and themetal components were stacked layer by layer” --- the current advantages and limitations of AM of Ti-6Al-4V must be mentioned using references: see doi DOI: 10.1016/j.scriptamat.2020.02.043
“its main advantage was 76higher concentration of vacancies and dislocations, due to the volume change caused by 77hydrogen decomposition in the sintering process of titanium hydride, which promoted 78the diffusion between elements.” --- please explain why is this an advantage in which sense. And how is this linked to the novelty of this work.
“The Central Composite Design (CCD) was an experimental design 91method based on a 2-level full factorial and partial test design.” --- this methods is not well explained in the introduction and must be clearly described for non experts.
The novelty of this works must be clearly explained in the introduction. Prior works using different approaches for the sintering optimization of Ti64 must be explained and compared to the approach used here.
It is not explained how the element analysis of Fig.1 are obtained. The limitations of the used technique must be explained and compared to other advanced methods using prior reports: see DOI: 10.1107/S1600577515023528. This is relevant in the discussion regarding the alloy’s homogenity.
“The sintered density of the samples was measured by the Archimedes method” --- please describe the accuracy of this method and the associated error with respect to other techniques such as micro-tomography.
Fig.2. close to 2theta 45deg peaks can be visualized though they must be indexed and identified to get insight about the coexisting phases in the material besides alpha and beta.
The sintering conditions such as the atmosphere used are not fully described. This is important owing to possible oxygen pick-up.
Can the authors comment on the higher differences of UTS between samples in 1100C compared to those of 1300C and 1200?
The cooling rate determine the microstructure characteristics of the lamellar microstructures obtained though is not analysed. The effect of this variable must be considered for the sintering optimization and its role explained in detail.
The authors investigate sintering variables such as porosity which are more critical for tensile testing. However, this work studies the compressive behaviour of this material. This is not a keen approach and the authors must explain which implications compression tests have over tensile tests. Tensile data must be provided.
Fig.7. the Main effect plots show the differences for the absolut values but it is not clear how important are these differences when comparing between the different parameters.
Why does the compressive strength vary between conditions but not the ductility? The effect of ductility must be explained in the text. Please explain this in the discussion in base of the possible deformation mechanisms.
The sintering parameters for ti64 available at the literature must be provided and compared to those obtained here.
Author Response
Dear Editor:
Thank you for your exciting letter of “Major Revisions”. According to the reviewers’ comments, we revised the manuscript carefully and examined it many times before submitted. Taking this opportunity, we also wish to thank the reviewers for their constructive comments and suggestions. The following is our detailed response to the reviewers’ comments.Since the picture is not displayed in the box, it is recommended that you directly view the attachment
Metals-1057679
Title: Sintering process optimization of PM Ti-6Al-4V alloy by Central Composite Designs (CCDs) method.
Answers to Reviewer 2:
Thank you very much for your useful comments and suggestions on the whole manuscript. We have checked the manuscript carefully and revised it (highlighted in yellow type in the "Revised Manuscript") according to the comments point by point, and the detailed answers and corrections are listed below:
Reviewer 2：
This work investigates the sintering parameters in Ti-6Al-4V. I strongly recommend considering the following points before publication:
1.The grammar has serious faults and must be revised along the manuscript. “Sintering process optimization of PM Ti-6Al-4V alloy by Cen-2tral Composite Designs method” – process can be avoided.
Response: Dear reviewers, thank you very much for your reminder. I have carefully revised the grammar in the manuscript and changed “Sintering process optimization of PM Ti-6Al-4V alloy by Cen-2tral Composite Designs method” to “Central composite experiment design (CCD)-response surface method (RSM) to optimize the sintering process of Ti-6Al-4V alloy”
- “In addition, verification experiments were carried out under the optimum condition,” --- to explain how these optimal conditions are reached
Response: Dear reviewers, thank you very much for your comment. In order to reach the optimal conditions, a number of experiments were conducted according to the design of experiment (DOE), then the optimal condition was reached through the analysis of response surface methodology (RSM). This specific process is described in the article.
