Review Reports

Point-by-point response to Comments and Suggestions for Authors

Comments 1: Both should be improved in terms of content as well as in terms of terminology. It is unusual to refer to the bead as seam, since it is mostly appeared in welding processes.

Response 1: Title is slightly changed and abstract is rewritten. After the discussion we have decided to use term “bead” in this work. Seam is more general term, and seam could consist of multiple beads.

Comments 2: Moreover, the authors do not state the used metal AM processes in which they refer. It is very important to have this information, since the process mechanism changes considerably between the metal AM processes. Therefore, it is critical to mention the process mechanism, the type of feedstock and the challenges that are faced in this process compared to the rest metal AM.

Response 2: This work is dedicated to better understanding of modification of heat transfer in melt pool. TEMC can be present in many different AM methods where melt pool is forming and differential Seebeck coefficient between solid and liquid is sufficient. Our guess is that it can be more applicable for the wire laser additive manufacturing. It is clarified in the text.  

Comments 3: The authors should also refer to the applicability of the proposed solution in different industrial requirements and common process-related and mechanism-induced issues.

Response 3: It is mentioned in the text that application of magnetic fields can be a way to control the heat transfer in the melt pool thus controlling the temperature and cooling rate. In general, application of magnetic fields during AM is promising research topic with many new articles. Wider applications in industry is still developing.

Comments 4: Moreover, why did the authors select to study this specific material? Is there any reason behind it?

Response 4: This Sn-10%wt.Pb alloy is chosen because it is commonly used model material of two component alloy. This alloy is a perfect two component alloy with traditional phase diagram. It is easy to etch it and there are many scientific papers about the grain development of this alloy in directional solidification and with electromagnetic fields It’s properties are well known, and Seebeck coefficient has been measured in high temperatures in solid and liquid states, which has been done only for wery small number of alloys and metals.

Comments 5: The authors state in line 33 that “The most common…. Powder [4]”. It is suggested to improve the used wording related to the terminology, since especially for metal AM processes, it is not often to be called 3D printing while wire and powder are feedstock types that can be met with different heat input methods and configurations. The authors may use the following sources to strengthen the arguments about the type of metal AM processes.

• Stavropoulos, P., & Foteinopoulos, P. (2018). Modelling of additive manufacturing processes: a review and classification. Manufacturing Review, 5, 2.

Response 5: The text has been clarified and terminology is now consistent through the text. We use “Additive manufacturing” or AM, which is more general term. In some sources 3D printing is used.  We are familiar with this review article and the differences and similarities between various AM methods.

Comments 6: Moreover, the authors mention “small amount of metal”. It is too generic and not suitable for scientific documents. Please be more specific in accordance with the used heat input and the type of feedstock.

Response 6: This is rephrased and clarified in the text. For this work it is important to highlight that small liquid meltpool is forming. It can be achieved in several ways (laser and electric arc are the most common ones).

Comments 7: “Because of these problems only certain metal alloys are suitable for additive manufacturing, while many commonly used alloys remain only machinable and castable”. The authors should clarify this statement, since there are significant efforts worldwide to adopt more functional materials, by introducing pre-heating and more focused heat input. “There are numerous different metal additive manufacturing methods, …process outcome”. The authors should describe with few words these processes, emphasizing on their main process mechanism”. They may consider consulting the following work.

• Biserova-Tahchieva, A. et al Additive Manufacturing Processes in Selected Corrosion Resistant Materials: A State of Knowledge Review. Materials2023, 16, 1893. https://doi.org/10.3390/ma16051893

Response 7: Yes, the methods and of AM develop rapidly and new approaches to use common or new materials appear. But there are still plenty of materials for which problems remain. This is good overview article describing some of the problems and solutions. We included it in the references.

Comments 8: At the end of this section, the authors should add the key points of this work, focusing on the used process, on the challenges as well as the expected output. They should also add the outline/structure of this work.

A: Article has been restructured and link between sections is improved.

Comments 9: It is advised to include the designed experimental work, the workflow that has been followed, the method which was used to extract the determination of the working parameters as well as the background for selecting the investigated geometry during these tests.

Response 9: Parameters are selected to demonstrate magnetic field and TEMC effects on low temperature alloys in regime close to actual AM. Text has been improved.

Comments 10: Until now, the connection of the used method for controlled microstructure is not clear how it is linked to metal AM, since the latter they assume addition of material rather than only remelting-pre-heating.

Response 10: Link to AM is described in the text now. Potential to use magnetic fields for bead size control and to compensate for the anisotropic heat flux. This can be valid for various AM methods. This article is mainly focused on melt pool flow analysis. We have tried this with wire feeder and the results are similar, but the wire disturbs the melt pool too much and the process in unstable with our equipment.

Comments 11: However, it not evident what is the useful output for industrial use from this specific workflow and the applicability of this solution in actual cases. Can the authors comment on that?

Response 11: This research is done to determine TEMC significance on the melt pool flow and how that affects the bead aspect ration and size. Application in AM can be magnetic field application during AM.

Comments 12: Can the authors explain why they have divided this section in three parts, introducing the development of a numerical tool (Comsol) for process simulation? Are they envisioning to develop reduced model based on the combination of experimental and simulation results? Please discuss.

Response 12: Experimental results, numerical results and analysis giving the analytical estimate and comparing the different approaches. Chapter is restructured and improved. In this fork fixed weld pool numerical model only serves as a tool indicating the flow velocity and flow pattern. Actual modeling of heat transfer caused by the bead increase is a complex modelling task.

Comments 13: This section includes a clear representation of the findings of the present work, but without clear focus on the added value of these outputs in the industrial status as well as direction for future work.

