© R. Shimaoka et al., Published by EDP Sciences 2025

1 Introduction

In recent years, micro electro mechanical systems (MEMS) have come to be used in wider fields, and at the same time, the demand for micro-metal parts is also increasing. These are manufactured by micro-forming processes which take into account the phenomenon of size effect, examples of which include micro-machining [1], micro-injection molding [2], photolithography [3], and chemical etching [4]. One of these, metal powder injection molding (MIM) is a technology that combines the advantages of powder metallurgy and plastic injection molding, and is excellent in terms of cost-effectiveness in mass production and high material flexibility [5–7]. MIM consists of four steps [8]: (I) Metal powder and a binder (thermoplastic resin) are mixed and granulated into pellets. (II) The mixture is injected into a mould to form the design. (III) The binder is removed by thermal evaporation and/or melting with solvent. (IV) Metal particles are defuse bonded by sintering. During step III, the areas where the binder was present become pores, and the pore size is reduced to the microscale by sintering. Among the MIM, focusing on single millimetre or smaller size production is called micro-MIM or µ-MIM and is applied in various industries but recently highly demanded in the medical field [9]. Products such as medical tweezers and forceps have very thin structures or designs yet must maintain high bending strength, which is an important requirement for medical parts. However, the micropore is unavoidable in MIM process and those micropores, especially existing surfaces, deteriorate the bending strength significantly. To solve this problem, we propose ultrasonic vibration assisted surface hardening process.

Since the Blaha effect was reported in 1955 [10], ultrasonic vibration has been used as an efficient and low-cost process assisting method in multiple metal processing such as welding [11], grinding [12], bending [13], and milling [14]. Furthermore, in the past 10 years, the application of ultrasonic vibration to micro-forming processes has attracted much attention in both academic and industrial fields. The effects of ultrasonic vibration have been reported to include enhancing forming limits [15], improving surface roughness [16], decreasing forming stress [17] reducing friction [18], and surface hardening [19]. Yang Bai et al. [20] revealed that the surface hardness of phosphor bronze C5191 was significantly improved when ultrasonic vibration assisted micro-forging was performed compared to that without ultrasonic vibration, and the hardness increased with enlarging vibration amplitude, up to a maximum of 37%. Recent studies have suggested that the effects of ultrasonic vibration in micro-forming processes are mainly due to two types of effects: acoustic softening and impact effect. Acoustic softening is a phenomenon in which vibration energy is transmitted to dislocations inside the material, decreasing the lattice resistance between dislocations and thereby reducing the flow stress [21]. And the impact effect, as stated by Jun Hu et al. [22], is a phenomenon that occurs when the amplitude is large enough that the vibrating punch periodically leaves from the surface of material, and the plastic strain on the surface increases significantly due to the vibration impact and local frictional heating caused by microslip between material surface and the punch. In the ultrasonic vibration assisted surface hardening, the product surface is struck multiple times with a vibrating punch, with the aim of generating an impact effect that increases plastic strain and work hardens the product.

The ultrasonic vibration assisted surface hardening is characterised by the fact that it is possible to locally strengthen the product by changing the size and shape of the punch, so that it can be applied to products with complex shape. In addition, this process can be performed at room temperature in a short time and does not require harmful chemicals and expensive, large-scale equipment, making it more economical and environmentally friendly than other surface strengthening methods.

In addition, porous metals have unique properties such as sound absorption, shock absorption, filtration, and thermal conductivity due to their internal structure [23,24], and we assumed that the sound and shock absorption properties are also effective in the ultrasonic vibration. According to Nishimaki S. [25], the mechanism of sound absorption is that the sound is consumed and attenuated by viscous resistance and thermal conduction when it propagates through a large number of pores. Also, Makoto K. [26] explains that the shock absorption is that when the porous metals are crushed, the cell walls bend and buckle, causing the entire cell to deform, and shock is absorbed by the existence of a deformation region called a plateau where strain increases at a certain compressive stress. Therefore, porous metals are expected to be able to efficiently absorb ultrasonic vibration energy and provoke a more effective impact effect and surface strengthening than materials without pores. There have been few studies in the past which have used ultrasonic vibration assisted processing on porous metals in microscales, so it can be said that this research is of academic interest as examines the effect of ultrasonic vibrations on porous structures and is expected to reveal new and unique phenomena.