- “AM took the high-energy beam (laser beam/electron beam/electric arc, etc.) as the heat source, and the metal components were stacked layer by layer” --- the current advantages and limitations of AM of Ti-6Al-4V must be mentioned using references: see doi DOI: 10.1016/j.scriptamat.2020.02.043
Response: Dear reviewers, thank you very much for your suggestion. According to your suggestions, I carefully read this article you recommended, and analyzed the advantages and disadvantages of AM in the introduction of manuscript. Then added the reference: [29] Barriobero-Vila, P.; Artzt, K.; Schel, N.; Siggel M.; Gussone, J.; Mapping the geometry of Ti-6Al-4V: From martensite decomposition to localized spheroidization during selective laser melting. Scr. Mater., 2020, Vol.182,48-52.
- “its main advantage was higher concentration of vacancies and dislocations, due to the volume change caused by hydrogen decomposition in the sintering process of titanium hydride, which promoted the diffusion between elements.” --- please explain why is this an advantage in which sense. And how is this linked to the novelty of this work.
Response: Dear reviewers, thank you very much for your comment. In recent years, titanium hydride powder metallurgy has become a hot topic in the world. The reversible alloying effect of hydrogen plays a role in the sintering process and promotes the sintering process. Compared with the sintered samples of hydrogenated dehydrogenated titanium, hydrogenated powder can obtain higher density and performance. In terms of cost and performance, products made from titanium hydride powder are better than traditional titanium powder metallurgy. The ultimate goal of optimizing sintering conditions in this research is to achieve the best performance of titanium products with high efficiency and low cost. Therefore, titanium hydride powder is used as a raw material to optimize the sintering process in a targeted manner while ensuring the advantages of the raw materials. The optimization result is a combination of raw materials and process advantages.
- “The Central Composite Design (CCD)was an experimental design method based on a 2-level full factorial and partial test design.” --- this method is not well explained in the introduction and must be clearly described for nonexperts.
Response: Dear reviewer, thanks for your suggestions. According to your suggestion, I gave a detailed introduction to the central compound experimental design in the experimental design section of the revised manuscript. In the experimental design part of the revised manuscript, the reasons and advantage for using CCD were added, and the experimental configuration table was supplemented. “Therefore, this study used the experimental design method of combining CCD and RSM to study the influence of the interactions of various factors (St, Ht, Hr) and the specific contribution of each parameter, then determined the optimal sintering parameters. First, a number of experiments were conducted according to the central composite experimental design (CCD), then response surface methodology (RSM) was adapted to analyze and optimize the data. RSM was a method under optimizing experimental conditions, suitable for solving related problems of nonlinear data processing. [26] The combination of central composite experimental design (CCD) and response surface method (RSM) could involve the cross-effects of variables on the overall response and obtain the specific contribution of each parameter. In addition, the correlation between the influencing factor and the response value could also be represented by a three-dimensional rendering diagram or a two-dimensional contour diagram.”
6.The novelty of this works must be clearly explained in the introduction. Prior works using different approaches for the sintering optimization of Ti64 must be explained and compared to the approach used here.