Response 13: Conclusions are improved by adding the indications for the future work and potential application of the results in metal AM under magnetic fields.

Point-by-point response to Comments and Suggestions for Authors

Comments 1: The value for the range in which the cooling rate must be 104–10610^4–10^6104–106 K/s is given once in the introduction, and a reference needs to be included adjacent to this value, indicating whether it’s for your laser settings (250 W, 1mm beam, 2-4mm/s) or generally for AM.

Response 1: In the introduction this broad range is given in general for various AM methods. For given work numerical model Fig. 6 shows precise thermal field from which thermal gradient magnitude and distribution can be obtained.

Comments 2: You provide information about a sample size with dimensions 10×30×70mm for Sn-10%wt.Pb samples, and it would be good to clarify whether these dimensions are with regard to welding direction, indicating the number of seams for each sample block to detect any edge effects.

Response 2: We did scanning along the longest edge of the sample. Each sample contained two seams 10 mm apart and 10 mm from edges. Bead size is typically 1-2 mm, thus we believe edge effects are minimal in this case. They were not observed in micrographs.

Comments 3: You could say in describing the laser, whereas in your welding, you are able to weld thicknesses up to 4mm, whereas your melt pool distances are about 0.8-0.9mm, whether you intentionally stayed below weld thicknesses closer to 4mm, given your power output decision to maintain at 250W.

Response 3: Mentioned 4 mm is max possible material which can be welded with given laser. For this work we were not aiming to go to extremely high temperatures and large melt pool. Power of 250 W were chosen because it cerates bead with aspect ratio close to 1 and does not heat material to extremely high temperatures.

Comments 4: The values for electrical conductivity given in Table 1 for solid: 4·10⁶ S/m, for liquid: 2·10⁶ S/m, for density: 7840 kg/m³, and for density again: 7480 kg/m³ are not accompanied by uncertainties or references, and it is requested to state whether these are mean values for the melting point or, in reference to [30], to add an explanation for their dependence on composition or temperatures.

Response 4: These are values used for this alloy in different other works or taken from general handbooks. Properties of Sn-10%wt.Pb are generally well known and in this article it is not too critical to use very accurate properties or to take into account temperature dependance.

Comments 5: The range given for beam diameter, 0.5–2mm, coupled with their chosen value of 1mm, could have been correlated better with their results for seam width, 1.58mm for no magnetic field, and 1.92mm for 0.2T; initially, there would seem to be some lateral conduction or perhaps defocusing in as much as their seams are wider than their beams.

Response 5: In this work we wanted to keep beads not too large. Width of 1 mm was chosen as optimal because it allowed to do the planned bead morphology analysis. Molten zone is defined by the heat transfer and melting point of the material. Bead size can be larger for low melting temperature materials. Narrower beam would cause smaller and hotter laser spot, but bead is mainly defined by total power input.

Comments 6: The comparison between the sizes of seams in width and depth (1.58mm/0.79mm Vs 1.92mm/0.91mm) could have been strengthened if you had provided figures for the uncertainty in measurement, for example, 0.02mm from the microscope, to enable an evaluation of its significance in regard to changes observed.

Response 6: It was measured with microscope. Repeatability of bead size measurements was good, so we believe the uncertainty is no larger than few percent, which does not significantly affects the effect we want to show with this paper or the conclusions.

Comments 7: Depth and width, you say, grow by "about 60% between 0 and 0.4 T” in depth and width, while the volume grows by "3–4 times” for the melt pool, although, strictly speaking, for 60% growth in depth and width, growth in volume with one dimension varying would be different, presumably either the actual cross-section areas could be presented, or there could be clarification on what exact growth in depth, width, and presumably in thickness, would correspond to growth factor of 3-4 times.

Response 7: It is assumed, that is the same aspect ratio is maintained than 1.6^3=4. Indeed comparing different cases in Fig 3. we see that cross section is 2-3x larger by area with magnetic field. Growth dependance vs magnetic field strength shown in Fig.4. shows similar growth in width and depth. We suspect for larger bead it may differ more.

Comments 8: The model’s highest value for thermoelectric current density, 1 A/mm², with an average value of 0.3 A/mm², matches expectations from an estimated value given in a thermal gradient measure of 500 K/mm; it might be useful to clarify the derivation for 500 K/mm (whether simulation or literature value, for example).

Response 8: Thermoelectric current is calculated and shown in Fig.6. Maximum current is in places where T gradient along the solid/liquid interface is maximum.

Comments 9: The value for maximum flow velocity with B = 0.2 T, which is 12 cm/s, is a significant finding, and I would like to suggest including a small table with flow velocity versus magnetic field strengths from 0 to 0.4 T, given your finding that flow velocity rises "almost proportionally" in this region.

Response 9: Parametric numerical model was done in our recent paper in JOM (Ref.27). There model results is normalized and plotted versus analythical estimate Fig.8. Agreement was good. In this work it was not done because aim was to quantify the bead sizes, but we calculated for 0.1, 0.2 and 0.4 and results agree well with expected values.

Comments 10: The value in Fig. 7 for maximal velocity is about 1 T, whereas you believe that with 0.4 T you are "quite close to maximal velocity" whereas 0.5 T is enough to achieve near maximal velocity; specifically, it would have been useful to report approximately what velocity could be expected at 0.4 T, 0.5 T, 0.8 T, and 1 T to have some idea just how close "quite close" really is.

Response 10: Velocity dependance on applied magnetic field is given in Fig.8. Wording is rephrased. What we wanted to say is that at the beginning velocity increases rapidly with field and 0.1- 0.4 T was sufficient to observe the impact of the field on the bead.

Point-by-point response to Comments and Suggestions for Authors