1 Introduction

Co–20Cr–15W–10Ni (CCWN) (mass pct, L605) alloys, registered as ASTM F90, are widely applicable to almost all balloon-expandable stent platforms owing to their excellent mechanical properties and corrosion resistance.[1,2,3] In particular, balloon-expandable stents require low yield stresses for expansion using a balloon with a low stress.[4] Recently, significant efforts have been devoted toward reducing the diameters of balloon-expandable stents to improve minimally invasive medical treatments.[5,6] However, to allow balloon-expandable stents to have a thinner wall design, the alloys used must demonstrate high strengths and high ductilities while maintaining low yield stresses.

In this regard, an effective approach to increase the strength of Co alloys involves the addition of interstitial elements,[7,8,9,10,11,12,13] including various light elements, such as nitrogen and boron.[14,15,16,17,18] However, carbon, known to be a face-centered cubic (fcc) stabilizing element,[19] is the most explored option owing to its higher solubility (1 mass pct) in Co matrices compared with other interstitial elements.[20]

For CCWN alloys, the registered carbon content in ASTM F90 ranges from 0.05 to 0.15 mass pct.[21] Several studies have investigated the precipitates of CCWN alloys with carbon contents ranging from 0.07 and 0.11 mass pct.[22,23,24,25,26] In particular, a Laves phase, such as Co2W, was observed in a CCWN alloy with a carbon content of 0.07 mass pct.[22] Further, α-Co3W lathes were detected in a CCWN alloy with 0.11 mass pct of carbon after annealing at 1073 K for 6 months.[23] Our group previously reported that the main phases of precipitates in CCWN alloys with 0.08 mass pct of carbon are Cr-rich M23C6 and W-rich η-phase (M6C–M12C type, M: metallic element) in the temperature range of 973 K to 1448 K.[24,25] Zhu and Chen detected the M23C6, M7C3, and η-phase after heat treating a CCWN alloy with 0.07 mass pct of carbon at 873 K–1273 K up to 86.4 ks.[26]

Notably, the addition of carbon to CCWN alloys is closely related to their mechanical properties, as carbide precipitation can increase their strength and suppress grain growth through pinning.[27] Here, solute carbon stabilizes the γ-phase with an fcc structure, which improves ductility.[28] Moreover, according to recent reports,[7,29] the effect of carbon addition on the microstructure and mechanical properties of high-entropy alloys (HEAs) containing Co and Cr is observed in the carbon content range of 0 to 0.40 mass pct. The addition of carbon to HEAs provides substantial strength through precipitation strengthening[9,12,30] and grain refinement without sacrificing ductility,[7,29,31,32] which can be achieved via conventional thermomechanical processing methods that combine cold working and annealing.

However, previous reports on the microstructures and mechanical properties of CCWN alloys have been limited to carbon contents ranging from 0.07 to 0.11 mass pct, which are included in ASTM F90. Thus, to improve the performance of CCWN alloys for thinner balloon-expandable stents, clarifying the effect of thermomechanical treatments on the microstructures and mechanical properties of CCWN alloys with a wide range of carbon contents is essential. With this background and from the perspective of balloon-expandable stent applications, this study investigated the microstructural changes, such as the formation of the ε-phase with a hexagonal close-packed (hcp) structure and the precipitation of carbides, during the thermomechanical treatment of CCWN alloys with carbon contents ranging from 0 to 0.2 mass pct.

2 Experimental

2.1 Materials

Approximately 200 g of ingots of the alloys with varying carbon contents (0, 0.05, 0.10, and 0.20 mass pct, referred to as 0C, 0.05C, 0.10C, and 0.20C alloys, respectively,) were prepared using an induction melting furnace. Pure metals (Co, Cr, W, Ni, Fe, and Mn) were used as raw materials for metallic elements, and Cr3C2 powder was used as the raw material for carbon. The target composition of metallic elements was Co:Cr:W:Ni:Fe:Mn = 51.5:20:15:10:2:1.5 in mass fraction. The raw materials were placed in an Al2O3 crucible, melted in an Ar atmosphere, and cast into an Fe mold with a diameter of 12 mm. The contents of metallic elements and carbon in the ingots were analyzed using inductively coupled plasma-atomic emission spectroscopy (ICP-AES, Agilent 5110, Agilent Technologies Inc., CA) and the combustion-infrared (IR) absorption method (LECO CS844, LECO, MI), respectively. Table I lists the obtained chemical compositions of the alloys, which were in good agreement with those designed.

