Effect of a Heterogeneous Nitrogen Diffusion Phase on Four-Point Bending Fatigue Properties in Commercially Pure Titanium

Abstract

The purpose of this study is to develop commercially pure (CP) titanium having a higher fatigue strength than titanium alloys developed via heterogeneous nitrogen diffusion. The microstructure of CP titanium having a heterogeneous nitrogen diffusion phase, which was fabricated by consolidating gas-nitrided powders, was characterized, and its fatigue properties were examined. The nitrogen content and hardness of CP titanium compacts having a heterogeneous nitrogen diffusion phase increased with increasing powder gas nitriding temperature and sintering temperature. The fatigue limit and fatigue life of CP titanium compacts increased with increasing sintering temperature and with decreasing powder gas-nitriding temperature. In particular, CP titanium having a heterogeneous nitrogen diffusion phase that is fabricated by high-temperature sintering of powders treated with low-temperature nitriding has a higher fatigue limit than un-nitrided bulk Ti–6Al–4V alloy. The fatigue limit of CP titanium can be controlled by optimizing the powder gas nitriding and sintering temperatures.

Fig. 12 (a) Optical micrograph and (b) nitrogen map of sintered compact fabricated from powder gas-nitrided at 873 K (sintering temperature: 1273 K) tested at σ_a = 240 MPa and N = 7 × 10³ cycles.

1. Introduction

Commercially pure (CP) titanium, which does not contain toxic elements such as Al or V, is used for bio-implants because of its high corrosion resistance, owing to the native oxide layer on its surface,¹⁾ and its superior biocompatibility.^2,3) Nevertheless, CP titanium has a lower fatigue strength than titanium alloys⁴⁾ and inferior tribological characteristics compared to ceramics and ferrous materials.^5,6) The mechanical and chemical properties of metallic materials are governed by their surface microstructure; therefore, surface modification has been shown to be effective in improving various characteristics of titanium-based materials, such as biocompatibility,^7,8) wear resistance,^9,10) and fatigue properties,¹¹⁾ to minimize the damage when used in the human body.

However, in some cases, surface modification has revealed tradeoffs between mechanical and chemical properties of titanium-based materials; for example, the high-corrosion-resistance titanium oxide layer on the surface disappeared during nitriding, which improved the tribological properties but reduced the corrosion resistance of titanium alloys.¹²⁾ Thus, multifunctional surfaces^13–16) should be formed to improve various characteristics of titanium-based materials for bio-implants, because in vivo damage to bio-implants is induced by various phenomena: cyclic loading, friction, wear, and corrosion. Morita et al.¹⁴⁾ and the present authors¹⁵⁾ reported that the mechanical properties, i.e., fatigue strength and wear resistance, of commercially pure (CP) titanium were improved by low-temperature nitriding, while conventional nitriding reduces the fatigue strength of titanium-based materials as a result of grain-coarsening.^17–20)

To overcome the tradeoff between properties in titanium-based materials, several concepts for designing a heterogeneous microstructure via powder metallurgy have been proposed.^21–23) Ameyama et al.^24–27) developed the concept of a bimodal structure for the fabrication of titanium-based materials with high strength and ductility based on powder metallurgy, whereby a coarse-grained structure surrounded by a network of fine grains is formed. The present authors^28,29) developed the concept of heterogeneous nitrogen diffusion into CP titanium using powder metallurgy, which enables the formation of both a titanium oxide layer with high corrosion resistance and a nitrided layer with high hardness. Furthermore, static mechanical properties, such as tensile strength^23,28) and compressive proof stress,³⁰⁾ and tribological properties^28,29) were improved compared to conventional titanium. To synthesize CP titanium having a heterogeneous nitrogen diffusion structure, with sufficient performance as a biomaterial, it is important to investigate the effects of the heterogeneous nitrogen diffusion phase on the fatigue resistance of CP titanium, free of toxic elements.

The purpose of this study is to employ heterogeneous nitrogen diffusion to develop CP titanium having a higher fatigue strength than titanium alloys. The microstructure of materials having a heterogeneous nitrogen diffusion phase fabricated by consolidating gas-nitrided CP titanium powder was characterized, and their fatigue properties were examined on the basis of this characterization and on fracture mechanics.

2. Experimental Procedures

2.1 Material and specimens

The raw material used in the current study was a CP titanium powder (25.6 µm particle diameter) with 0.122% O, 0.008% N, 0.005% H, 0.037% Fe, and 0.003% C (on a mass basis), with the remainder consisting of Ti.³¹⁾ Figure 1 illustrates the process used to form the CP titanium, which had a heterogeneous nitrogen diffusion phase, and two types of specimens fabricated by gas nitriding and spark plasma sintering (SPS). Gas nitriding was conducted at 823 K or 873 K for 18 ks to produce a nitrided layer on the CP titanium powder particle surfaces (see Figs. 1(a) and (b)).

