Effect of Thermal Aging on the Interfacial Reaction Behavior and Failure Mechanism of Ni-xCu/Sn Soldering Joints under Shear Loading
Round 1
Reviewer 1 Report
This manuscript studied the interfacial reaction behavior and failure mechanism of 2 Ni-xCu/Sn soldering joints under thermal aging. The manuscript needs thorough corrections in all sections, from abstract to conclusion. The following are specific comments on all sections:
1. Title. Avoid not-sound abbreviation in the title.
2. Several typos, some grammar mistakes and incomplete sentences are found in the text.
3. The research significance displays the gap more than the significance of the research.
4. Style of figures, tables, and references did not match with journal format.
5. On what basis do the authors consider the range or level of each of the studied variables? Would you please elaborate on the manuscript to better understand the readers?
6. The materials' mechanical properties of each material should be provided with a reasonable explanation.
7. The authors should discuss and compare the failure modes for each specimen.
8. It is missing an analysis of the cost.
Minor editing of English language required
Author Response
Point 1: Title. Avoid not-sound abbreviation in the title.
Response 1: Thank you for your suggestion. It is common practice to use abbreviated forms for elements like "Ni," "Cu," and "Sn" in articles and titles. Therefore, the "Ni-xCu/Sn" in the title is appropriately written. We appreciate your understanding in this matter.
Point 2: Several typos, some grammar mistakes and incomplete sentences are found in the text.
Response 2: Thank you for your careful review. We have modified the typos, grammar mistakes and incomplete sentences in the text. Relevant revisions are marked in red in the manuscript.
Point 3: The research significance displays the gap more than the significance of the research.
Response 3: Thank you for your attention to the significance of this study. We have further revised the research significance of this manuscript as "This study provides potential guidance for the composition optimization of micro-soldering substrates in high-power chips", which may be more in line with the significance of the research. The research conducted in this paper addresses a critical issue concerning significant defects observed in Cu/Sn micro-solder joints after aging in microelectronic packaging. Previous studies, including our own research, have indicated that these Cu/Sn system reaction defects can be effectively eliminated by designing the Cu substrate as a Cu-Ni alloy. The primary focus of this paper is to examine the influence of Cu content on the growth of intermetallic compounds (IMCs) at the Ni-xCu/Sn interface, the phase composition, and the mechanical properties of the aged joint. The findings demonstrate that solder joints exhibit improved mechanical properties when the composition of the Cu substrate is designed as Ni-60Cu.
Point 4: Style of figures, tables, and references did not match with journal format.
Response 4: Thank you for your careful review. We have restyled the figures, tables and references in this article to match the journal.
Point 5: On what basis do the authors consider the range or level of each of the studied variables? Would you please elaborate on the manuscript to better understand the readers?
Response 5: Thank you for your attention to this question. The variables examined in this study primarily encompass the Cu content within the Ni-xCu alloys and the aging parameters. The range of Cu content is chosen from 0 to 100 wt.% primarily due to the scarcity of research on the interfacial reaction of the Ni-xCu/Sn system across the entire composition. The selection of a 10 wt.% increment is aimed at obtaining consistent and reliable reaction outcomes. The aging temperature is set at 200 °C, considering that the ambient temperature experienced by solder joints in high-power chips during operation can exceed this threshold. The variation in aging time, specifically 24 h, 48 h, and 72 h, is intended to explore the stability of the reaction products over time.
Point 6: The materials' mechanical properties of each material should be provided with a reasonable explanation.
Response 6: Thanks for your suggestion. We have refined the explanation of the evolution of the mechanical properties of each joint. Actually, the explanation of the mechanical properties of each joint is based on the microscopic topography of the fracture and the displacement-shear force curve. With the increase of Cu content in the Ni-xCu alloys, the decline of joints’ mechanical properties is mainly caused by the decrease of solder brittleness and the existence of particle phase. When the Cu content is 60 wt.%, the mechanical properties are the best. This can be explained from two aspects. On one hand, the IMCs at the Ni-60Cu/Sn interface are single-phase with fine grains; on the other hand, the IMCs have slower growth rate. Compared with the Cu/Sn/Cu joints, the stress value generated during the phase transformation process is reduced. When the Cu content is higher than 70 wt.%, the mechanical properties of the joint is decreased compared with the Cu/Sn/Cu joints. This can be attributed to the formation of full IMC, the presence of numerous shrinkage cavities in the center of the weld, and the occurrence of brittle fracture.
Point 7: The authors should discuss and compare the failure modes for each specimen.
Response 7: Thanks for your suggestion. We have optimized the discussion of failure modes. From the fracture micrographs and displacement-shear force curves of each sample, it can be seen that the fracture morphology and fracture mode of some samples are basically consistent. Therefore, this manuscript focuses on the occurrence and the transition mechanism of failure modes in different intervals.
Point 8: It is missing an analysis of the cost.
Response 8: Thanks for your suggestion. Generally speaking, the Cu substrate of Cu/Sn micro-solder joints is realized by an electroplating process. For the Ni-xCu alloy substrate, it can be prepared by the metal co-deposition method. The cost is mainly due to the fact that the price of Ni is higher than that of Cu.
Point 9: Minor editing of English language required.
Response 9: Thank you for your careful review. We have polished the language of the entire presentation.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Overall the paper is organized properly. Specimen details and test setup are explained adequately. However, there are few issues for which the authors have not provided convincing scientific explanation-
1. Why do Ni-80Cu and Ni-90Cu specimens have full IMC in entire joint, while 100% Cu has small layer of IMC? The specimens do not follow a trend, which is not normal. Were these specimens fabricated incorrectly? Were all 5 samples of Ni-80Cu and Ni-90Cu have full IMC? If yes, can authors provide those images. In case the fabrication process malfunctioned, have authors created new batch of Ni-10Cu and Ni-80Cu and verified if the fabrication process is consistent?