Table I Chemical Compositions of the Alloys Fabricated via Induction Melting (mass pct)
Full size table

2.2 Thermomechanical Treatment

The as-cast alloys were machined into cylindrical shapes (∅9 mm) to remove surface-reacted layers. The machined alloy ingots were subjected to a solution treatment at 1573 K for 7.2 ks. To avoid decarburization during heat treatment, the alloys were sealed inside SiO2 ampoules with Ar gas at a pressure of 0.02 to 0.03 MPa.[33] After the solution treatment, the alloys were water-quenched by breaking their SiO2 ampoules. Cold swaging was performed at room temperature, and the cross-sectional area reduction of bars was set to 58.5 pct. During the cold swaging process, the diameter of the jig was reduced from ∅8 to ∅6 mm, resulting in swaged bars measuring ∅5.8 mm in diameter. The as-swaged alloy bars were cut into cylindrical shapes with dimensions of ∅5.8 mm × 5 mm and heat treated at 1173 K–1473 K for 0.15–7.2 ks, followed by water quenching.

2.3 Analysis of the Microstructures

The microstructures of the ∅5.8 mm × 5 mm heat-treated specimens of the alloys were observed using scanning electron microscopy-backscattered electron mode (SEM-BSE, JSM-7800F, JEOL Ltd., Tokyo, Japan) and electron backscattering diffraction (EBSD, OIM analysis version 7, EDAX Inc., NJ). Specifically, the microstructures were observed in the center of the cut surface, which was perpendicular to the swaging direction. Precipitates were electrolytically extracted from the alloys in a methanol-10 vol pct acetylacetone-1 vol pct tetramethylammonium chloride solution (AA solution, Fujiwara Co., Ltd., Tokyo, Japan) at 2 V for 10.8 ks. They were collected as residues after filtering the electrolyte.[34] The obtained precipitates were analyzed by X-ray diffraction (XRD, SmartLab 9 kW, Rigaku, Tokyo, Japan) and SEM. The area fraction of the precipitates was calculated using the ImageJ image-analysis software (Version 1.53, NIH, Bethesda, MD) using 30 SEM-BSE images (8000× magnification). The crystal structure and composition of the precipitates were determined using transmission electron microscopy (TEM; JEM-2100, JEOL Ltd., Tokyo, Japan) and TEM-energy dispersive X-ray spectroscopy (TEM-EDS), respectively. The equilibrium constituent phases were investigated using the Thermo-Calc software and compared with the experimental results. The thermodynamic data sets used in this calculation were obtained from SSOL5 (SGTE General Alloy Solutions Database, version 5.0).

3 Results

3.1 As-Cast and Solution-Treated Alloys

Figure 1 shows the XRD patterns of the metallic matrix and the precipitates electrolytically extracted from the as-cast alloys. All as-cast alloys consisted of a single γ-phase. The phase of the precipitates detected in the 0.10C and 0.20C alloys was the η-phase (Co4W2C-type, JCPDS card No. 002-1198). The γ-Co peaks in Figure 1(b) are not from a precipitate but from a Co-based metallic matrix detached from the alloys during electrolytic extraction. Figure 2 shows the BSE images of the as-cast alloys. No precipitates were observed in the 0C and 0.05C alloys, whereas precipitates were detected in the 0.10C and 0.20C alloys. The inset figures in Figures 2(c) and (d) show SEM images of the precipitates on the filters after electrolytic extraction. The precipitates were indicated to be the η-phase by XRD analysis (Figure 1(b)), and their appearance was dendritic.

Fig. 1
figure 1

XRD patterns of the (a) metallic matrix and (b) precipitates of the as-cast alloys

Full size image
Fig. 2
figure 2

SEM-BSE images of the as-cast (a) 0C, (b) 0.05C, (c) 0.10C, and (d) 0.20C alloys. Inset figures present SEM images of the precipitates electrolytically extracted from the as-cast alloys

Full size image

Figure 3 shows the XRD patterns of the metallic matrix of the alloys after solution treatment at 1523 K for 7.2 ks. After the solution treatment, the matrix of all alloys was a single γ-phase, and no precipitates were detected. These results reveal that the precipitates dissolved entirely in the metallic matrix during solution treatment in all alloys, including the 0.20C alloy.

Fig. 3
figure 3

XRD patterns of the metallic matrix of the solution-treated alloys

Full size image

3.2 Change in the Phase of Matrix During Cold Swaging

Figure 4 shows the EBSD phase and grain orientation spread (GOS) maps of the as-swaged alloys. After cold swaging, the 0C and 0.05C alloys consisted of both the γ-phase and ɛ-phase, whereas the single γ-phase was maintained in 0.10C and 0.20C alloys, as shown in the phase maps. Furthermore, according to the GOS maps, swaging attained a strained microstructure in all alloys. In the CCWN alloy, the ε-phase is formed by the strain-induced martensitic transformation (SIMT).[4] These results indicated that adding 0.10 mass pct carbon stabilized the γ-phase and suppressed the SIMT during cold swaging to achieve a 58.5 pct reduction in area.

