2017 Volume 58 Issue 12 Pages 1695-1701
In zinc alloy die casting, the thin oxide layer generated on a mold surface causes soldering reactions, and the outer surface becomes coarser as the soldering proceeds. To minimize casting defects such as zinc deposits, we propose a surface modification of the mold to prevent soldering reactions. In this study, we observed the diffusion state of constituents generated between a molten zinc alloy (ZDC1) and the surface-modified mold material (maraging steel) using secondary ion mass spectrometry (SIMS). Furthermore, we analyzed the interfacial reaction layer between the surface-modified mold and the zinc deposits observed on the mold by transmission electron microscopy (TEM) and energy dispersive X-ray spectroscopy (EDS), and considered the effect of surface modification on suppressing soldering reactions.
To improve the quality of castings at the manufacturing sites of die casting, it is necessary to increase the mold temperature to minimize casting defects such as blisters and cold shuts. However, this is considered to be the cause of soldering reactions. Therefore, soldering reactions on the mold surface in die casting are still a severe problem. We observed the interface between a zinc alloy (ZDC1) and the mold material (maraging steel) by TEM and EDS, which allowed us to determine the cause of the soldering reactions between the zinc alloy and the mold material.
A thin film oxide layer of Al and Fe was found to cause adhesion, and it consisted of two thin layers. Fe was found in the lower layer of the thin film oxide layer. On the other hand, Al was observed in the inside of the mold. These results revealed that Al and Fe mutually diffused1). Therefore, it is considered very important to suppress the diffusion of Al from the molten zinc alloy to the mold and to prevent the generation of the thin film oxide layer at the interface as effective countermeasures for suppressing soldering reactions1).
Castings are generally released into the atmosphere. Therefore, as a means of preventing the oxidation of the mold surface, surface modification of the mold is effective.
There are various surface modification methods depending on the purpose, such as plating, thermal spraying, enameling, carburizing, and nitriding. The nitriding treatment method is widely applied because it is capable of improving fatigue resistance and wear resistance, as well as being useful for prolonging the life of various materials since a hardened layer having high hardness and compressive residual stress is obtained2).
Nitriding at 773 to 843 K is commonly used to harden the surface layer, but nitrogen combines with chromium in the base material to form chromium nitride. Chromium nitride reduces the chromium concentration of the base material, resulting in lower corrosion resistance, which is regarded as a drawback. On the other hand, in low-temperature (673–723 K) nitriding, a hard reformed layer called the S phase, which has superior corrosion resistance, is formed in the outermost layer3).
Moveover, it has been reported that this S phase has excellent corrosion resistance4–10). A carburized layer having the same structure as the S phase, which also has excellent corrosion resistance and hardness, has also been found11,12). As a surface treatment method to improve the resistance against high-temperature corrosion and heat resistance, the diffusion penetration treatment method has been proposed1,12).The diffusion penetration treatment generates an alloy layer or an intermetallic compound in the mold surface by diffusion and penetration.
Nitriding treatment is a method with less deformation than heat treatment, and there have been many studies on its application to stainless steel13–18). This is because the resulting surface hardness is higher than those obtained by other heat treatments owing to the generation of a nitrided layer. The plasma nitriding technique, which is a type of nitriding technology, can be used to perform uniform nitriding treatment even on unprocessed materials with complicated shapes19–24).
Meanwhile, surface reforming techniques applying ion implantation, physical vapor deposition (PVD), and chemical vapor deposition (CVD) techniques which have recently been developed in the electronics field are also utilized.
In aluminum die casting, the effect of TiN on preventing soldering reactions has been reported25), but there are few research reports on zinc alloy die casting. In particular, it has been reported that the amorphous carbon film has some excellent properties as a lubricant, such as a low friction coefficient, high hardness, and chemical stability, and can act as a gas barrier against oxygen and nitrogen25–36).
In this study, we compared two types of surface modification, namely, plasma nitriding and amorphous carbon film. These treatment and coating are used to suppress the generation of a thin oxide layer, which is a cause of soldering reactions in zinc alloy die casting1).A noncontact-type surface roughness meter and SIMS were used for the quantitative evaluation of zinc deposits generated by soldering reactions on a mold surface in die casting. Furthermore, we analyzed the interfacial reaction layer between the mold material and the zinc alloy by TEM, EDS, and SIMS, and report its effectiveness for suppressing soldering.
ZDC1 alloy (3.5% to 4.3% Al) and maraging steel were used as test materials. Two types of surface modification were performed for comparison of the surface of the mold, and two materials were prepared. One was a plasma nitriding material, in which we set the surface roughness Sa to approximately 0.3 μm by polishing after plasma nitriding the mold material (hereinafter referred to as a nitriding material). Nitriding layer was approximately 30 μm to 50 μm.