In this study, we verified the effects of the ultrasonic vibration assisted surface hardening, the influence of the porous structure, and the surface strengthening mechanism through the following two types of experiments.

Experiment I: Verification of the effect of ultrasonic vibration assisted surface hardening on porous structure.

Ultrasonic vibration assisted surface hardening was performed on MIM processed pure copper specimens and rolled specimens without a porous structure. After hardening, the effects were investigated by Vickers surface hardness testing, cross-sectional electron backscatter diffraction analysis (EBSD), and surface roughness measurement.

Experiment II: Verification of the influence of surface roughness and surface strengthening mechanism.

To eliminate the effect of surface roughness on hardening, two types of MIM processed stainless steel 316L specimens (one is electropolished and the other is untreated) were used, and surface strengthening was performed on each. After hardening, the influence was investigated by cross-sectional EBSD, and the mechanism of surface strengthening was verified.

2 Materials and methods

2.1 Material

2.1.1 Experiment I

In experiment I, commercially available rolled pure copper specimens and MIM processed pure copper specimens that intentionally increase the number of pores provided by Taisei Kogyo Co., Ltd. were used. The specifications of each specimen are shown in Table 1.

2.1.2 Experiment II

In experiment II, with/without surface-treated MIM processed stainless steel 316L specimens, that intentionally increase the number of pores provided by Taisei Kogyo Co., Ltd., were used. Electropolishing was deployed as a surface treatment in this work. The specifications of each specimen and average surface roughness measured by AFM are shown in Table 2.

Figure 1 shows cross-sectional SEM images of a pure copper MIM specimen and a SUS316L MIM specimen taken at 450× magnification.

Fig. 1 450× cross-sectional SEM images of pure copper MIM specimen and SUS316L MIM specimen.

2.2 Experimental method

The ultrasonic vibration assisted surface hardening system is shown in Figure 2. The system was incorporated inside a small tabletop servo press designed by Micromachining Laboratory LLC, and the minimum step movement is 1 µm.

As shown in Figure 2b, the system consists of an upper die (including an ultrasonic generator and a horn connected to it, a stress load cell, and a punch) and a lower die (including a specimen fixing plate and a dynamic load cell). The ultrasonic generator generates vibrations of 60 kHz, which are transmitted to the horn, which resonates and amplifies the amplitude before transmitting the vibrations to the punch. The horn is specially designed to have a natural frequency of 60.5264 kHz, and the punch vibrates in the range of 60 ± 2 kHz by resonance. The maximum amplitude of the punch tip when unloaded was 3.21 µm as measured by a laser displacement metre. The punch is made of tool steel (SKD11) and has a diameter of 5 mm.

Figure 3 shows the plates used to fix specimens. To prevent the specimen from warping and shifting during the process, the upper plate, which has a through hole for the punch in the centre, can be firmly locked to the lower plate by its four corner bolts.

The contact state between the punch and the specimen can be measured by a stress load cell (Nihon Tokushu Sokki, LCS-H-30 kN) on the top of the punch. The load cell is fixed to the punch with a preload of about 300 N to improve response. Additionally, a dynamic load cell (Kistler, 9132B) is installed under the specimen fixing plate, and the vibration waveform when ultrasonic vibration is applied can be obtained. When the punch leaves from the specimen and the impact effect occurs, the acquired waveform is distorted from a sine wave, therefore the occurrence of the impact effect can be confirmed [22].

The experiment was performed in the following procedure: (I) Fix the specimen. (II) Lower the punch and apply a preload to prevent it from shifting during the process. (III) Apply ultrasonic vibration. As a lubricant, graphite spray was applied to the surfaces of the specimen and punch. Table 3 shows the ultrasonic application conditions and Table 4 shows the preload values used for each specimen.