Fig. 4
figure 4

EBSD phase and GOS maps of the as-swaged 0C, 0.05C, 0.10C, and 0.20C alloys. The inset shows a schematic of the observed area on the swaged bar

Full size image

3.3 Microstructure of the Heat-Treated Alloys

XRD patterns of the metallic matrix of the alloys heat treated at 1323 K for 0.3 ks are shown in Figure 5. All the heat-treated alloys consisted of a single γ-phase matrix. The XRD patterns of the precipitates electrolytically extracted from the alloys heat treated at various temperatures and holding times are summarized in Figure 6. As illustrated in Figure 6(a), after heat treatment at 1273 K for 0.6 ks, the μ-phase (Co7W6-type, JCPDS card No. 002-1091)[35] was detected as a precipitate in the 0C alloy. By contrast, η-phase precipitates were detected in all carbon-added alloys. The SEM-BSE images of the alloys heat treated at 1273 K for 0.6 ks are shown in Figure 7. White precipitates with spherical shapes and sizes of several hundred nanometers were observed. The number of precipitates increased with increasing carbon content in the alloys.

Fig. 5
figure 5

XRD patterns of the metallic matrix of alloys heat treated at 1323 K for 0.3 ks

Full size image
Fig. 6
figure 6

XRD patterns of the precipitates electrolytically extracted from (a) all alloys heat treated at 1273 K for 0.6 ks, (b) 0C alloys, and (c) 0.10C alloys after heat treatment at 1173–1423 K for 0.6 ks

Full size image
Fig. 7
figure 7

SEM-BSE images of (a) 0C, (b) 0.05C, (c) 0.10C, and (d) 0.20C alloys heat treated at 1273 K for 0.6 ks

Full size image

The changes in the XRD patterns of the μ-phase precipitates of the 0C alloys that underwent heat treatment are shown in Figure 6(b). With an increase in the heat treatment temperature, the peak intensity of the μ-phase diminished and could no longer be detected at temperatures of 1323 K or higher. Figure 6(c) shows the change in the XRD patterns of the η-phase precipitates electrolytically extracted from the 0.10C alloy with the heat treatment temperature. η-phase precipitates were detected in the heat treatment temperature range of 1173 K–1373 K. The peak intensity reaches its maximum at 1323 K and disappears at 1473 K.

3.4 Recrystallization

The degree of recrystallization of the alloys heat treated at various temperatures was characterized by GOS maps, as shown in Figure 8. As the GOS threshold of recrystallized grains was determined to be less than 2 deg in Co-based alloys,[36,37,38,39] the same value (2 deg) was used for the threshold value in this study. Here, the recrystallized fraction (FR) was calculated using the area where the GOS value was less than 2 deg (SR) and the total area (Stotal) by Eq. [1].

$$ F_{{\text{R}}} = S_{{\text{R}}} /S_{{{\text{total}}}} $$
(1)
Fig. 8
figure 8

EBSD GOS maps of the alloys heat treated at various temperatures for 0.3 ks. *At the heat treatment temperature of 1273 K, the holding time was 0.6 ks

Full size image

The effect of the heat treatment temperature on the recrystallized fraction (FR) of the alloys is shown in Figure 9. At 1273 K, the FR value was less than 0.3 independent of the carbon content in the alloys. In addition, the FR values drastically increased at 1323 K or more, and they reached almost 1.0. These results suggested that full recrystallization occurred at 1323 K, independent of the carbon content in the alloys. A longer holding time of 7.2 ks at 1273 K was still insufficient to complete recrystallization. At 1323 K, the holding time of 0.15 ks was inadequate to complete recrystallization, but 0.3 ks was enough to complete recrystallization. The EBSD IPF maps of the alloys heat treated at 1323 K–1473 K for 0.3 ks are shown in Figure 10. All the alloys presented completely recrystallized structures. The average grain sizes of the recrystallized alloys shown in Figure 10 were calculated and plotted as a function of the heat treatment temperature (Figure 11). Here, the open and closed marks indicate the microstructures with and without precipitates, respectively. As indicated, the average grain size decreased with increasing carbon content and increased with increasing heat treatment temperature.

