Effect of Beam Power on Intermetallic Compound Formation of Electron Beam-Welded Cu and Al6082-T6 Dissimilar Joints

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

(1) Suggest adding EBSD analysis of microstructure of welded joints with different power levels;

(2) Suggest adding different power welding joint tensile properties and analyzing the relationship between power, intermetallic compounds, and mechanical properties

Author Response

Dear reviewer,

Thank you for your time! Your suggestions helped us to improve our manuscript. According to your remarks the following answers were generated:

 

Reviewer suggestion: Suggest adding EBSD analysis of microstructure of welded joints with different power levels.

Answer to reviewer: Thank you for this suggestion! Currently, it is not possible for us to study the samples using EBSD. This, however, is an excellent suggestion, which would help us significantly in the future for structure characterization of all sorts of samples. In order to gain better knowledge and support our claims for specific texturing of the different samples their preferred crystallographic orientation was determined using the pole density equation. This also helped us gain more insight on the relationship between structure, technological conditions and properties.  

Reviewer suggestion: Suggest adding different power welding joint tensile properties and analyzing the relationship between power, intermetallic compounds, and mechanical properties.

Answer to reviewer:  As mentioned in the article no tensile properties such as yield strength, ultimate tensile strength, elongation, etc. were observed in the case of using a beam power of 1800 and 2400 W. The samples broke apart immediately during the pre-loading stage, thus a stress-strain curve was not obtained. In the case of the sample prepared with a power of 3000 W, only a breaking force was detected and measured. Due to the low overall pressure the software of the machine also did not manage to determine the specific properties such as yield strength, elongation and others, and a stress-strain curve was also not build in this case. Only a braking force value was registered along with an UTS one. Regardless, the relationship between power, formation of IMCs and mechanical properties was elaborated and extended. 

 

Once again the author team would like to thank the respected reviewer for their comments and suggestions!

Best regards,

Georgi Kotlarski

Reviewer 2 Report

Comments and Suggestions for Authors

The presented manuscript "Effect of the Beam Current on the Intermetallic Compounds Formation of Electron Beam Welded Cu and Al6082-T6 Dissimilar Joints" is devoted to the current issue of obtaining joints of aluminum and copper alloys. Scientists have been studying this issue for a long time, but it is difficult to develop a technology for fusion welding of this pair of materials, since they form brittle intermetallic compounds that greatly reduce the ductility of the weld and lead to cracks under the action of their own welding stresses without applying an external load.

Notes:

1. There is no clearly formulated objective of the study.

2. The authors compare welded joints of 8 mm thick plates welded at different power, thereby obtaining different penetration depths, different relative degrees of material penetration, chemical composition of the weld, stress level, etc. The degree of edge penetration and change in the chemical composition of the weld metal, which in this case is more important than the beam power. When welding thinner plates, the results will be different, which the authors also mention in the discussion, referring to the work of Lee et al. [24], so power cannot be considered as a determining factor. From this point of view, the research methodology does not seem to be fully thought out.

3. In lines 81-83, the authors write: "Prior to the welding procedure, both plates were cleaned with acetone and preheated to a temperature of 300 °C. The preheating process was performed in order to reduce the thermal internal stresses caused by the high thermal gradient formed during the EBW process." It is not clear from the text whether this refers to preheating before installing the plates in the vacuum chamber, or heating during the welding process? If the first option, how was the temperature of the samples controlled during welding? What is the actual temperature during welding, taking into account cooling of the samples during the installation of the samples, pumping out the vacuum chamber and aiming at the weld joint? 4. In lines 93-94 the authors write: "The 2 °C per minute cooling process was determined to be sufficient enough to avoid the formation of cracks in the weld seam caused by rapid cooling." Please describe in more detail here the basis on which the cooling rate was determined and how the authors see the causes of cracks in a dissimilar joint during rapid cooling.

5. In lines 245-246 the authors write: "The copper plate exhibited an yield strength (YS) of 262 MPa and an ultimate tensile strength (UTS) of 278 MPa." Judging by this, the copper was in a hardened state due to strain hardening, prolonged heating at a temperature of 300 C should have led to an increase in copper grain size and a strong decrease in hardness, which is apparently confirmed by Fig. 8. I recommend that the authors compare the microstructure of the copper plate before and after welding, estimate the expected temperature and heating time during the welding process and, based on this, determine the causes of the decrease in copper hardness. 6. When describing the results of mechanical tests, the authors provide data on the applied force before failure, which is not quite true, since only the magnitude of stresses can be compared.