Fig. 1

Fabrication process of commercially pure titanium having a heterogeneous nitrogen diffusion phase and specimen preparation.

The gas-nitrided powders were subsequently consolidated by SPS at 1073 K, 1273 K, or 1473 K for 1.8 ks under a vacuum of less than 15 Pa and a pressure of 50 MPa. CP titanium compacts having a heterogeneous nitrogen diffusion phase (10 mm thickness, 25 mm diameter) were produced using a graphite die with an internal diameter of 25 mm (see Fig. 1(c)). The as-received CP titanium powder was also consolidated for comparison (the un-nitrided series). The CP titanium compacts were machined into specimens for four-point bending fatigue tests and fatigue crack propagation tests, and their surfaces were polished with emery paper (see Figs. 1(d) and (e)). The four-point bending fatigue test specimens were subsequently polished using a SiO₂ suspension to obtain a mirror finish.

2.2 Microstructural characterization

The surface hardness of CP titanium compacts having a heterogeneous nitrogen diffusion phase was measured using a nano-indenter (FISCHERSCOPE HM2000) at a load of 10 mN. The microstructures of the powders and sintered compacts were characterized using an electron probe micro analyzer (EPMA) at an accelerating voltage of 15 kV to examine the effect of gas nitriding and SPS temperatures on the nitrogen diffusion behavior in CP titanium.

2.3 Four-point bending fatigue tests

The sintered compact (25 mm diameter) was machined by wire electric discharge machining to produce plate-type specimens with dimensions of 18 × 3 × 1 mm, as shown in Fig. 1(d). Four-point bending fatigue tests were conducted in an electrodynamic fatigue testing apparatus under a stress ratio, R, of 0.1 using plate-type specimens in air at room temperature. The frequency of stress cycling was 10 Hz. In this study, the fatigue limit, σ_w, was defined as the average of the maximum stress amplitude without specimen failure and the minimum stress amplitude at which the specimens failed. To determine the mechanisms of fatigue failure in CP titanium compacts having a heterogeneous nitrogen diffusion phase, optical microscopy was conducted on the surfaces of some specimens during fatigue testing, i.e., with the specimens installed in the fatigue testing machine.³²⁾ The stress intensity factor, K, was calculated using the formula proposed by Newman and Raju,³³⁾ and the aspect ratio of the fatigue crack under bending was estimated on the basis of a previous study.³⁴⁾

After testing, the fracture surfaces were observed using scanning electron microscopy (SEM) and crack initiation sites were analyzed by EPMA to elucidate the mechanism of fatigue failure in CP titanium compacts having a heterogeneous nitrogen diffusion phase.

2.4 Fatigue crack propagation tests

The sintered compact (25 mm diameter) was machined by wire electric discharge machining to produce disk-shaped compact (DC(T)) specimens (2 mm thickness, 18.7 mm width) based on the ASTM standard, as shown in Fig. 1(e). K-increasing tests were conducted in ambient air at a force ratio, R, of 0.1 to examine the fatigue crack propagation behaviors of the CP titanium compacts having a heterogeneous nitrogen diffusion phase. The frequency of stress cycling was 30 Hz. The details can be found elsewhere.^34,35)

3. Results

3.1 Microstructural characterization of CP titanium powder

Figure 2 shows the cross-sectional nitrogen profiles measured by EPMA for gas-nitrided and as-received CP titanium powders to characterize the nitrided layer under a step size of 0.1 µm. Powders embedded in resin were polished before EPMA analysis. The maximum nitrogen content of CP titanium powders increased with increasing powder gas nitriding temperature at the surface. In addition, the nitrogen content of gas-nitrided CP titanium powders tended to gradually decrease with increasing distance from the surface, and became constant near the un-nitrided interior, although the nitrogen element was detected even for the un-nitrided powders, in which 0.008% N contained, and varied slightly. It was confirmed that no titanium nitrides formed on CP titanium powders nitrided below 873 K;²⁸⁾ thus, a nitrogen diffusion layer was produced on the gas-nitrided CP titanium powder particle surfaces. The thickness of this layer increased with increasing powder gas nitriding temperature.

Fig. 2

Cross-sectional nitrogen profiles for un-nitrided powder and gas-nitrided powders, as measured by EPMA.