The other was amorphous carbon film (a-C film), in which the nitriding material was further coated with a carbon film having a total thickness of approximately 2 μm after plasma nitriding (hereinafter referred to as an a-C film material). The film was coated by physical vapor deposition (PVD) techniques and it consists of two layers. The upper layer was composed of amorphous with electron configurations consists of sp2 and sp3, and any metal components other than carbon was not intentionally doped. The lower Layer is an intermediate layer.
The chemical compositions of the zinc alloy and maraging steel used in this experiment are shown in Tables 1 and 2, respectively. In this study, a casting test was carried out using a hot-chamber die casting machine.
| Al | Cu | Sn | Pb | Mg | Cd | Zn |
|---|---|---|---|---|---|---|
| 3.5–4.3 | <0.35 | <0.005 | <0.007 | <0.05 | <0.005 | Bal. |
| Si | Mn | P･S | Ni | Co | Mo | Ti | Fe |
|---|---|---|---|---|---|---|---|
| <0.15 | <0.10 | <0.010 | 17.00～ 18.50 |
7.50～ 8.50 |
4.70～ 5.20 |
0.50～ 0.70 |
Bal. |
We observed time sequences of the amount of adhesion and the surface roughness of the zinc deposits induced by soldering reactions as the number of shots increased. The height of the zinc adhesion layer was measured using a non-contact type surface roughness meter. The mold surface before die casting was set as the reference surface, and the difference of heights between the outermost surface after die casting and the reference surface was regarded as the thickness of the zinc adhesion layer. The thickness and roughness of the zinc adhesion layer were repeatedly measured simultaneously at the fixed point.
Figure 1 shows a schematic of the casting product obtained using an actual die casting machine. The temperature of the molten metal in the melting furnace was 703 K and the injection time per shot was set to around 0.01 s. Shots were repeated at intervals of 4.5 s 100,000 times.
Schematic drawing of the castings.
Figure 2 shows the appearances of a fixed die surface and a movable die surface after 100,000 shots of die casting. We can see the flow mark on the fixed surface. However, no soldering can be seen. On the other hand, soldering can be confirmed at points A and B near the gate on the movable die surface. In particular, a large amount of zinc deposits was observed at A.
Microscope image of die surface after die casting. (a) Fixed die side, (b) Movable die side.
The components of the zinc deposit layer at observation points A and B are exactly the same, however the thickness of the deposition layer at point A was greater than that of B.
Therefore, it is relatively easy to measure the time history of the thickness of zinc deposits at A.
On the other hand, since the analysis time in the depth-direction by SIMS depends on the thickness of the zinc deposits, it is better to select point B where the zinc deposits layer was relatively thin as the observation point for the depth-direction analysis.
Then, we measured the thickness of the zinc deposits using a sample obtained at A and analyzed the amorphous oxide layer between the zinc alloy and the mold surface using a sample obtained at B.
Next, to quantitatively evaluate the diffusion of atoms on the mold surface after the die casting test, depth-direction analysis by SIMS was carried out. The constituents for analysis were C, O, Mg, Al, Cr, Fe, Zn, W, and so forth. The measurement conditions were as follows: irradiation region, about 100 × 100 μm; analysis region, around 33 μmφ. TEM observation and EDS analysis of the reaction layer generated in the vicinity of the die surface were carried out. A focused ion beam (FIB) device was used to fabricate the thin film samples for TEM observation.
The thickness of the deposited zinc was evaluated as the amount of zinc deposits near the gate where the soldering was observed. Figure 3 shows the relationship between the thickness of zinc deposits induced by soldering reactions and the shot number in die casting. Figure 4 shows the relationship between the surface roughness of the adhesion layer of zinc and the shot number at the same observation position near the gate.
Time history of height of zinc deposits.
Time history of surface roughness of zinc deposits.
In the case of the nitriding material, it was confirmed that soldering progresses and the amount of zinc deposits increases as the number of casting shots increases as shown in Figs. 3 and 4, respectively. In particular, the amount of zinc deposits increases rapidly until 40,000 shots. It is considered that the growth of zinc deposits is suppressed by the cleaning effect of the molten metal in the die casting process when the number of shots exceeds 40,000. On the other hand, it was confirmed that the thickness of the zinc deposits does not increase with increasing number of shots in the case of the a-C film material. The average roughness on the outermost surface of the zinc deposits was 0.2 μm. This is due to the suppression of soldering.