Response: Dear reviewers, thank you very much for your suggestion. According to your suggestion, I have made supplements and corrections in the introduction to the different optimization processes and the novelty of my work. In the introduction, the advantages compared to the work discussed by Zhang et al and J.-M. Oh. et al. on the sintering process were added: The process of sintering was the essential step in traditional powder metallurgy. J.-M. Oh. et al. have studied the effect of sintering temperature on the sintering performance of TC4, and the result shown that sintering temperature has a major impact on sintering performance [35]. In addition, Zhang et al. have also studied the influence of the main process factors on the sintering process of titanium hydride powder, the control of variables such as sintering temperature (St), holding time (Ht) and heating rate (Hr) were the main parameters that affect the performance of the sample during the sintering process. However, according to previous studies [31,33,34,35], these studies did not specifically analyze the degree of impact of each parameter and lack of research on the effect of the interaction of three process parameters. Therefore, this study used the experimental design method of combining CCD and RSM to study the influence of the interactions of various factors (St, Ht, Hr) and the specific contribution of each parameter, then determined the optimal sintering parameters. First, a number of experiments were conducted according to the central composite experimental design (CCD), then response surface methodology (RSM) was adapted to analyze and optimize the data. RSM was a method under optimizing experimental conditions, suitable for solving related problems of nonlinear data processing. [26] The combination of central composite experimental design (CCD) and response surface method (RSM) could involve the cross-effects of variables on the overall response and obtain the specific contribution of each parameter. In addition, the correlation between the influencing factor and the response value could also be represented by a three-dimensional rendering diagram or a two-dimensional contour diagram [27, 32].
7.It is not explained how the element analysis of Fig.1 are obtained. The limitations of the used technique must be explained and compared to other advanced methods using prior reports: see DOI: 10.1107/S1600577515023528. This is relevant in the discussion regarding the alloy’s homogenity.
Response: Dear reviewers, thank you very much for your comment. According to your suggestion, I supplemented the working principle of EDS in the revised version of the manuscript, and carefully referenced the literature you provided, and cited this article. “EDS could perform element analysis by measuring the energy of the excited X-ray photons of solid materials, for samples with less complex element distribution, this measure was efficient and stable.”
- “The sintered density of the samples was measured by the Archimedes method” --- please describe the accuracy of this method and the associated error with respect to other techniques such as micro-tomography.
Response: Dear reviewers, thank you very much for your comment. In order to avoid the measurement error of Archimedes drainage method, I used seven parallel samples for measurement and took the average to reduce errors.
9.Fig.2. close to 2theta 45deg peaks can be visualized though they must be indexed and identified to get insight about the coexisting phases in the material besides alpha and beta.
Response: Dear reviewers, thank you very much for your reminder. Considered your question about the weak peaks near 45 degrees in the XRD spectrum, I re-use the Jade software to perform peak-finding operations on the XRD data, but I did not find the corresponding phase. I guess it may be caused by an error in the operation. If necessary, I could provide a complete XRD diagram in the follow-up manuscript.
10.The sintering conditions such as the atmosphere used are not fully described. This is important owing to possible oxygen pick-up.
Response: Dear reviewers, thank you very much for your comment. I described the sintering atmosphere in 2.1: these samples were sintered under high vacuum (5×10-3 Pa).
11.Can the authors comment on the higher differences of UTS between samples in 1100℃ compared to those of 1300℃ and 1200℃?
Response: Dear reviewers, thank you very much for your comment. The compressive strength of the material was very sensitive to the presence of internal pores. When the temperature was 1100°C, the internal porosity of the material was higher, and the pore size was too large, and it was easier to initiate cracks and cause fracture. Moreover, the existence of large pores reduces the true cross-section of the material, which makes the critical fracture strength of the material drop significantly. When the temperature was between 1200°C and 1300°C, the probability of large holes decreases sharply, and the density fluctuations are relatively moderate, so the intensity fluctuations were smaller than those at 1100°C.
12.The cooling rate determine the microstructure characteristics of the lamellar microstructures obtained though is not analyzed. The effect of this variable must be considered for the sintering optimization and its role explained in detail.
Response: Dear reviewers, thank you very much for your comment. Due to equipment limitations, the sintering process in this experiment was carried out in a vacuum molybdenum wire furnace, and the cooling rate was fixed, which could only be cooled by furnace cooling. Hope to get your understanding.
13.The authors investigate sintering variables such as porosity which are more critical for tensile testing. However, this work studies the compressive behaviour of this material. This is not a keen approach and the authors must explain which implications compression tests have over tensile tests. Tensile data must be provided.