3.2 Microstructural characterization of CP titanium having a heterogeneous nitrogen diffusion phase

Figure 3 shows nitrogen maps obtained by EPMA for CP titanium compacts made by consolidating gas-nitrided powders over a 40,000 µm² area under a step size of 1 µm. The colors in the EPMA maps indicate the nitrogen content. The signal intensities for CP titanium samples fabricated by sintering powders gas-nitrided at 823 K and 873 K differed in Fig. 3. The nitrogen content of CP titanium compacts fabricated by sintering gas-nitrided powders tended to increase with increasing powder gas nitriding temperature owing to the high nitrogen content at the surface of the CP titanium powders, as shown in Fig. 2. Furthermore, nitrogen diffusion behavior in CP titanium compacts depended on the sintering temperature. In the images of compacts consolidated at 1073 K for gas-nitrided powders (see Fig. 3(a)), the high-nitrogen-content region forms a continuous connected network structure around the substrate with low nitrogen content. The size of this network structure corresponds to the particle diameter of the gas-nitrided CP titanium powder (see Fig. 3 for a schematic of the gas-nitrided powder (25.6 µm particle diameter) used in this study). This is because all of the nitrogen-diffused regions, which were formed at the surface of the CP titanium powders containing nitrogen, bonded together during SPS. In contrast, a characteristic nitrogen diffusion phase with an almost lamellar structure was formed in the CP titanium fabricated by consolidating gas-nitrided powders at 1473 K, as shown in Fig. 3(c).

Fig. 3

Nitrogen maps for sintered compacts fabricated from gas-nitrided powders, as measured by EPMA.

Figure 4 shows the nitrogen profiles measured by EPMA for compacts fabricated by sintering as-received powder and gas-nitrided CP titanium powders to investigate the effect of powder gas nitriding temperature on the nitrogen profiles. The maximum nitrogen content of CP titanium increased with increasing powder gas nitriding temperature. In addition, the nitrogen intensity detected in CP titanium compacts varied as a function of position, because the nitrogen contained in gas-nitrided CP titanium powder diffused into the core of the powder particles during sintering, as shown in Fig. 3(c). Figure 5 shows the nitrogen profiles measured by EPMA for compacts fabricated by sintering CP titanium powders gas-nitrided at 873 K to investigate the effect of sintering temperature on the nitrogen profiles. The maximum nitrogen content of CP titanium having a heterogeneous nitrogen diffusion phase increased with increasing the sintering temperature. Thus, a heterogeneous nitrogen profile was formed in CP titanium compacts fabricated by sintering gas-nitrided powders.

Fig. 4

Nitrogen profiles for sintered compacts fabricated from un-nitrided powder and gas-nitrided powders, obtained by EPMA (sintering temperature: 1473 K).

Fig. 5

Nitrogen profiles for sintered compacts fabricated from gas-nitrided powders, obtained by EPMA (nitriding temperature: 873 K).

The Vickers hardness of CP titanium compacts was measured to investigate the effect of powder gas nitriding temperature on the hardness distributions. Figure 6 shows hardness profiles for the un-nitrided compact and compacts made by sintering at 1473 K powders that were gas-nitrided at 823 K or 873 K. The hardness values for CP titanium compacts fabricated by consolidating gas-nitrided powders varied as a function of position, and exhibited higher hardness than the average value for the un-nitrided compact (246 ± 20 HV). The maximum hardness of CP titanium compacts having a heterogeneous nitrogen diffusion structure increased with increasing powder gas nitriding temperature. This was because the maximum nitrogen content of CP titanium compacts increased with increasing powder gas nitriding temperature, as shown in Fig. 4. To examine the effect of the sintering temperature on hardness, Fig. 7 shows the hardness profiles for compacts fabricated by sintering powders gas-nitrided at 873 K. The maximum hardness of CP titanium compacts having a heterogeneous nitrogen diffusion phase depended on the sintering temperature and increased with increasing sintering temperature. The unique hardness variations correspond to the heterogeneous nitrogen diffusion phase, as observed for the nitrogen profiles in Fig. 5.

Fig. 6

Hardness distributions of sintered compacts fabricated from un-nitrided powder and gas-nitrided powders (sintering temperature: 1473 K).

Fig. 7

Hardness distributions of sintered compacts fabricated from gas-nitrided powders (nitriding temperature: 873 K).

Thus, the nitrogen analyses and hardness variations suggest that the heterogeneous profile of the nitrogen diffusion phase depends on both the powder gas nitriding and sintering temperatures.