As a result:

1. The objectives of the study are unclear.

2. In the conclusions, the authors indicate an increase in the depth of penetration with an increase in the power of the electron beam, which is obvious without experiments. The authors associate the change in microstructure and properties with an increase in power, on the basis of which they state (lines 352-353): "the application of a higher power of the electron beam is preferential". At the same time, the authors do not consider the change in the degree of penetration of the edges of a dissimilar joint and the chemical composition of the weld.

3. It is not clear what new knowledge was obtained as a result of this work.

4. The article is methodologically incorrect, requires extensive revision and is not suitable for publication in its current form.

Author Response

Dear reviewer,

Thank you for your time and your honest and extensive opinion of the presented work. After reading carefully your concerns and remarks the manuscript was edited accordingly. According to your remarks the following answers were generated:

 

Reviewer comment: There is no clearly formulated objective of the study.

Answer to reviewer: Thank you for this remark! The objectives of the article were modified.

Reviewer comment: The authors compare welded joints of 8 mm thick plates welded at different power, thereby obtaining different penetration depths, different relative degrees of material penetration, chemical composition of the weld, stress level, etc. The degree of edge penetration and change in the chemical composition of the weld metal, which in this case is more important than the beam power. When welding thinner plates, the results will be different, which the authors also mention in the discussion, referring to the work of Lee et al. [24], so power cannot be considered as a determining factor. From this point of view, the research methodology does not seem to be fully thought out.

Answer to reviewer: Thank you for this remark! The power of the electron beam does lead to the formation of weld seams of differing size. Varying the power would imminently lead to a change in the microstructure as well. The exact influence of the power of the electron beam on the resultant microstructure was of interest. The primary mechanical property investigated in this work was the microhardness. The tensile properties (good or bad) were added additionally to supplement to the results. It was also mentioned in the article that no actual comparison can be made between the different specimens even if the experiments were successful because the cross-section (the thickness) of the specimens is compromised by the lesser penetration. In that regard the respected reviewer is correct, since only one of the specimens would produce accurate representative data.    

Reviewer comment: In lines 81-83, the authors write: "Prior to the welding procedure, both plates were cleaned with acetone and preheated to a temperature of 300 °C. The preheating process was performed in order to reduce the thermal internal stresses caused by the high thermal gradient formed during the EBW process." It is not clear from the text whether this refers to preheating before installing the plates in the vacuum chamber, or heating during the welding process? If the first option, how was the temperature of the samples controlled during welding? What is the actual temperature during welding, taking into account cooling of the samples during the installation of the samples, pumping out the vacuum chamber and aiming at the weld joint?

Answer to reviewer: Thank you for this remark! Indeed, this detail was no specified previously. The text was edited and the process was clarified. Currently the temperature of the welding plates immediately before the welding process occurs is unknown for us. It is indeed advisable to monitor this as well, however, an attempt to heat the substrates as much as possible with minimal oxidation and contamination was made. For this reason, this temperature was chosen. The processes of natural cooling were indeed not controlled. It would be interesting indeed to vary the temperature of the preheating process instead of just maxing it out to investigate whether that would have an effect (or rather what effect it would have) on the microstructure, properties, welding depth, etc. Also previous research indicates that it is indeed possible to preheat the samples using the electron beam in a vacuum environment which would not only increase the purity of the samples, but also help with decreasing the time between the heating and welding processes. However, such a methodology was not used in the present research. The author team will save this great suggestion for future research.     

Reviewer comment: In lines 93-94 the authors write: "The 2 °C per minute cooling process was determined to be sufficient enough to avoid the formation of cracks in the weld seam caused by rapid cooling." Please describe in more detail here the basis on which the cooling rate was determined and how the authors see the causes of cracks in a dissimilar joint during rapid cooling.

Answer to reviewer: Thank you for this comment! As a research group that investigates electron beam welding we have performed a large quantity of experiments. Previous experience, not only with copper and aluminum, indicated that during the welding process plastic deformation of the plates occurs (along the axis of the plates). In some cases, this deformation is highly pronounced, but in some cases not so much, depending on the materials. Regardless, as a result of this deformation high residual stresses form in the fusion zone. Defects in the form of cracks form along the length of the weld seam in the case of uncontrollable cooling (even at low temperatures). The applying cooling process was implemented in order to improve and slow down the cooling process, thus gradually reduce the residual stresses (stress relieving). It indeed would be interesting how the cooling speed would affect defect propagation and formation, however, this was not investigated in the current research. Instead the most favorable case was chosen (the slowest possible cooling). This was clarified in the text as well.  