Response: Dear reviewers, thank you very much for your suggestion. It is a very meaningful idea. The tensile test was indeed an effective way to test the mechanical properties of materials, especially sensitive to porosity. At the beginning of the design experiment, I also considered doing a tensile experiment, but at that time, there was no suitable mold to carry out the compaction to prepare the tensile sample. Due to the epidemic, a lot of related work in this article was postponed, and it was too late to make the mold of tensile sample, and finally decided to conduct a compression experiment test. Moreover, titanium alloy products not only receive tensile stress during service, but some titanium alloy products also work under compressive stress for a long time. Therefore, it is also meaningful to optimize the sintering parameters for the compressive properties of the material. If possible, we will make compact molds for tensile specimens, apply the optimized parameters, and test the tensile properties in subsequent work.
14.Fig.7. the Main effect plots show the differences for the absolute values but it is not clear how important are these differences when comparing between the different parameters.
Response: Dear reviewers, thank you very much for your comment. Due to the newly added pictures in the manuscript (Figure 7 was changed to Figure 10), this Figure directly reflected the influence trend of each parameter on each performance, the difference in absolute value could reflect the effect of each sintering parameter on performance. For instance, as shown in Fig. 10 (c1, c2, c3), there was an almost linear relationship between sintering temperature and compressive strength, compression ratio, sintering density. This obvious change means that the influence of sintering temperature on the sintering process was very important. The influence of other parameters on sintering could also be visually observed in the part 3.4.2 of the manuscript mainly analyzes these trends.
15.Why does the compressive strength vary between conditions but not the ductility? The effect of ductility must be explained in the text. Please explain this in the discussion in base of the possible deformation mechanisms.
Response: Dear reviewers, thank you very much for your comment. For the cases of the compressive strength vary between conditions but not the ductility, which can be clarified as follows: Since the strength of the material was greatly affected by internal defects (pores) during compression and deformation, the strength changes significantly with density. Because the mechanism of compression deformation is not as sensitive to plasticity as strength. Compared with strength, the change of plasticity with density was not obvious, and the change of grain size and microstructure has a greater impact on plasticity. (Mainly reflected in the change with the holding time). Based on your suggestion, I explained the change in compression ratio in the manuscript and added it in 3.4.2. Because of compression ratio was a reflection of the ability of a metal to resist plastic deformation without breaking. When the grains were coarsened, the number of grains in a certain volume became less, the slip system during plastic deformation was less, and the deformation could not be uniformly dispersed, which was not conducive to plastic deformation. Therefore, when the sample was kept at a sufficiently high sintering temperature for too long, the compression rate would decrease due to the coarsening of the crystal grains.
16.The sintering parameters for Ti64 available at the literature must be provided and compared to those obtained here.
Response: Dear reviewers, thank you very much for your comment. After consulting the literature, I listed the parameter results of other researchers for vacuum sintering Ti6Al4V, which proved the advantage of my parameter optimization.