In Experiments I and II, due to time constraints, it was not possible to completely match the thickness, physical properties, and materials of the specimens. However, the purpose of this study is to qualitatively verify the effects and mechanisms of ultrasonic vibration assisted surface hardening of porous metals, and it was determined that this would not interfere with the research for the following reasons.

Fig. 2 Overview of the ultrasonic vibration assisted surface hardening system. (a)External appearance, (b) Schematic diagram of the system [27].

Fig. 3 Specimen fixing plate [27].

2.2.1 Difference in thickness between rolled and MIM specimens in Experiment I

According to J. Hu et al. [28], the impact effect occurs when the maximum strain caused by the peak-to-peak amplitude in each period of vibration exceeds the yield strain of the material. Therefore, by adjusting the preload and aligning the strain so that the initial elastic deformation and springback amounts are approximately the same, the punch separation displacement during process is set to be the same, minimizing the effects of differences in specimens' thickness and physical properties. Previous research [29] has revealed that the impact effect occurs when a preload of 30 N is applied to the same rolled pure copper specimen, and the strain was calculated by determining the deformation amount of the specimen under preload pressure from the displacement of the punch, and the preload of the MIM specimen that would result in the same strain was calculated and adopted. In addition, in this experimental device, a standing wave is stably generated by the horn, and since the preload is small compared to the inertial force of the punch, it is expected that the change in the amplitude of the ultrasonic vibration when the preload change can be neglected.

2.2.2 Change of the specimen material to SUS316L from experiment I to II

The effect of the surface roughness on the deformation form of the surface layer of the specimen is due to the size effect caused by the shape of the surface asperities and the bulk part. If the surface asperities and the bulk part are made of the same material, it is expected that it does not depend on the material strength, so this time we used SUS316L material, which is a different material from the experiment I. Additionally, according to Y. Bai et al. [30], it has been reported that the effect of ultrasonic vibration is largely due to the movement of dislocations, so the higher the strength of the specimen material and the less likely the dislocations are to move, the smaller the effect. Subsequent research has revealed that the larger the surface area/total grain ratio, that is, the thinner the material is while maintaining its surface area, the larger the proportion of surface grains, which have fewer dislocation entanglements and are easier for the slip system to move through than internal grains, and the stronger the effect of ultrasonic vibration is [31]. Based on these results, in order to more easily obtain the effect of ultrasonic vibration on SUS316L specimens, which are harder than pure copper, the thickness was made thinner.

2.3 Evaluation method

The surface hardness of the specimens after hardening was measured using a micro-Vickers hardness tester (Shimadzu Corporation, HMV-G31ST-HC). The surface was not polished before the test in order to avoid the layer strengthened by ultrasonic vibration being scraped off. In the test, five data points were obtained from the hardened surface area and the surrounding unprocessed area, and the average value of the three data points excluding the maximum and minimum values was used. The test was performed at six test forces: HV0.002 (19.61 mN), HV0.005 (49.03 mN), HV0.01 (98.07 mN), HV0.025 (245.2 mN), HV0.05 (490.3 mN), HV0.1 (980.7 mN).

The changes in the internal structure of the specimens after hardening were observed by cross-sectional SEM and EBSD. Figure 4 shows an overview of the observation sections. To compare the effects of surface strengthening, we obtained data for two sections, the hardened section and the unprocessed section, and the inverse pole figure (IPF) map and the kernel average misorientation (KAM) map were obtained. The IPF map shows the three-dimensional crystal orientation information as a two-dimensional map according to a colour key. And the KAM map shows the difference in crystal orientation between adjacent measurement points, and there is a quantitative correspondence with the plastic strain caused by dislocation motion [32].

The surface roughness of the specimens was measured by AFM (KEYENCEN, VN-8010).

Fig. 4 Schematic illustration of EBSD sections.