2. Fig. 6 shear strength for 60% Cu seems incorrect and authors need to repeat the experiments. There is no way solder alloys behave so randomly deviating from overall trend. Can authors provide scientific reasoning for such high shear strength for 60% Cu case?
The title of the paper is a bit technically incorrect. It give impression that solder joints fail by aging. On the contrary, solder joints fail under monotonic shear loading. So, for more clarity, the title should be something like- Effect of thermal aging on Interfacial reaction behavior and failure mechanism of Ni... solder joints under shear loading.
Author Response
Point 1: Why do Ni-80Cu and Ni-90Cu specimens have full IMC in entire joint, while 100% Cu has small layer of IMC? The specimens do not follow a trend, which is not normal. Were these specimens fabricated incorrectly? Were all 5 samples of Ni-80Cu and Ni-90Cu have full IMC? If yes, can authors provide those images. In case the fabrication process malfunctioned, have authors created new batch of Ni-10Cu and Ni-80Cu and verified if the fabrication process is consistent?
Response 1: Thank you for your attention to these questions. The Ni-80Cu and Ni-90Cu specimens have full IMC in the entire joint due to the IMCs growing faster at the Ni-80Cu/Sn and Ni-90Cu/Sn interface compared to Cu/Sn interface. Previous studies have reported that the antiphase boundary array of the (Cu,Ni)6Sn5 lattice provided a high-density short-circuit diffusion path for Sn at the Ni-80Cu/Sn and Ni-90Cu/Sn interface. Therefore, these specimens are fabricated correctly. Five groups are prepared for each parameter to obtain the average shear strength value, and the sample close to the average value is selected for morphology observation. The reliability of the experimental data can also be determined from the morphology of the IMCs at the interface of each sample after different aging periods.
Point 2: Fig. 6 shear strength for 60% Cu seems incorrect and authors need to repeat the experiments. There is no way solder alloys behave so randomly deviating from overall trend. Can authors provide scientific reasoning for such high shear strength for 60% Cu case?
Response 2: Thank you for your attention to this question. According to the experimental results, the average shear strength of the joint when the substrate is Ni-60Cu is indeed slightly higher than that of the joint with pure Cu as the substrate. On the one hand, the anomalous phenomenon of shear strength at Cu content of 60 wt.% is explained in the manuscript based on the fracture morphology and fracture curve. When the Cu content is 40 wt.% and 50 wt.%, there are some second particle phases with larger sizes in the joint IMCs, which will promote the fracture initiation during shearing, so that the fracture is mainly located between the solder and the IMCs boundary. This phenomenon can be seen in Fig. 7 observed in. When the Cu content is 70 wt.% and 80 wt.%, the joint quickly forms a full IMC, and there is a large number of shrinkage cavities in the center of the weld, which leads to a decrease in the mechanical properties of the joint and brittle fracture. This phenomenon can be observed in Figures 7 and 8. On the other hand, the increase in joint shear strength when the Cu content is 60 wt.% is not only because the IMCs at the Ni-60Cu/Sn interface are single-phase with fine grains, but also because the IMCs have a slower growth rate compared to Cu/Sn. We have further refined the discussion of this part in the manuscript.
Point 3: The title of the paper is a bit technically incorrect. It give impression that solder joints fail by aging. On the contrary, solder joints fail under monotonic shear loading. So, for more clarity, the title should be something like- Effect of thermal aging on Interfacial reaction behavior and failure mechanism of Ni... solder joints under shear loading.
Response 3: Thanks for your advice. The Ni-xCu/Sn soldering joints in this article do fail under shear loading, so changing the title to " Effect of thermal aging on the interfacial reaction behavior and failure mechanism of Ni-xCu/Sn soldering joints under shear loading " as you suggested, would be more fitting and accurate. We appreciate your valuable input in refining the title to accurately reflect the focus of the study.
Author Response File: Author Response.pdf
Reviewer 3 Report
I olnly miss the electrical characterization of the solder joints: what is the influence of the aging on the contact resistance? (Since these material structures are used in micoelectronical systems, not only the mechanical characterization is impoprtant.)
Author Response
Point 1: I olnly miss the electrical characterization of the solder joints: what is the influence of the aging on the contact resistance? (Since these material structures are used in micoelectronical systems, not only the mechanical characterization is important.)
Response 1: Thank you for your attention to this issue. The electrical characteristics of micro-solder joints also have a great influence on their electrical interconnection and signal transmission. According to current research, Cu/Sn micro-solder joints often form a porous Cu3Sn layer in the late aging period, which will significantly increase the electrical resistance and thermal resistance of the solder joints. Since the lattice constant of Ni is similar to that of Cu, it can form an infinite solid solution with Cu. For the Cu substrate, the electrical properties may not change significantly after Ni is solid-dissolved into Cu to form a Ni-Cu alloy. However, if the reaction product of the Ni-xCu/Sn interface is adjusted to be a single and uniform (Cu,Ni)6Sn5 phase with a slower growth rate, the pores can be reduced. Correspondingly, the electrical characteristics of the micro-solder joints can be further improved.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The paper can be accepted in its current form.
Minor editing of English language required
Reviewer 2 Report
Satisfied with authors response and modifications.
Quality of English language is fine.
Reviewer 3 Report
The authors altered the title of the paper which covers the content much better.
Other improvement were also made.