|
1. 45 μm dehydride powder, compacted at 400 MPa to 77% green density, heated ingraphite furnace first in vacuum of 0.1 Pa at 5 °C/min to 850 °C, then pressurized furnace using 0.11 MPa atmosphere argon, continued heating to 1400 °C for 480 min, final density 92.6%. [Yu, C., Cao, P., M.I. Jones, Titanium powder sintering in a graphite furnace and mechanical properties of sintered parts, Metals 7 (2017) 67–80] |
|
2. Titanium hydride powder and aluminum-vanadium alloy were used as raw materials, and Ti-6Al-4V alloy was directly sintered by powder metallurgy prealloying method. Direct degreasing and dehydrogenation during sintering were carried out to obtain Ti-6Al-4V alloy sintered body samples. when the sintering temperature is 1200 ℃, the holding time is 240 min, the density reaches 95.47%; [Zhu, Y.L.; Yang, S.L.; Ma, L. Direct sintering of hydrogenated titanium powder and vanadium aluminum alloy powder experimental study of TC4 alloy. Sichuan Metall. 2019, Vol. 41, 28-31.] |
|
3. The density of TC4 alloy prepared by cold pressing and vacuum sintering is basically between 94% and 96%. [Kumar, P.; Ravi Chandran, K. S.; Strength-Ductility Property Maps of Powder Metallurgy (PM) Ti-6Al-4V Alloy: A Critical Review of Processing-Structure-Property Relationships, Metall. Mater. Trans. A, 2017, Vol. 48, 2301-2319.] |
|
4. Pressure-less microwave sintering of titanium alloy: sintering parameters identified were: St of 1280°C, Hr of 8°C/minute, and Ht of 78 minutes. The compressive yield strength of 541.34 ± 48.81 MPa, density of 90.8%. [Singh, D.; Pandey, P. M.; Sundaram, D. K.; Optimization of Pressure-Less Microwave Sintering of Ti6Al4V by Response Surface Methodology. Mater. Manu. Process, 2018, Vol. 33, 1835-1844.] |
The references added in the revised manuscript are as follows:
[29] Barriobero-Vila, P.; Artzt, K.; Schel, N.; Siggel M.; Gussone, J.; Mapping the geometry of Ti-6Al-4V: From martensite decomposition to localized spheroidization during selective laser melting. Scr. Mater., 2020, Vol.182,48-52.
[30] Alfeld, M.; Wahabzada M.; Bauckhage, C.; Kersting K.; Wellenreuther G.; Non-negative matrix factorization for the near real-time interpretation of absorption effects in elemental distribution images acquired by X-ray fluorescence imaging. J. Synchrotron Rad., 2016, Vol. 23, 579-589.
[31] Zhang, J.M.; Yi, J.H.; Lei, T.; Fang, Z.G.; Dehydrogenation and sintering process of titanium hydride for manufacture titanium and titanium Alloy. Adv. Mater. Res.,2012, Vol. 616-618, 1823-1829.
[32] Lenth, R.V.; Response-Surface Methods in R, Using rsm. J. Stat. Softw., 2009, Vol. 30, 7-15.
[33] Zhu, Y.L.; Yang, S.L.; Ma, L. Direct sintering of hydrogenated titanium powder and vanadium
aluminum alloy powder experimental study of TC4 alloy. Sichuan Metall. 2019, Vol. 41, 28-31.
[34] Ivasishin, O.; Moxson, V.; Low-cost titanium hydride powder metallurgy. In Titanium powder metallurgy (Science, Technology and Applications); Ma, Q., Ed.; Butterworth-Heinemann.: Oxford, UK, 2015; pp.115-148.
[35] Oh, J.M., Heo, K.H.; Kim, W.B.; Sintering Properties of Ti–6Al–4V Alloys Prepared Using Ti/TiH2 Powders. Mater. Trans., 2013, Vol. 54, 119-221.
We hope that our revised version will be satisfactory for publishing in Metals. Anyway, more suggestions and comments are welcome. Great thanks to you and the reviewers for the time and effort you expend on this paper.
Best regards.
Yungui Chen ([email protected])
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
There is a big methodological mistake in this work: in table 4 are summarized all the points of the CCD used to calculate the ANOVA but the replications of the central point are missing. The way you have carried out the ANOVA is uncorrect, that’s why you have even some weird results like for example interaction effects significant while their main factors are not. Please make experiments for at least 4 more central points and calculate again the ANOVA and the linear models and all the interpretations of the models.