3 Result and discussion

3.1 Result of experiment I

3.1.1 Changes in surface hardness

Figure 5 shows the relationship between the indentation depth and the surface hardness of the hardened and unprocessed areas of the rolled and MIM specimens obtained from the Vickers surface hardness test.

Figure 5a indicates that there is no obvious difference in surface hardness between the hardened and unprocessed areas of the rolled specimen, whereas Figure 5b shows that the surface hardness of the MIM specimen apparently increases after hardening. By weakening the test force and reducing the indentation depth, it is possible to investigate the hardness near the surface, but an indentation depth of 2 µm or less results in large errors due to the influence of surface roughness and internal pores. Excluding the two points with the smallest indentation depths in Figure 5b where the error is large, it can be seen that the rate of increase in surface hardness improves as the indentation depth decreases and approaches the surface, with a maximum increase of 43.3%.

Fig. 5 Relationship between indentation depth and surface hardness of hardened and unprocessed areas of pure copper MIM specimen and rolled specimen.

3.1.2 Changes in internal structure

Figure 6 shows the IPF and KAM map of the hardened section and unprocessed section of the rolled specimen obtained by cross-sectional EBSD. Figure 7 also shows the IPF and KAM map of the MIM specimen.

In the case of the rolled specimen, a comparison of the IPF maps in Figures 6a and 6b indicates that after hardening, the grain boundaries near the surface become unclear and refinement is progressing. Furthermore, a comparison of the KAM maps shows that strain increases significantly within about 10 µm of surface after hardening. On the other hand, in the case of the MIM specimen, a comparison of the IPF maps in Figures 7a and 7b indicated that refinement of grains was not confirmed. Furthermore, a comparison of the KAM maps shows that after hardening, strain increases significantly within about 3 µm of surface.

Fig. 6 IPF and KAM maps of hardened and unprocessed sections of pure copper rolled specimen obtained by cross-sectional EBSD.

Fig. 7 IPF and KAM maps of hardened and unprocessed sections of pure copper MIM specimen obtained by cross-sectional EBSD.

3.1.3 Changes in surface roughness

Table 5 shows the change in average surface roughness before and after hardening obtained by AFM.

From Table 5, it can be seen that the average surface roughness of the rolled specimens is decreased by 35.3% after hardening, and the average surface roughness decreasing rate of the MIM specimens reaches 57.9%.

3.1.4 Discussion

As for the effect of surface strengthening on the rolled specimen, Figure 6 shows that after hardening, strain increased about 10 µm from the surface, and grain refinement progressed in the same area. This is thought to be because ultrasonic vibration was applied to the material during process, and the vibration energy increased strain and formed new grain boundaries. This phenomenon indicates the occurrence of the impact effect. However, when compared with Figure 7, it can be seen that the increased strain is distributed over a wider range than in the results for the MIM specimen. In contrast, as for the effect of surface strengthening on the MIM specimen, Figure 7 shows that strain increased significantly within the surface approximately 3 µm after hardening, but the grain refinement was not confirmed. This is presumably because the strain increased by surface strengthening was concentrated only near the surface and did not cause grain refinement in other regions. In addition, since the increase in strain is clearly denser than the results for the rolled specimen shown in Figure 6, it is supposed that this concentrated strain causes work hardening, which is why the increase in surface hardness was observed, as shown in Figure 5b and the rate of increase in surface hardness improved closer to the surface. On the other hand, since the strain is dispersed in the rolled specimen, the increase in surface hardness was not as significant as in the MIM specimen.

As for the reason why the increase in strain in the MIM specimen was concentrated on the surface within a few micrometres, unlike the rolled specimen, one possible reason is the effect of plastic strain when the surface irregularities are flattened, since the change in surface roughness due to hardening is greater in the MIM specimen than in the rolled specimen, as shown in Table 5. However, since the surface roughness is small compared to the depth of the range where strain increased, it remains to be investigated whether this is solely due to the effect of differences in surface roughness. To investigate the cause of this phenomenon in more detail, surface strengthening was carried out on two types of MIM specimens with different surface roughness in experiment II.

