Nanoarchitectured Co–Cr–Mo orthopedic implant alloys: Nitrogen-enhanced nanostructural evolution and its effect on phase stability

Abstract

Our previous studies indicate that nitrogen addition suppresses the athermal γ (face-centered cubic, fcc) → ε (hexagonal close-packed, hcp) martensitic transformation of biomedical Co–Cr–Mo alloys and ultimately offers large elongation to failure while maintaining high strength. In the present study, structural evolution and dislocation slip as an elementary process in the martensitic transformation in Co–Cr–Mo alloys were investigated to reveal the origin of their enhanced γ phase stability due to nitrogen addition. Alloy specimens with and without nitrogen addition were prepared. The N-doped alloys had a single-phase γ matrix, whereas the N-free alloys had a γ/ε duplex microstructure. Irrespective of the nitrogen content, dislocations frequently dissociated into Shockley partial dislocations with stacking faults. This indicates that nitrogen has little effect on the stability of the γ phase, which is also predicted by thermodynamic calculations. We discovered short-range ordering (SRO) or nanoscale Cr₂N precipitates in the γ matrix of the N-containing alloy specimens, and it was revealed that both SRO and nanoprecipitates function as obstacles to the glide of partial dislocations and consequently significantly affect the kinetics of the γ → ε martensitic transformation. Since the formation of ε martensite plays a crucial role in plastic deformation and wear behavior, the developed nanostructural modification associated with nitrogen addition must be a promising strategy for highly durable orthopedic implants.

Introduction

Co–Cr–Mo alloys with exceptional corrosion and wear resistances have been used in orthopedic implants such as artificial hip and knee joints [1], [2], [3], [4]. Although as-cast or hot-forged Co–Cr–Mo alloys are practically used for these applications, their strength and elongation to failure are sometimes low due to a coarse microstructure, solidification defects, precipitations, and so on. Thus, there has been a strong demand to improve the mechanical reliability of biomedical Co–Cr–Mo alloys to obtain orthopedic implants with long lifetimes. The strengthening by carbide precipitation has been commonly employed in the present alloys. Recent research, however, suggests that the hard precipitation phases (e.g. M₂₃C₆-type carbides) with sizes from 100 nm to several tens of micrometers are detrimental to wear performance, corrosion resistance and biocompatibility [3], [5], [6]. Therefore, much effort has gone into improving the mechanical properties of the Co–Cr–Mo alloys by alternative approaches such as the alloy design and optimization of the hot deformation process [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19].

Biomedical-grade Co–Cr–Mo alloys generally exhibit a phase transformation from a face-centered cubic (fcc) γ matrix to a hexagonal close-packed (hcp) ε martensite during quenching, plastic deformation and isothermal heat treatment [3], [10], [11], [12], [13], [19], [20]. In previous studies, we found that ε martensite has conflicting effects on the mechanical properties. It produces hardening as a result of the strain-induced martensitic transformation that enhances the wear properties of Co–Cr–Mo alloys [3], but ε martensite formation reduces the tensile elongation to failure and the cold workability [10], [11], [12]. These findings suggest that optimization of the martensitic transformation behavior is important for enhancing mechanical performance of the Co–Cr–Mo alloys. Nickel is known to stabilize γ phase by suppressing the martensitic transformation of cobalt and cobalt-based alloys; the Ni-containing alloys exhibit excellent elongation to failure [21]. Large amounts of Ni, which may cause allergies and cancer in living organisms [22], have been incorporated in the biomedical Co–Cr-based alloys when a high deformability is required (designated ASTM F90 and F562) [23], [24].

The addition of nitrogen, which can substitute for Ni, to the Co–Cr–Mo alloys is attractive from both practical and scientific viewpoints [11], [12], [13], [14], [15], [16], [17]. The authors have found that just a small amount of nitrogen can stabilize the γ phase by suppressing the athermal ε martensitic transformation on quenching [11], [12], [13], [14]. We have recently demonstrated that an alloy design that combines hot deformation with nitrogen addition dramatically improves mechanical properties [11], [13]. Fig. 1 shows the relationship between the 0.2% proof stress and tensile elongation of the newly developed N-doped Co–29Cr–6Mo (wt.%)¹ alloys processed by hot deformation, which is reproduced from Ref. [11]. Here, the results of conventionally prepared alloy specimens are also plotted in this figure. As shown in Fig. 1, the developed alloys (shown by green symbols) exhibit a superior strength–elongation to failure balance compared with other specimens. Since the hot-forged N-doped alloy has an ultrafine-grained structure [11], the strength of this specimen is extremely high. It should be noted that the alloys without nitrogen addition show a severe loss of elongation to failure with increasing the strength. Thus, nitrogen addition is a promising method for developing the high performance components in artificial joint replacement.

The improvement of mechanical properties of N-doped alloys is due to the N addition eliminating athermal ε martensite plates from the microstructure, which induce premature fracture [11], [12]. It is therefore important to clarify the underlying mechanisms for the enhanced γ phase stability on nitrogen doping. Li et al. [25] have recently performed three-dimensional atom probe tomography (3D-APT), which indicated the presence of short-range ordering (SRO) between Cr and N atoms in the γ matrix of a nitrogen-doped Co–29Cr–6Mo alloy. They speculated that this SRO increases the energy barrier for the γ → ε transition. However, the details have not been clearly elucidated.

It is well accepted that nitrogen in steels forms SRO [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37]. The SRO due to chemical affinity between alloying elements in steels has been intensively studied by various approaches such as electron structure calculations [26], [27], X-ray absorption fine structure spectroscopy [28], small-angle neutron scattering [29] and 3D-APT [30]. In addition, transmission electron microscopy (TEM) has provided indirect information about SRO by revealing a planar array of dislocations [31], [32]. There are only a few TEM studies that have directly imaged SRO domains in aged microstructures (e.g. Ref. [33]).

One of the most important aspects of the N-containing steels can be characterized by the plastic flow behavior [34], [35], [36], [37]. Flow stress is generally categorized into two components: thermal and athermal stresses (i.e., effective and internal stresses, respectively). The former depends on the temperature and the strain rate, whereas the latter is independent of both these parameters. Previous studies of high-N austenitic steels [34], [35], [36], [37] suggested that SRO enhances both the thermal and athermal components of flow stress.

In an fcc alloy with a low stacking fault energy (SFE), perfect dislocations with Burgers vectors a/2[$\overline{1}01$] tend to dissociate into Shockley partial dislocations bounding stacking faults (SFs) in the {1 1 1}_γ plane, in accordance with the following equation:

$$\frac{a}{2}[\overline{1}01]\to \frac{a}{6}[\overline{2}11]+\text{SF}+\frac{a}{6}[\overline{1}\overline{1}2]$$

where a is the lattice constant of the fcc phase. In addition, ε martensite develops due to regular overlapping of SFs in every second {1 1 1}_γ plane [38]. In other words, glide of Shockley partial dislocations can be considered an elementary step in this martensitic transformation. Thus, an in-depth understanding of SRO is necessary to elucidate not only strength/deformation mechanisms, but also the origin of γ phase stabilization on nitrogen addition. However, no correlations have been found among SRO, dislocation glide, and ultimately the γ → ε martensitic transformation for Co–Cr–Mo alloys.

In the present study, we investigated the nanostructures of hot-forged and subsequently heat-treated Co–29Cr–6Mo–N alloys by TEM for interpreting the origin of nitrogen-associated stabilization of γ phase. Compression tests at room temperature were performed at various strain rates and the relationship between the thermal activation of dislocation glide and nanostructure evolution is discussed.