Author Response
Dear Editor:
Thank you for your exciting letter of “Major Revisions”. According to the reviewers’ comments, we revised the manuscript carefully and examined it many times before submitted. Taking this opportunity, we also wish to thank the reviewers for their constructive comments and suggestions. The following is our detailed response to the reviewers’ comments.（For your specific view, you'd better view it in the attachment or manuscript）
Metals-1057679
Title: Central composite experiment design (CCD)-response surface method (RSM) to optimize the sintering process of Ti-6Al-4V alloy
Answers to Reviewer 1:
Thank you very much for your useful comments and suggestions on the whole manuscript. We have checked the manuscript carefully and revised it (The changes in the second round are marked in red font) according to the comments point by point, and the detailed answers and corrections are listed below:
Reviewer 1: There is a big methodological mistake in this work: in table 4 are summarized all the points of the CCD used to calculate the ANOVA but the replications of the central point are missing. The way you have carried out the ANOVA is uncorrect, that’s why you have even some weird results like for example interaction effects significant while their main factors are not. Please make experiments for at least 4 more central points and calculate again the ANOVA and the linear models and all the interpretations of the models.
Response: Dear reviewers, thank you very much for your reminder. According to the problems you pointed out, we have carefully reorganized all the experimental data, and according to your suggestions, we have added the central point of the experiment. According to the new data, re-analyze the surface. After reanalysis, the abnormal result problem you mentioned was solved, and the verification experiment was re-run after obtaining new optimization data. And the related charts have been corrected (marked in red font). Data about the center point (for example: hardness, strength, porosity, etc.) in the manuscript is replaced by the average value. Due to the re-analysis of the experimental data, the following charts and equations have been changed. The specific analysis was added to the original manuscript (red font);
Table 4. The input factors and output results of sintering experiments.
|
Running sequence |
Point type |
St (℃) |
Ht (min) |
Hr (℃/min) |
Compression Strength (MPa) |
Density (kg.m-3) |
Compression Ratio |
|
1 |
1 |
1100 |
120 |
5 |
1090 |
4106.6 |
0.310 |
|
2 |
1 |
1300 |
120 |
5 |
1343 |
4321.1 |
0.330 |
|
3 |
1 |
1100 |
240 |
5 |
1120 |
4145.3 |
0.330 |
|
4 |
1 |
1300 |
240 |
5 |
1390 |
4338.0 |
0.265 |
|
5 |
1 |
1100 |
120 |
10 |
1101 |
4105.1 |
0.320 |
|
6 |
1 |
1300 |
120 |
10 |
1381 |
4317.5 |
0.323 |
|
7 |
1 |
1100 |
240 |
10 |
1113 |
4153.0 |
0.320 |
|
8 |
1 |
1300 |
240 |
10 |
1357 |
4362.0 |
0.268 |
|
9 |
-1 |
1100 |
180 |
7.5 |
1100 |
4134.0 |
0.325 |
|
10 |
-1 |
1300 |
180 |
7.5 |
1387 |
4341.7 |
0.300 |
|
11 |
-1 |
1200 |
120 |
7.5 |
1260 |
4251.2 |
0.310 |
|
12 |
-1 |
1200 |
240 |
7.5 |
1270 |
4276.0 |
0.280 |
|
13 |
-1 |
1200 |
180 |
5 |
1291 |
4273.9 |
0.330 |
|
14 |
-1 |
1200 |
180 |
10 |
1267 |
4271.9 |
0.310 |
|
15 |
0 |
1200 |
180 |
7.5 |
1275 |
4262.9 |
0.325 |
|
16 |
0 |
1200 |
180 |
7.5 |
1262 |
4262.5 |
0.315 |
|
17 |
0 |
1200 |
180 |
7.5 |
1269 |
4262.8 |
0.325 |
|
18 |
0 |
1200 |
180 |
7.5 |
1265 |
4261.8 |
0.318 |
|
19 |
0 |
1200 |
180 |
7.5 |
1271 |
4262.7 |
0.326 |
Figure 8. Histogram of the contribution of St, Ht, and Hr to the performance of sintered samples.
|
Density= |
(1) |
|
Compression Ratio = |
(2) |
|
Compressive Strength = |
(3) |
Fig.9 Normal Probability Plot for process parameters
Table 5. ANOVA results of CCDs experiments.