3.3 Four-point bending fatigue properties of CP titanium having a heterogeneous nitrogen diffusion phase

Fatigue tests were conducted on CP titanium compacts fabricated using gas-nitrided powders to investigate the effect of the heterogeneous nitrogen diffusion phase on fatigue properties. Figure 8 shows the results of four-point bending fatigue tests on CP titanium fabricated using powders gas-nitrided at (a) 823 K and (b) 873 K. The plots with an arrow represent a run-out specimen without failure at N = 10⁷ cycles, and the data for the un-nitrided compact and conventionally nitrided bulk CP titanium¹⁵⁾ are also shown in Fig. 8 for comparison. The fatigue limit and fatigue life for sintered compacts tended to increase with increasing sintering temperature at a comparable powder gas nitriding temperature. In addition, the fatigue limit for sintered compacts tended to increase with decreasing powder gas nitriding temperature at a comparable sintering temperature. Thus, the fatigue properties of the CP titanium compact having a heterogeneous nitrogen diffusion phase depend on both the powder gas nitriding and sintering temperatures. Furthermore, the fatigue limit of the CP titanium having a heterogeneous nitrogen diffusion phase was higher than that of conventionally nitrided bulk CP titanium,¹⁵⁾ but CP titanium compacts having a continuous connected network structure of the high-nitrogen-content region forms (see Fig. 3) had a lower fatigue limit than the un-nitrided compact. On the other hand, the fatigue limit of the CP titanium compacts having lamellar microstructures (see Fig. 3) was higher than that of the un-nitrided compact. In the present study, CP titanium compacts fabricated by sintering at 1473 K a powder gas-nitrided at 823 K exhibited the highest fatigue limit, 1.5 times higher than that of the un-nitrided compact, as shown in Fig. 8(a).

Fig. 8

Results of four-point bending fatigue tests for sintered compacts fabricated from powders gas-nitrided at (a) 823 K and (b) 873 K.

4. Discussion

4.1 Fractography

The fracture surfaces of the failed specimens were observed by SEM to investigate the fracture mechanism of CP titanium compacts fabricated using gas-nitrided powders. Figure 9 shows an SEM fractograph for each specimen fabricated with gas-nitrided powders near the crack initiation site. All specimens in the present study failed as a result of surface-initiated fracture. The tensile stress was applied to the upper surface in this figure and fracture surfaces were observed at the same magnification as the nitrogen maps obtained by EPMA analysis shown in Fig. 3. Characteristic facets were observed on the fracture surfaces of the compacts fabricated with gas-nitrided powders and corresponded to the nitrogen maps in Fig. 3. In particular, a rough morphology was clearly observed for the entire fracture surface (not only in the instantaneous fracture region but also in the fatigue crack propagation region) of the compact that was formed with powder gas-nitrided at 823 K and that was sintered at 1473 K (see Fig. 10). Lamellar microstructures were formed in the CP titanium compact that was fabricated by sintering at 1473 K powder gas-nitrided at 823 K (see Fig. 3(c)) and that had the highest fatigue limit. Nalla et al.³⁶⁾ reported that the lamellar microstructure exhibited a high fatigue crack propagation resistance, typical of titanium alloys, higher than that of equiaxed microstructures because higher levels of crack tip shielding were promoted by roughness-induced crack closure. Non-propagating crack was observed at the surface of the compact sintered at 1473 K of powder gas-nitrided at 823 K without failure at N = 10⁷ cycles; therefore, the fatigue crack path in the compact sintered at 1473 K of powder gas-nitrided at 823 K was influenced by the microstructure: a rough morphology was clearly observed for the fracture surface and roughness-induced crack closure would be expected to occur in the CP titanium having a lamellar nitrogen diffusion phase, which resulted in a high fatigue limit.

Fig. 9

SEM fractographs of sintered compacts fabricated from gas-nitrided powders.

Fig. 10

Macroscopic SEM fractograph of the compact with the highest fatigue limit (σ_a = 380 MPa, N_f = 2.2 × 10⁵ cycles). It was fabricated by gas nitriding powders at 823 K followed by sintering at 1473 K.

Thus, it was confirmed that microstructures having a heterogeneous nitrogen diffusion phase strongly influence the fatigue properties of CP titanium and that the fatigue limit can be controlled by optimizing the powder gas nitriding and SPS temperatures, i.e., through high-temperature sintering of powders subjected to low-temperature nitriding.

4.2 Fatigue crack initiation and propagation behaviors in CP titanium having a heterogeneous nitrogen diffusion phase

To explain the microstructure near the crack initiation site, the fatigue crack initiation and propagation behaviors were examined for compacts fabricated from gas-nitrided powder. In particularly, CP titanium compacts having a continuous connected network structure of the high-nitrogen-content region forms had a low fatigue limit (see Fig. 8). In order to clarify the reason why the continuous connected network structure reduced the fatigue limit of CP titanium, the surface of the compact that was formed with powder gas-nitrided at 873 K and that was consolidated at 1273 K, which had a lower fatigue limit than the un-nitrided compact (see Fig. 8(b)), was examined by optical microscopy after varying number of cycles, because we can clearly observe the effect of the continuous connected network of the nitrogen diffusion phase on the fatigue fracture mechanism of CP titanium. From this viewpoint, discussions were made based on the results for the compact that was formed with powder gas-nitrided at 873 K and that was consolidated at 1273 K as typical examples in this section.