Reviewer comment: In lines 245-246 the authors write: "The copper plate exhibited an yield strength (YS) of 262 MPa and an ultimate tensile strength (UTS) of 278 MPa." Judging by this, the copper was in a hardened state due to strain hardening, prolonged heating at a temperature of 300 C should have led to an increase in copper grain size and a strong decrease in hardness, which is apparently confirmed by Fig. 8. I recommend that the authors compare the microstructure of the copper plate before and after welding, estimate the expected temperature and heating time during the welding process and, based on this, determine the causes of the decrease in copper hardness. 6. When describing the results of mechanical tests, the authors provide data on the applied force before failure, which is not quite true, since only the magnitude of stresses can be compared.

Answer to reviewer: Thank you for this comment! The microstructure of the welded plated was investigated before the welding process. Preheated samples were also investigated and compared to the as-delivered ones. The obtained microstructure regarding the aluminum plates is not so representative, but evident enough that a change in the size of the aluminum crystals occurred. This influenced the aluminum plate and reduced its mechanical properties before the welding process. The preheating did not affect the structure and properties of the Cu plate. It is possible some heat treatment process was applied to the Cu bar, which was used to form the plates, however, the manufacturer did not specify such details. The borders between the fusion zone and the aluminum and copper plates were studied and also included in the paper. The paper was overall overhauled and the correlation between the technological conditions, structure and mechanical properties (or lack thereof) was elaborated and extended.

 

 

Once again the author team would like to thank the respected reviewer for their comments and suggestions. This helped us to revise and greatly improve the quality of our manuscript.

Best regards,

Georgi Kotlarski 

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript presents the Cu/Al dissimilar joints using electron beam welding. The following concerns must be addressed:

(1) The abstract is logical and clear. It would be better if the common devices (i.e., XRD, SEM, EDS) used were not described. Generally, the abstract does not describe exactly what devices are used.

(2) At 3000W, only two phases were detected. It is unreasonable not to include the Cu phase, one of the welded materials. Moreover, the Cu side was not fully melted in both cases of low energy (1800 W and 2400 W), and the Cu phase was detected in the weld seam. It is necessary to specify the test area.

 

(3) The description of joint mechanical performance is lacking in this manuscript. Comparison of loads, fracture surfaces, etc. could be considered. Currently, the optimum ultimate tensile strength of the Cu/Al joints is only 50 MPa, much lower than that of the two substrates. To explain why. Also, is there any improvement strategy?

Author Response

Dear reviewer,

Thank you for your time and excellent suggestions! They helped us to significantly improve our manuscript. According to your remarks the following answers were generated:

 

Reviewer suggestion: The abstract is logical and clear. It would be better if the common devices (i.e., XRD, SEM, EDS) used were not described. Generally, the abstract does not describe exactly what devices are used.

Answer to reviewer: Thank you for this remark! The description of all methods used during the experimental work were removed from the abstract.

Reviewer suggestion: At 3000W, only two phases were detected. It is unreasonable not to include the Cu phase, one of the welded materials. Moreover, the Cu side was not fully melted in both cases of low energy (1800 W and 2400 W), and the Cu phase was detected in the weld seam. It is necessary to specify the test area.

Answer to reviewer: Thank you for this remark! The test area was specified and a figure was added to assist readers with a visual representation.  

Reviewer suggestion: The description of joint mechanical performance is lacking in this manuscript. Comparison of loads, fracture surfaces, etc. could be considered. Currently, the optimum ultimate tensile strength of the Cu/Al joints is only 50 MPa, much lower than that of the two substrates. To explain why. Also, is there any improvement strategy?

Answer to reviewer: Thank you for this remark! As mentioned in the article the tensile test samples prepared from the plates welded using powers of 1800 W and 2400 W broke apart immediately during the preloading stage. No representative data regarding their yields strength, ultimate tensile strength, elongation, breaking force or other quantitative data was registered. In the case of the sample prepared using a power of 3000 W the pressure was also too low for the software of the machine to determine any values for the yield strength and elongation, so only the UTS and braking force were registered. A stress-strain curve was not obtained in any of the cases. However, the relationship between the technological conditions, structure and mechanical properties (or lack thereof) were elaborated and further explained in the text. A potential for future improvement was also discussed.   