|
|
Compressive Strength |
Density |
Compression Ratio |
|
Regression |
|
|
|
|
(P value) |
0.001 |
0.001 |
0.001 |
|
DF |
9 |
9 |
9 |
|
(Residual error) |
10 |
10 |
10 |
|
(Lack of fit and pure error) |
5 and 5 |
5 and 5 |
5 and 5 |
|
P-value for each response |
|
|
|
|
Linear terms |
0.000 |
0.000 |
0.002 |
|
St |
0.000 |
0.000 |
0.001 |
|
Ht |
0.151 |
0.028 |
0.009 |
|
Hr |
0.018 |
0.103 |
0.934 |
|
Square terms |
0.000 |
0.000 |
0.013 |
|
St*St |
0.001 |
0.000 |
0.885 |
|
Ht*Ht |
0.204 |
0.124 |
0.004 |
|
Hr*Hr |
0.399 |
0.279 |
0.196 |
|
Interactions |
0.068 |
0.081 |
0.001 |
|
St*Ht |
0.553 |
0.120 |
0.000 |
|
St*Hr |
1.000 |
0.413 |
0.799 |
|
Hr*Ht |
0.012 |
0.043 |
0.734 |
|
Total (AdjSS) |
185655 |
116578 |
0.007412 |
|
R-sq (Adsj) |
98.90% |
99.51% |
83.16% |
Fig. 811 3D surface diagram of (a) density and (b) compression ratio vs. sintering temperature and holding time; (c) compressive strength vs. sintering temperature and heating rate
Figure 12. The optimal response result graph of this experiment.
Table 6: Analysis of the verification experiments
|
Compressive Strength |
Compression Ratio |
Density |
||||||
|
Predict |
Actual |
Error |
Predict |
Actual |
Error |
Predict |
Actual |
Error |
|
1370MPa |
1388 MPa |
1.3% |
0.326 |
0.310 |
4.9% |
4330 kg.m-3 |
4332 kg.m-3 |
0.04% |
We hope that our revised version will be satisfactory for publishing in Metals. Anyway, more suggestions and comments are welcome. Great thanks to you and the reviewers for the time and effort you expend on this paper.
Best regards.
Yungui Chen ([email protected])
Author Response File: Author Response.pdf
Reviewer 2 Report
The comments written by the reviewer have been addressed.
Author Response
Dear Editor:
Thank you for your exciting letter of “Major Revisions”. According to the reviewers’ comments, we revised the manuscript carefully and examined it many times before submitted. Taking this opportunity, we also wish to thank the reviewers for their constructive comments and suggestions. The following is our detailed response to the reviewers’ comments.（For your specific view, you'd better view it in the attachment or manuscript）
Metals-1057679
Title: Central composite experiment design (CCD)-response surface method (RSM) to optimize the sintering process of Ti-6Al-4V alloy
Answers to Reviewer 2:
Thank you very much for your useful comments and suggestions on the whole manuscript. We have checked the manuscript carefully and revised it (The changes in the second round are marked in red font) according to the comments point by point, and the detailed answers and corrections are listed below:
Response: Dear reviewers, thank you very much for your reminder. According to the problems you pointed out, we have carefully reorganized all the experimental data, and according to your suggestions, we have added the central point of the experiment. According to the new data, re-analyze the surface. After reanalysis, the abnormal result problem you mentioned was solved, and the verification experiment was re-run after obtaining new optimization data. And the related charts have been corrected (marked in red font).
We hope that our revised version will be satisfactory for publishing in Metals. Anyway, more suggestions and comments are welcome. Great thanks to you and the reviewers for the time and effort you expend on this paper.
Best regards.
Yungui Chen ([email protected])
Author Response File: Author Response.pdf
Round 3
Reviewer 1 Report
For next paper I suggest to analize densities and possible lack of fusion by using tomographic analysis.