Figure 11 shows (a) an optical micrograph and (b) the nitrogen map of the surface of the sintered compact fabricated from powder gas-nitrided at 873 K, tested at a stress amplitude of 240 MPa. Figure 11 reveals the presence of a 20 µm long fatigue crack after 2.00 × 10³ cycles. This crack was initiated in the low-nitrogen-content region of the CP titanium having a continuous connected network of nitrogen diffusion phase during the fatigue test (see Fig. 11(b)). The present authors have previously reported^{27,34,37–40)} that a relatively coarse microstructure with low strength was the predominant crack initiation site in CP titanium and a Ti–6Al–4V alloy with a bimodal grain size distribution during fatigue tests. Uematsu et al.⁴¹⁾ reported that a fatigue crack was initiated at the untransformed coarse grains in the CP titanium heat-treated at a temperature above the β transus. These studies indicate that the weak region, which is the un-nitrided region with low hardness and low nitrogen content in the present study, acts as a crack initiation site in CP titanium having a heterogeneous microstructure. Thus, the continuous connected network of the nitrogen diffusion phase does not increase the resistance to fatigue crack initiation in CP titanium.

Fig. 11

(a) Optical micrograph and (b) nitrogen map of sintered compact fabricated from powder gas-nitrided at 873 K (sintering temperature: 1273 K) tested at σ_a = 240 MPa and N = 2 × 10³ cycles.

In addition, fatigue crack propagation was examined in CP titanium having a continuous connected network of a nitrogen diffusion phase. Figure 12 shows (a) an optical micrograph and (b) the nitrogen map of the compact surface tested at σ_a = 240 MPa. A fatigue crack was initiated near the specimen edge and gradually propagated with increasing number of cycles. In Fig. 12(b), the crack profile is represented by the black line and the crack tip is indicated by arrows after each cycle. The fatigue crack grew with increasing number of cycles and the crack profile was not influenced by the nitrogen map. However, the rate of increase in the surface crack length, Δ2c, tended not to increase with increasing number of cycles, although Δ2c generally does increase with increasing number of cycles. These differences observed between the fatigue crack propagation behavior of CP titanium having a heterogeneous nitrogen diffusion phase and those of conventional materials were attributed to the dependence of the local crack propagation rate on the nitrogen diffusion phase: Δ2c was small when a fatigue crack propagated through the un-nitrided region (e.g., from 1000 to 2000 cycles), whereas Δ2c was large when a fatigue crack propagated through the nitrogen diffusion phase (e.g., from 0 to 1000 and from 5500 to 6000 cycles). Figure 13 shows SEM fractographs of the compact failed at σ_a = 240 MPa after testing. In Fig. 13, the crack tip is shown by arrows after each cycle, and the relatively large facets on the fracture surface correspond to the high-nitrogen-intensity region in Fig. 12(b).

Fig. 12

(a) Optical micrograph and (b) nitrogen map of sintered compact fabricated from powder gas-nitrided at 873 K (sintering temperature: 1273 K) tested at σ_a = 240 MPa and N = 7 × 10³ cycles.

Fig. 13

Typical fracture surface features of sintered compact that was fabricated from powder gas-nitrided at 873 K (sintering temperature: 1273 K) and that failed at σ_a = 240 MPa.

The effect of the continuous connected network of the nitrogen diffusion phase on fatigue properties was examined on the basis of fracture mechanics. Figure 14 shows the crack growth rate, da/dN, for a small crack in the compact formed by consolidating at 1273 K powder gas-nitrided at 873 K against the stress intensity range, ΔK, under four-point bending fatigue tests at 240 MPa. da/dN varied with ΔK, even though the da/dN generally tends to increase with increasing ΔK. This was because the continuous connected network of the nitrogen diffusion phase facilitated local fatigue crack propagation, as shown in Fig. 12(b).

Fig. 14

Relationship between da/dN and ΔK in sintered compacts fabricated from powder gas-nitrided at 873 K tested at σ_a = 240 MPa under four-point bending and for long cracks in sintered compacts fabricated from un-nitrided powder and powder gas-nitrided at 873 K (sintering temperature: 1273 K).

To compare fatigue crack propagation in the un-nitrided compact and the compact having a nitrogen diffusion phase, we conducted K-increasing tests using the DC(T) specimens. Figure 14 plots the crack growth rate, da/dN, versus the stress intensity range, ΔK, for long cracks in un-nitrided specimens as well as CP titanium powder that was gas-nitrided at 873 K and consolidated at 1273 K. The da/dN for CP titanium having a nitrogen diffusion phase were consistently higher than those for un-nitrided specimens at comparable ΔK levels. Thus, heterogeneous nitrogen diffusion accelerates the crack propagation in CP titanium. Furthermore, the da/dN values for small cracks were higher than those for long cracks under comparable ΔK in CP titanium fabricated with gas-nitrided powders. The short crack effect was still observed in the fatigue crack propagation behavior of the newly developed CP titanium with a heterogeneous nitrogen diffusion phase as well as conventional materials.