Figure 5 shows the relationship between the amount of zinc deposits and the surface roughness shown in Figs. 3 and 4. In the case of the nitriding material, a correlation was found between the amount of zinc deposits and the surface roughness of the soldering layer. This cleaning effect has also been reported in aluminum alloy die casting37). After the thickness of the soldering layer reached 4 μm, the surface roughness Sa of zinc deposits did not increase beyond 2 μm owing to the cleaning effect of the molten metal. On the other hand, in the case of the a-C film material, there is no correlation between the thickness of the zinc deposits and the surface roughness of the soldering layer. It was confirmed that the thickness of zinc deposits generated on the mold surface of the a-C film material does not increase owing to the cleaning effect of the molten metal since the adhesion is weak on the die surface of the a-C film material.
Relationship between height and roughness of zinc deposits on die surface.
Figure 6 shows SEM images of the die surface for the nitriding material and the a-C film material after 5,000 and 100,000 shots. We can see that more soldering occurs and the outermost surface becomes coarser as the number of shots increases for the nitriding material, compared with the results for the a-C film material.
Comparison of SEM images of zinc deposits. (a) Nitriding: 5,000 shots, (b) Nitriding: 100,000 shots (c) a-C film: 5,000 shots, (d) a-C film: 100,000 shots.
It is considered that the adhesion of zinc deposits is reflected by the composition of the soldering layer on the mold surface and its crystal structure. Therefore, we investigated the interfacial reaction products generated on the mold surface by performing a die casting test. First, we compared the composition of the materials that diffused into the interface between the zinc deposits and the mold surface in the cases of the nitriding material and a-C film material. Since oxygen (O), aluminum (Al), iron (Fe), and zinc (Zn) are constituents of the thin oxide layer generated at the interface of the molten zinc/mold material, we focused on the increase and decrease in the amounts of these four constituents. Figure 7 shows the SIMS analysis result in the depth direction below the mold surface of the nitriding material. The horizontal axis shows the depth direction and the vertical axis shows the relative secondary ion intensity. (a) shows the result of unused material, (b) shows that after 5,000 shots, and (c) shows that after 100,000 shots. (a), (b), and (c) show the time sequences of diffusion for each constituent.
Results of SIMS analysis of Nitriding material. (a) unused item, (b) 5,000 shots, (c) 100,000 shots.
The thickneses of the soldering layers after 5,000 and 100,000 shots were about 0.1 and 0.8 μm, respectively, and we observed that the outermost surface of the soldering layer was laminated as the number of shots increased.
It can be seen that the zinc soldering layer consists of an oxide of Zn and Al. By focusing on the diffusion phenomena near the interface, it was confirmed that Al was derived from the ZDC1 alloy and Fe originated from the mold as time proceeds. Especially in (c), it was confirmed that O diffused from the outermost surface of the mold to a depth of about 0.8 μm. From the results shown in Fig. 7, if the mold consists of a nitriding material, the interdiffusion of Al and Fe progresses as the oxidation of the mold surface proceeds, and the adhesion between the soldering layer and the outermost surface of the mold is considered to be strong.
The zinc deposits with strong adhesion remain on the outermost surface of the mold. Then, the soldering layer develops and the outermost surface becomes coarser with increasing shot number.
This phenomenon is consistent with the results shown in Figs. 3 to 5.
Figure 8 shows SIMS measurement results of the a-C film material. This material consists of two layers. One is an a-C only layer, which is the upper layer, and the other is an intermediate layer, which is the lower layer. The thickness of the a-C only layer is considered to be around 0.6 μm for Figs. 8(a)–8(c). Forthermore, the thickness of the intermediate layer is consideared to be around 1.8 μm for Figs. 8(a)–8(c). The total film thickness of the a-C layer and the intermediate layer is around 2.4 μm, which does not decrease when the number of shots increases.
Results of SIMS analysis of a-C film material. (a) unused item, (b) 5,000 shots, (c) 100,000 shots.
It was confirmed that the a-C layer does not wear away with increasing number of shots, even if die cast shots are repeated. We can see that the thickness of the zinc soldering layer is maintained at about 0.5 μm when the number of shots increases as shown in Fig. 8. Although the surface of the a-C layer was slightly oxidized, the diffusion of Al and Zn into the a-C layer was hardly observed compared with the result shown in Fig. 7.
Figure 9 shows the values obtained by integrating the intensity from the bottom of the zinc deposits to a certain depth of the reacting layer, based on SIMS measurement results. In detail, these values were obtained by integrating the difference between the SIMS profiles after 5,000 and 100,000 shots and the SIMS profile of the unused material from the bottom of the zinc deposits to a certain depth of the reacting layer based on SIMS measurement results. By comparing the result after 5,000 shots with that after 100,000 shots, it was confirmed that the amount of the oxygen constituent increases as the number of shots increases in the case of the nitriding treatment. On the other hand, no increase in the oxygen constituent was observed in the case of the a-C film material when the number of shots increased. Although there are some differences in the evaluated values for other constituents, similar tendencies were observed.