Fig. 9
figure 9

Effect of heat treatment temperature on the recrystallized fraction of the alloys. The holding time was 0.3 ks at all temperatures except for 1273 K, which was held for 0.6 ks, as indicated by *

Full size image
Fig. 10
figure 10

EBSD IPF maps of the alloys heat treated at various temperatures for 0.3 ks

Full size image
Fig. 11
figure 11

Effect of heat treatment temperature on the average grain size of the alloys after heat treatment for 0.3 ks

Full size image

4 Discussion

4.1 Effect of Carbon Content on the Matrix Phase and Precipitates

Figure 12 summarizes the area fraction of the ε-phase in the as-swaged alloys as a function of the carbon content. The formation of the ε-phase deteriorates the workability of CCWN alloys.[40] During the balloon-expandable stent fabrication process, cold working is essential; therefore, carbon should be added at a concentration of more than 0.05 mass pct to increase the workability of the alloys through the suppression of the formation of the ε-phase.

Fig. 12
figure 12

Effect of carbon content on the area fraction of the ɛ-phase of the as-swaged alloys

Full size image

Several studies have reported the phases of precipitates detected in Co-based alloys,[22,24,26,29,31] as summarized in Table II. Although the presence of M23C6-type precipitates has been observed in previous studies, no M23C6-type precipitates were observed in this study. Zhu and Chen reported that M23C6-type precipitates decomposed during heat treatment above 1073 K and formed M6C-type (η-phase) precipitates in CCWN alloys with a carbon content of 0.07 mass pct, as shown in Eq. [2]:[26]

$$ {\text{6M}}_{{{23}}} {\text{C}}_{{6}} \to {\text{23M}}_{{6}} {\text{C }} + {\text{ 13C}} $$
(2)
Table II Phase of Precipitates in Co-Based Alloys
Full size table

They confirmed that M6C-type precipitates formed on M23C6-type precipitates. In this study, heat treatment was conducted at temperatures higher than 1073 K; hence, no M23C6-type precipitates could be observed, as they were already decomposed into M6C-type precipitates.

The precipitates were determined to be the μ-phase and η-phase in 0C and carbon-added alloys, respectively, by XRD analysis after the extraction (Figure 6). To further support these findings, a TEM analysis was carried out (refer to the electronic supplementary material, Supplementary Figure S-1). The TEM-selected area electron diffraction (SAED) analysis confirmed that μ-phase and η-phase precipitates were present in the 0C and 0.20C alloys heat treated at 1173 K and 1373 K, respectively, for 7.2 ks. Table III summarizes the chemical compositions of the μ-phase and η-phase precipitates quantitatively analyzed via TEM-EDS. For the η-phase precipitates, although the carbon content has not been reported in our previous studies,[25] the carbon content of the η-phase was determined to be ~ 7 mol pct. Given the limitations of TEM-EDX, however, this value was not reliable, however. Nonetheless, the detection of carbon in the precipitates was evidence that the η-phase was a carbide. In contrast, no carbon was detected in the μ-phase; thus, it was intermetallic.

Table III Chemical Compositions of the μ-Phase and η-Phase Precipitates (mol pct)
Full size table

This is the first study revealing the formation of μ-phase precipitates in CCWN alloys. The μ-phase chemical formula is generally expressed as Co7W6.[25] In this study, the (Co + Ni):(W + Cr) atomic ratio was almost equal to 7:6. This result suggests that Ni atoms were introduced into the Co sites, and Cr atoms were introduced into the W sites.

The phases of the precipitates observed after heat treatment are summarized in Figure 14 as a time-temperature-precipitation diagram. The mark at each experimental point in the figure corresponds to the phase of the precipitates, as determined from the XRD analysis of the precipitates electrolytically extracted from the alloys. η-phase precipitates were detected after holding times of up to 0.3 ks at 1323 K and 1373 K for the 0.05C alloy, and at 1423 K for the 0.10C alloy. However, the precipitates were not detected after a long holding time at the same temperature, implying that they dissolved in the metallic matrix during heat treatment. These results indicate that the η-phase precipitates detected under these conditions are unstable and form during the temperature rise during heat treatment.

In the 0C alloys, the μ-phase was stable at temperatures up to 1273 K (Figure 13(a)). The Thermo-Calc calculations for the equilibrium of the μ-phase with the carbon content are shown in Figure 14. According to the calculated phase diagram, the μ-phase was unstable with increasing temperature and carbon content, and it dissolved in the γ matrix. This result is in agreement with the results of the present study. In the calculation, the μ-phase in the 0.10C alloys was stable up to 1100 K. In this study, the μ-phase was not detected in 0.05C alloys, even at 1173 K.