Thus, a continuous connected network of a nitrogen diffusion phase does not increase the resistance to fatigue crack initiation; rather, it reduces the resistance to fatigue crack propagation in CP titanium, which results in a lower fatigue limit (see Fig. 8(b)). This is consistent with the fact that CP titanium without a continuous connected network of nitrogen diffusion phase (powder gas-nitrided at 823 K, consolidated at 1473 K) exhibited the highest fatigue limit, as shown in Fig. 8(a).

4.3 Comparison of fatigue properties of CP titanium having a heterogeneous nitrogen diffusion phase and Ti–6Al–4V alloy

It appears that CP titanium with superior fatigue properties can be fabricated by optimizing the powder gas nitriding and SPS temperatures, as described in the previous sections. In this section, we compare the fatigue properties of CP titanium having a heterogeneous nitrogen diffusion phase with those of a bulk titanium alloy. The results are plotted as S-N diagrams in Fig. 15 for un-nitrided bulk Ti–6Al–4V⁴²⁾ and CP titanium compacts that were made from powder gas-nitrided at 823 K and that were then consolidated at 1473 K, which showed the highest fatigue limit in this study. A fatigue limit is clearly observed for each specimen; therefore, the S-N curves can be drawn using a bilinear S-N model with a fatigue limit (JSMS standard regression models).⁴³⁾ The regression S-N curves for each specimen series could be represented by the following equations:

Fig. 15

Results of four-point bending fatigue tests for sintered CP titanium (sintering temperature: 1473 K) made from powder gas-nitrided at 823 K and for un-nitrided bulk Ti–6Al–4V alloy.⁴²⁾

CP titanium powder gas-nitrided at 823 K and consolidated at 1473 K:   

\begin{equation} \log 10\sigma_{\text{a}} = -42.242 \log 10(\mathit{N}_{\text{f}}) + 574.96, \end{equation} (1)

un-nitrided bulk Ti–6Al–4V:⁴²⁾   

\begin{equation} \log 10\sigma_{\text{a}} = -35.196 \log 10(\mathit{N}_{\text{f}}) + 561.53, \end{equation} (2)

where σ_a is the stress amplitude (MPa), which is higher than the fatigue limit, and N_f is the number of cycles. In the present study, the single-log model was adopted. The fatigue limit of the CP titanium having a heterogeneous nitrogen diffusion phase (325 MPa) was slightly higher than that of un-nitrided Ti–6Al–4V (319 MPa), even though the former had a slightly shorter fatigue life at high stress amplitudes.

Clearly, forming a heterogeneous nitrogen diffusion phase is effective in increasing the fatigue limit of CP titanium. CP titanium contains no toxic elements, such as aluminum or vanadium, which are present in Ti–6Al–4V. Thus, the heterogeneous nitrogen diffusion approach, involving high-temperature sintering of powders treated with low-temperature nitriding, enables the fabrication of biomaterials with superior mechanical properties.

5. Conclusion

The microstructure of CP titanium having a heterogeneous nitrogen diffusion phase fabricated by consolidating gas-nitrided CP titanium powders was characterized, and the fatigue properties were determined. The main conclusions of the present work are as follows:

1. (1)    The heterogeneous nitrogen profile and hardness of the nitrogen diffusion phase depend on both the powder gas nitriding and sintering temperatures.
2. (2)    The fatigue limit for CP titanium compacts increases with decreasing powder gas nitriding temperature and with increasing sintering temperature.
3. (3)    Formation of the continuous connected network of a nitrogen diffusion phase decreases the fatigue limit and increases the fatigue crack growth rate, da/dN, of CP titanium.
4. (4)    CP titanium with a lamellar nitrogen diffusion phase fabricated by high-temperature sintering of powders first subjected to low-temperature nitriding has a higher fatigue limit than un-nitrided bulk Ti–6Al–4V alloy.

Acknowledgements

The authors would like to thank JSPS KAKENHI (Grant No. 19H02024), Light Metal Educational Foundation, Inc., and Nippon Sheet Glass Foundation for Materials Science and Engineering for their financial support.