Results of SIMS analysis at interface of ZDC1/Maraging steel. (a) O, (b) Al, (c) Zn, (d) Fe.
From the above results, we can confirm the following. In the case of the nitriding treatment, the diffusion of four constituents namely O, Al, Zn, and Fe, to the surface layer on the mold proceeds as the number of shots increases. The diffusion of Zn and Al is derived from the molten metal ZDC1 and that of Fe is derived from the mold. That is the mutual diffusion between the constituents of ZDC1 and those of the mold was observed. On the other hand, it was clearly observed that the diffusion of O, Al, Zn, and Fe can be suppressed by the a-C film material. From the above-mentioned findings, we consider that the mechanism of soldering in zinc alloy die casting is as follows. The mechanism of soldering on molds in aluminum die casting has already been considered from physical and chemical viewpoints38–40). There are several reports on chemical combushon mechanisms: (1) the generation of “physical initial adhesion”, (2) the generation of a reaction layer by the adhesion of aluminum alloy and the mutual diffusion between the aluminum alloy and the mold, (3) a “reaction layer growth” stage38), and (4) the rapid growth of the reaction layer caused by the disappearance of the oxide film. It is reported that the oxide film plays an important role in the suppression of soldering39).
Furthermore, Nishi40) reported that a reactive layer composed of an Al-Fe-Si-based intermetallic compound is generated at the interface between the aluminum metal and the mold, and an aluminum alloy adheres to the upper part of the reaction layer. It is reported that the aluminum alloy often adheres firmly to the mold surface in the soldering regions of the mold.
As described above, in aluminum die casting, it is considered that a reaction layer that consists of an interlayer compound is generated at the interface between the mold and the casting. On the other hand, there have been extremely few detailed reports on the soldering reaction of a zinc alloy. From the experimental results descriped in sections 3.1 and 3.2, similar mechanisms are expected for the chemical soldering mechanism of zinc alloy die casting.
In zinc alloy die casting, the mechanism of the growth of soldering in the nitride treatment can be considered as follows: (1) the generation of “physical initial adhesion of zinc alloy”, (2) the generation of a reaction layer by soldering of the zinc alloy and interdiffusion between the zinc alloy and the mold material, and (3) the “growth of a reaction layer” stage. On the other hand, in the a-C film material, although (1) the generation of the “early-stage physical zinc alloy adhesion” occurs, the subsequent (2) interdiffusion between the zinc alloy and the mold material does not occur. Even if zinc deposits are generated in the first stage, their adhesion is weak and this leads to their spontaneous peeling every 30,000 shots. Therefore, it is considered that the generation of zinc deposits is suppressed in the a-C film material.
3.3 TEM observation of soldering layerTo clarify the difference in the interface reaction layer between the nitriding material and the a-C film material, the electron diffraction pattern obtained by TEM observation was analyzed.
From Figs. 10 (a) and 10(b), the interface between zinc deposits and the mold material seemed to separated in the nitriding treatment. However, we can see that the interface of the two regions is not clarified from the plaid stripes obtained under the high-magnification view in Figs. 10 (c) and 10(d).
Diffraction pattern of interface of ZDC1/Nitriding material.
This observation seems to be due to the fact that there are no obvious transition zones between the two regions and that compounds are continuously generated and change. Actually, the electron diffraction pattern for the interfacial reactive layer was unclear, and this means that the interfacial reactive layer consists of compounds. These compounds were indexed as Fe, ZnO, Fe3Zn10, Al2O3, and AlFe3 from the electron diffraction pattern.
On the other hand, the interface between the a-C film material and the mold material, shown in Fig. 11 (c), is clear. In the electron diffraction pattern shown in Fig. 11 (e), amorphous-like halo patterns41,42) were observed, and only the ZnO structure could be indexed. These results show that the interdiffusion of Al and Fe was suppressed. The results obtained by TEM support the results obtained by SIMS shown in Fig. 9.
Diffraction pattern of interface of ZDC1/a-C film material.
The quantification of diffusion by SIMS and the observation of the interfacial layer by TEM were carried out to clarify the cause of the adhesion of the zinc alloy generated by soldering, and the following results were obtained.
The average thickness of the zinc deposit was maintaind within 0.5 μm, and the average roughness on the outermost surface of the zinc deposits was 0.2 μm. This result showed that the a-C film could suppress the growth of the zinc deposits caused by soldering.