Fig. 13
figure 13

Phases of the precipitates formed in the as-swaged and heat-treated (a) 0C, (b) 0.05C, (c) 0.10C, and (d) 0.20 C alloys

Full size image
Fig. 14
figure 14

Thermo-Calc calculations for the stable equilibrium μ-phase with varying carbon contents

Full size image

On the contrary, as per the calculations, the μ-phase was stable up to 1223 K in the 0.05C alloy. However, the CCWN alloy is a complicated material; therefore, a more precise database is required to more accurately calculate the phases in this alloy system. In general, however, the Thermo-Calc calculations showed a similar trend to the experimental results and can serve as guidelines.

The precipitation temperature range of the η-phase detected in the carbon-added alloys increased with increasing carbon content (Figures 13(b) through (d)). Figure 15 shows the effect of the carbon content on the area fraction of the η-phase precipitates. The area fraction of the η-phase precipitates increased with increasing carbon content, which can be qualitatively explained by the calculated phase diagram (Figure 14).

Fig. 15
figure 15

Effect of carbon content on the area fraction of the η-phase precipitates in the alloys heat treated at 1323–1473 K for 0.3 ks

Full size image

4.2 Recrystallized Grain Size

Several studies have reported that the recrystallization temperature of the CoCrFeMnNi HEA increases with the addition of carbon.[7,29] This can be attributed to the carbon in the solid solution rather than to carbides precipitating in the matrix. The migration of grain boundaries is impeded by strong interactions between interstitial solutes and grain boundaries, which increase the energy barrier for recovery and recrystallization during annealing.[41] However, in this study, no significant effect of carbon content on the recrystallization temperature was detected in the CCWN alloys. All the alloys were fully recrystallized after heat treatment at 1323 K for 0.3 ks with a homogeneous structure and not fully recrystallized at 1273 for 7.2 ks. The carbon content in the CCWN alloy exhibited no effect on recrystallization.

The strut diameter of balloon-expandable stents, currently in practical use, is 80–110 µm. The requirement for a fine grain size corresponds to a minimum of three grains passing through the strut. Therefore, the grain size of the alloys is less than 30 μm for balloon-expandable stent applications.[42] In this study, the grain size required for a balloon-expandable stent (< 30 μm) was acquired after the recrystallization of carbon-added alloys at 1323 K–1473 K for 0.3 ks as illustrated in Figure 11.

As shown in Figure 15, in the 0.20C alloy, grain growth was suppressed even at 1473 K, which is consistent with the existence of precipitates with sizes of approximately 220 nm (see Electronic Supplementary Figure S-2). In the CoCrFeMnNi HEA and Ni-based superalloys, refinement by recrystallization and subsequent grain growth were both controlled by finely dispersed precipitates in the size range of several hundred nanometers, which provided a strong barrier to dislocation motion and impeded grain boundary motion.[30,43]

Hence, the findings of this study suggested that grain growth during the heat treatment was suppressed by the pinning effect of the precipitates in the matrix. In our following paper, we will report the effect of grain size, precipitates, and carbon content on the mechanical properties of these alloys by focusing on alloys with a grain size of less than 30 μm.

5 Conclusions

Co–Cr–W–Ni alloys with carbon contents of 0, 0.05, 0.10, and 0.20 mass pct were prepared using an induction melting furnace, and the observed microstructural changes in these alloys after solution treatment, cold swaging, and heat treatment were studied, focusing on the phase of the matrices and precipitates. The following conclusions were drawn:

  1. 1.

    After solution treatment at 1523 K for 7.2 ks, the precipitates present in the as-cast alloys were completely dissolved in all the alloys.

  2. 2.

    A single γ-phase was observed in the 0.10C and 0.20C alloys after cold swaging with an area reduction of 58.5 pct, whereas both the γ-phase and ε-phase were detected in the 0C and 0.05C alloys. It is suggested that the ε-phase is formed by SIMT. Thus, adding 0.10 mass pct of carbon effectively stabilized the γ-phase during cold swaging.

  3. 3.

    The phases of the precipitates formed during heat treatment were the μ-phase and η-phase in the 0C and carbon-containing alloys, respectively. Both precipitates dissolved completely in the matrix at higher temperatures. The precipitation temperature range of the η-phase increased with increasing carbon content. The stability of the precipitates qualitatively agreed with that of the obtained phase diagram.

  4. 4.

    The heat treatment conditions required to obtain the grain size essential for balloon-expandable stents (< 30 μm) were clarified for completely recrystallized alloys. Grain growth in the carbon-added alloys was inhibited by the η-phase precipitates owing to the pinning effect of the precipitates.