REFERENCES

• 1)   R.  Strietzel,  A.  Hösch,  H.  Kalbfleisch and  D.  Buch: Biomaterials 19 (1998) 1495–1499. doi:10.1016/S0142-9612(98)00065-9
• 2)   Y.  Kurashina,  A.  Ezura,  R.  Murakami,  M.  Mizutani and  J.  Komotori: J. Mater. Sci. Mater. Med. 30 (2019) 57. doi:10.1007/s10856-019-6259-1
• 3)   S.  Kikuchi,  S.  Yoshida,  Y.  Nakamura,  K.  Nambu and  T.  Akahori: Surf. Coat. Technol. 288 (2016) 196–202. doi:10.1016/j.surfcoat.2016.01.023
• 4)   M.  Niinomi: Int. J. Fatigue 29 (2007) 992–1000. doi:10.1016/j.ijfatigue.2006.09.021
• 5)   K.G.  Budinski: Wear 151 (1991) 203–217. doi:10.1016/0043-1648(91)90249-T
• 6)   S.  Kikuchi,  K.  Katahira and  J.  Komotori: J. Japan Inst. Met. Mater. 82 (2018) 341–348. doi:10.2320/jinstmet.J2018017
• 7)   D.R.  Bloyer,  J.M.  Gomez-Bega,  E.  Saiz,  J.M.  McNaney,  R.M.  Cannon and  A.P.  Tomsia: Acta Mater. 47 (1999) 4221–4224. doi:10.1016/S1359-6454(99)00280-3
• 8)   R.S.  Bedi,  D.E.  Beving,  L.P.  Zanello and  Y.  Yan: Acta Biomater. 5 (2009) 3265–3271. doi:10.1016/j.actbio.2009.04.019
• 9)   H.  Dong and  T.  Bell: Wear 238 (2000) 131–137. doi:10.1016/S0043-1648(99)00359-2
• 10)   K.  Farokhzadeh,  A.  Edrisy,  G.  Pigott and  P.  Lidster: Wear 302 (2013) 845–853. doi:10.1016/j.wear.2013.01.070
• 11)   T.  Morita,  A.  Miyatani,  S.  Kariya,  M.  Kumagai,  S.  Takesue and  J.  Komotori: Results Mater. 7 (2020) 100128. doi:10.1016/j.rinma.2020.100128
• 12)   B.S.  Yilbas,  A.Z.  Sahin,  Z.  Ahmad and  B.J.  Abdul Aleem: Corros. Sci. 37 (1995) 1627–1636. doi:10.1016/0010-938X(95)00065-R
• 13)   Y.  Tsutsumi,  M.  Niinomi,  M.  Nakai,  M.  Shimaburuku,  M.  Ashida,  P.  Chen,  H.  Doi and  T.  Hanawa: Metals 6 (2016) 76. doi:10.3390/met6040076
• 14)   T.  Morita,  H.  Takahashi,  M.  Shimizu and  K.  Kawasaki: Fatigue Fract. Eng. Mater. Struct. 20 (1997) 85–92. doi:10.1111/j.1460-2695.1997.tb00404.x
• 15)   S.  Kikuchi,  A.  Ueno and  H.  Akebono: Int. J. Fatigue 139 (2020) 105772. doi:10.1016/j.ijfatigue.2020.105772
• 16)   I.  Apachitei,  A.  Leoni,  A.C.  Riemslag,  L.E.  Fratila-Apachitei and  J.  Duszczyk: Appl. Surf. Sci. 257 (2011) 6941–6944. doi:10.1016/j.apsusc.2011.03.038
• 17)   S.  Kikuchi,  S.  Yoshida and  A.  Ueno: Int. J. Fatigue 120 (2019) 134–140. doi:10.1016/j.ijfatigue.2018.11.005
• 18)   S.  Takesue,  S.  Kikuchi,  H.  Akebono,  T.  Morita and  J.  Komotori: Results Mater. 5 (2020) 100071. doi:10.1016/j.rinma.2020.100071
• 19)   K.  Tokaji,  T.  Ogawa and  H.  Shibata: J. Mater. Eng. Perform. 8 (1999) 159–167. doi:10.1361/105994999770346990
• 20)   K.  Farokhzadeh and  A.  Edrisy: Mater. Sci. Eng. A 620 (2015) 435–444. doi:10.1016/j.msea.2014.10.008
• 21)   E.  Ma and  T.  Zhu: Mater. Today 20 (2017) 323–331. doi:10.1016/j.mattod.2017.02.003
• 22)   J.  Umeda,  T.  Mimoto,  K.  Kondoh and  B.  Fugetsu: Trans. JWRI 41 (2012) 49–52.
• 23)   Y.  Yamabe,  J.  Umeda,  H.  Imai and  K.  Kondoh: Mater. Trans. 59 (2018) 61–65. doi:10.2320/matertrans.Y-M2017842
• 24)   T.  Sekiguchi,  K.  Ono,  H.  Fujiwara and  K.  Ameyama: Mater. Trans. 51 (2010) 39–45. doi:10.2320/matertrans.MB200913