 

Once again the author team would like to thank the respected reviewer for their comments and excellent suggestions!

Best regards,

Georgi Kotlarski

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revised manuscript of the paper reflects the research purpose well and is acceptable

Author Response

Dear reviewer,

Thank you for your upmost positive review of the submitted paper! You valueable advices helped us improve the quality of our manuscipt significantly, and are much appreciated! 

Best regards,

Georgi Kotlarski 

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have extensively revised the manuscript, reworked the structure of the manuscript, added photographs of the microstructure, and expanded the discussion. 

However, the study is still methodologically incorrect, since it places the primary cause of the influence on the formed joint on the power of the electron beam, rather than the change in the chemical composition of the weld metal caused by different degrees of the edges melting of a dissimilar joint, as well as different thermokinetic conditions for the formation of a weld.

Author Response

Dear reviewer, 

Thank you for your honest review! It is exactly the different thermokinetic conditions that are responsible for the formation of a differing structure in the formed joints. This, we hoped, was reflected clearly enough in the manuscript. The chemical composition is in a direct correlation to the thermal conditions during the welding process. However, the different thermokinetic conditions were generated at the basis of using differing powers of the electron beam welding process. Logically the processes are interconnected. Despite not achieving full penetration of the materials in all cases, the change in the structure of the samples is evident. Indeed, the micorhardness and the tensile properties were not very representative, but we still think the change in the microstructure of the samples, and the observed mechanism of chemical bonding and formation of IMCs may still be valuable data for some specific readers. 

Regardless, we once again thank you for your honest opinion and wish you all the best! 

Best regards, 

Georgi Kotlarski 

Reviewer 3 Report

Comments and Suggestions for Authors

The response to Comment 2 could not be found in the manuscript. If possible, please indicate the specifics of the change.

 

For Comment 3, the authors mentioned that they cannot obtain the stress-strain curves. But, you can compare loads, fracture surfaces, etc. Please provide more details.

Author Response

Dear reviewer, 

Thank you for your time and efforts in preparing this review! According to your suggestions the following responses were generated. 

Reviewer comment: The response to Comment 2 could not be found in the manuscript. If possible, please indicate the specifics of the change.

Answer to reviewer: Please, find the specified test area in Figure 2. The XRD analysis was always performed in the middle of the weld seam. In the text it is hypothesised that the presence of the copper phase is due to the rapid cooling of the melt pool, which resulted in incomplete bonding between the copper and aluminum crystals. As a result unreacted copper particles were introduced in the weld seam. We cannot add the copper phase in the last case because it is not present in the weld seam. This is a result of the higher input power, which resulted in excellent bonding between the copper and aluminum particles, resulting in the formation of the CuAl2 phase, which is proven by the perfromed crystallographic analysis and the indicated texture formation. 

Reviewer comment: For Comment 3, the authors mentioned that they cannot obtain the stress-strain curves. But, you can compare loads, fracture surfaces, etc. Please provide more details.

Answer to reviewer: We are unable to comapre loads (we presume you mean the maximum load the samples could hold before the fracture) since the machine did not measure any data due to the low mechanical strength of the samples (samples 1 and 2). Currently, we are unable also to provide fracture surface SEM images. This would be interesting to observe indeed, however, we are unsure if it is necessary since the first two samples broke immediately. This would not provide much more information regarding the structure. In terms of determining the fracture mechanism it is, as far as I know, incorrect to use the static stress-strain method we used for this purpose. Not to mention that, as previosly stated, even if we decided to do so we could not obtain the stress-strain curves. In summary, we are aware that the harndess values and the tensile properties are not very representative, however, this article focues more on the structural changes that occur during the welding procedure, more spefically the change in the microstructure and the crystallographic properties. 

The author team would like to express their gratitude for your positive review and hope we have managed to imrpove the manuscript in agreement with your comments. 

Best regards, 

Georgi Kotlarski 

Round 3

Reviewer 3 Report

Comments and Suggestions for Authors

The article mainly introduces the formation of electron beam welds between copper and aluminum plates under different power modes (1800 watts, 2400 watts, and 3000 watts), and studies related performance. Low power leads to defects in the weld, high power improves material intermixing and structure, and the sample with a power of 3000 watts has the best performance. After two rounds of revisions, the article has been modified accordingly according to the suggestions of the reviewers, and the overall article has met the publication requirements of the journal.