3.2 Result of experiment II

3.2.1 Effect of surface roughness on surface strengthening

Figure 8 shows the IPF and KAM map of the hardened section and unprocessed section of the surface electropolished MIM specimen obtained by cross-sectional EBSD. Figure 9 also shows the IPF and KAM map of the untreated MIM specimen.

In the case of the electropolished MIM specimen, a comparison of the IPF maps in Figures 8a and 8b indicates that there is no clear change in the grain size. In addition, a comparison of the KAM maps shows that after hardening, strain increases at about 2 µm from the material surface. Similarly, in the case of the untreated MIM specimen, a comparison of the IPF maps in Figures 9a and 9b indicates that there is no clear change in the grain size, and a comparison of the KAM maps shows that strain increases at about 1 µm from the surface.

Fig. 8 IPF and KAM maps of the hardened and unprocessed sections of the SUS316L surface electropolished MIM specimen obtained by cross-sectional EBSD.

Fig. 9 IPF and KAM maps of the hardened and unprocessed sections of the SUS316L untreated MIM specimen obtained by cross-sectional EBSD.

3.2.2 Effect of porous structure on surface strengthening

Figure 10 shows a comparison of the KAM maps of the hardened and unprocessed sections from the surface to a depth of 90 µm for electropolished and untreated MIM specimens.

Comparing the pores in the electropolished and untreated MIM specimens with the untreated and hardened sections, it can be seen that both specimens show increased strain around the pore walls after hardening, especially at the top (surface side of the specimens).

Fig. 10 Comparison of KAM maps of the hardened and unprocessed sections from the surface to a depth of 90 µm on electropolished and untreated SUS316L MIM specimens.

3.2.3 Discussion

As for the effect of surface strengthening on the MIM specimens with different surface roughness, no clear change in grain size was observed after hardening, and an increase in strain was concentrated within a few micrometres on the surface. In addition, the electropolished MIM specimen with a smaller surface roughness had a deeper range of increased strain than the untreated MIM specimen. From these results, it is presumed that the effect of surface roughness on hardening is that the smaller the roughness, the greater the increase in strain from the surface to a deeper area. This is supposed to be because the smaller the surface roughness, the stronger the adhesion between the punch and the material surface, and the easier it is for the ultrasonic vibration energy to be transmitted into the material. On the other hand, despite the change in the surface roughness of the specimen, the strain increased after hardening concentrated in the surface several micrometres, and no grain refinement was observed inside the material, and the same tendency was confirmed in Figure 7. Therefore, this phenomenon can be said to be a characteristic of MIM specimens. Furthermore, since an increase in strain was confirmed around the pores after hardening in Figure 10, it is supposed that the porous structure was the main factor. The sound and shock absorption properties of porous metals, including MIM products, are thought to be achieved by converting sound into thermal energy through friction with the pore walls, or by the pores collapsing. Based on this theory, it is supposed that the ultrasonic vibration energy was converted into thermal energy as it was transmitted through the pore walls, activating the movement of dislocations, and/or that the pores underwent slight plastic deformation due to the impact from the punch, increasing strain around the pore walls. In this way, the vibration energy transmitted inside the material was absorbed by the pores, and did not cause an increase in strain over a depth of several tens of micrometres and refinement of crystal grains, as confirmed by the results of the rolled specimen in Figure 6. On the other hand, within a few micrometres from the surface, the impact effect activated dislocations and increased strain locally, which is the cause of the phenomenon that appeared when the MIM specimens were surface-hardened in Experiments I and II, and is the mechanism of surface strengthening.

4 Conclusion

In this study, we verified the influence of the porous structure and the mechanism of surface strengthening when ultrasonic vibration assisted surface hardening was performed on porous metals. In experiment I, surface strengthening was performed on rolled pure copper specimens and MIM processed pure copper specimens with a porous structure, and they were analysed through Vickers hardness tests, cross-sectional EBSD, and AFM. In experiment II, surface strengthening was performed on two types of MIM processed SUS316L specimens with different surface roughness, and they were analysed through cross-sectional EBSD.