• 25)   M.  Ota,  S.K.  Vajpai,  R.  Imao,  K.  Kurokawa and  K.  Ameyama: Mater. Trans. 56 (2015) 154–159. doi:10.2320/matertrans.M2014280
• 26)   H.  Fujiwara,  T.  Sekiguchi and  K.  Ameyama: Int. J. Mater. Res. 100 (2009) 796–799. doi:10.3139/146.110116
• 27)   S.  Kikuchi,  Y.  Hayami,  T.  Ishiguri,  B.  Guennec,  A.  Ueno,  M.  Ota and  K.  Ameyama: Mater. Sci. Eng. A 687 (2017) 269–275. doi:10.1016/j.msea.2017.01.076
• 28)   S.  Kikuchi,  H.  Akebono,  A.  Ueno and  K.  Ameyama: Powder Technol. 330 (2018) 349–356. doi:10.1016/j.powtec.2018.02.047
• 29)   S.  Kikuchi,  S.  Suzuki and  H.  Akebono: Mater. Trans. 61 (2020) 2284–2291. doi:10.2320/matertrans.Z-M2020854
• 30)   Y.S.  Zhang,  Y.H.  Zhao,  W.  Zhang,  J.W.  Lu,  J.J.  Hu,  W.T.  Huo and  P.X.  Zhang: Sci. Rep. 7 (2017) 40039. doi:10.1038/srep40039
• 31)   K.  Fujita,  M.  Ijiri,  Y.  Inoue and  S.  Kikuchi: Adv. Mater. 33 (2021) 2008298. doi:10.1002/adma.202008298
• 32)   H.  Hirai,  H.  Kurita,  S.  Gourdet,  K.  Fujita,  K.  Nakazawa and  S.  Kikuchi: Eng. Fract. Mech. 238 (2020) 107284. doi:10.1016/j.engfracmech.2020.107284
• 33)  J.C. Newman and I.S. Raju: Stress-intensity factor equations for cracks in three-dimensional finite bodies subjected to tension and bending loads, (NASA Langley Res. Ctr., Hampton, VA, 1984) (NASA Technical Memorandum 85793).
• 34)   S.  Kikuchi,  H.  Kubozono,  Y.  Nukui,  Y.  Nakai,  A.  Ueno,  M.O.  Kawabata and  K.  Ameyama: Mater. Sci. Eng. A 711 (2018) 29–36. doi:10.1016/j.msea.2017.11.010
• 35)   S.  Kikuchi,  T.  Imai,  H.  Kubozono,  Y.  Nakai,  M.  Ota,  A.  Ueno and  K.  Ameyama: Int. J. Fatigue 92 (2016) 616–622. doi:10.1016/j.ijfatigue.2016.02.038
• 36)   R.K.  Nalla,  B.L.  Boyce,  J.P.  Campbell,  J.O.  Peters and  R.O.  Ritchie: Metall. Mater. Trans. A 33 (2002) 899–918. doi:10.1007/s11661-002-0160-z
• 37)   K.  Osaki,  S.  Kikuchi,  Y.  Nakai,  M.O.  Kawabata and  K.  Ameyama: Mater. Sci. Eng. A 773 (2020) 138892. doi:10.1016/j.msea.2019.138892
• 38)   S.  Kikuchi,  K.  Takemura,  Y.  Hayami,  A.  Ueno and  K.  Ameyama: J. Soc. Mater. Sci. Jpn. 64 (2015) 880–886. doi:10.2472/jsms.64.880
• 39)   B.  Guennec,  T.  Ishiguri,  M.O.  Kawabata,  S.  Kikuchi,  A.  Ueno and  K.  Ameyama: Metals 10 (2020) 636–657. doi:10.3390/met10050636
• 40)   Y.  Nukui,  H.  Kubozono,  S.  Kikuchi,  Y.  Nakai,  A.  Ueno,  M.O.  Kawabata and  K.  Ameyama: Mater. Sci. Eng. A 716 (2018) 228–234. doi:10.1016/j.msea.2018.01.054
• 41)   Y.  Uematsu,  C.A.  Huang,  T.  Kakiuchi,  T.  Mizutani and  M.  Nakajima: Fatigue Fract. Eng. Mater. Struct. 43 (2020) 2800–2811. doi:10.1111/ffe.13292
• 42)   S.  Kikuchi,  Y.  Nakamura,  K.  Nambu and  T.  Akahori: Int. J. Automat. Technol. 11 (2017) 915–924. doi:10.20965/ijat.2017.p0915
• 43)  T. Sakai (Chair of Editorial Committee) et al.: Standard evaluation method of fatigue reliability for metallic materials —standard regression method of S-N curves—, JSMS-SD-11-07, (The Society of Materials Science, Japan, 2007).