As a result, in experiment I, the surface hardness of the MIM specimens increased by up to 43.3%, and the average surface roughness decreased by 57.9%. It was also observed that the strain was concentrated within a few micrometres of the material surface. On the other hand, there was no obvious increase in surface hardness for the rolled specimen, and the increase in strain was observed to be distributed over a wider range than for the MIM specimen. In experiment II, regardless of the surface roughness, the increase in strain was concentrated within a few micrometres of the surface for both MIM specimens, and an increase in strain was also observed near the walls of the internal pores after hardening. From these results, it was found that when surface strengthening is performed on porous metals, the increase in strain is more concentrated within a few micrometres of the surface compared to metals without a porous structure. This can be inferred to be because the ultrasonic vibration energy transmitted inside the material is converted into heat in the internal pores and/or absorbed by deformation of the pores, and strain increases only on the very surface due to the impact from the punch. The strain concentrated on the surface caused work hardening, increasing the surface hardness of the MIM specimen, whereas the increase in hardness was not apparent in the rolled specimen, in which the strain was distributed over a wider range.

This study demonstrated that the ultrasonic vibration assisted surface hardening is a method that can effectively strengthen the surface of porous metals locally in a short time without the use of toxic chemicals or large-scale equipment. This is expected to be a method to solve the problem of insufficient bending rigidity due to internal pores in µ-MIM products used in the medical field. In addition, the newly discovered phenomenon in which porous metals absorb ultrasonic vibrations and only a few micrometres on the surface becomes work-hardened is expected to be used as a method to improve only the bending rigidity of materials without affecting their internal ductility or crystal structure. In future research, we plan to investigate the effects of porosity, pore size, and material to quantitatively clarify the mechanism of surface strengthening in order to advance practical application.

Acknowledgments

The authors would thank Taisei Kogyo Co. for providing the MIM processed specimens and electropolishing the stainless steel 316L specimens.

Funding

None.

Conflicts of interest

The authors declare no conflict of interest.

Data availability statement

Not applicable.

Author contribution statement

Ryo Shimaoka; Conceptualization, Methodology and Writing—original draft preparation, Zidong Yin; Resources and Review & Editing, Shigeo Tanaka; Preparing specimens and Review & Editing, Ming Yang; Supervision and Review & Editing. All authors have read and agreed to the published version of the manuscript.

References

All Tables

All Figures

	Fig. 1 450× cross-sectional SEM images of pure copper MIM specimen and SUS316L MIM specimen.
In the text

	Fig. 2 Overview of the ultrasonic vibration assisted surface hardening system. (a)External appearance, (b) Schematic diagram of the system [27].
In the text

	Fig. 3 Specimen fixing plate [27].
In the text

	Fig. 4 Schematic illustration of EBSD sections.
In the text

	Fig. 5 Relationship between indentation depth and surface hardness of hardened and unprocessed areas of pure copper MIM specimen and rolled specimen.
In the text

	Fig. 6 IPF and KAM maps of hardened and unprocessed sections of pure copper rolled specimen obtained by cross-sectional EBSD.
In the text

	Fig. 7 IPF and KAM maps of hardened and unprocessed sections of pure copper MIM specimen obtained by cross-sectional EBSD.
In the text

	Fig. 8 IPF and KAM maps of the hardened and unprocessed sections of the SUS316L surface electropolished MIM specimen obtained by cross-sectional EBSD.
In the text

	Fig. 9 IPF and KAM maps of the hardened and unprocessed sections of the SUS316L untreated MIM specimen obtained by cross-sectional EBSD.
In the text

	Fig. 10 Comparison of KAM maps of the hardened and unprocessed sections from the surface to a depth of 90 µm on electropolished and untreated SUS316L MIM specimens.
In the text