Elsevier

Intermetallics

Volume 139, December 2021, 107349
Intermetallics

Mechanical property enhancement of the Ag–tailored Au–Cu–Al shape memory alloy via the ductile phase toughening

https://doi.org/10.1016/j.intermet.2021.107349Get rights and content

Highlights

  • Ductile phase toughening was practiced in the Au–Cu–Al–based alloys by Ag addition.

  • The Au–28Cu–16Al–28Ag was optimized by the brittle β phase and the ductile α1 phase.

  • Three–dimensional diagram of the quaternary Au–Cu–Al–Ag system was constructed.

  • The mechanical properties and shape memory effect of Au–Cu–Al–Ag were investigated.

  • A strategy for the fabrication of the Au–Cu–Al–Ag were shown in the first time.

Abstract

The Au–Cu–Al biocompatible shape memory alloys (SMAs) have attracted much attention due to the requirements of biomedical applications. However, brittleness remains a critical issue in the intermetallics of Au–Cu–Al alloys. In this study, to solve the dilemma of embrittlement, Ag was introduced into the Au–Cu–Al alloy to realize the ductile phase toughening (DPT) since the ductile disordered–fcc phase was predicted to precipitate in the brittle parent β phase based on the Ag–Au–Cu phase diagram. Microstructure observations, chemical composition analysis, crystal structure identifications, and thermal analysis revealed that the Au–28Cu–16Al–28Ag (mol.%) alloy was mainly composed of the Ag–rich α1 primary solidification phase with disordered–fcc structure and the eutectic structure of the α1 phase and the β phase with L21 structure. The ductile phase was successfully inserted into the brittle Au–Cu–Al intermetallic while the martensitic transformation temperature and crystal structures remained almost uninterrupted. In the cyclic loading–unloading tensile tests, the alloy mainly composed of the β phase failed in the early elastic region due to its embrittlement; while the alloy mainly composed of the α1 phase performed 29% in the total strain deformation. The Au–28Cu–16Al–28Ag alloy, which is composed of the β phase and the α1 phase, solve the issue of embrittlement; in addition, shape recovery was also found in this alloy during unloading. According to the microstructure observations, cyclic tensile tests, and fracture surface observations, the aforementioned improvement of strengthening and enhancement of ductility were brought from the insertion of the ductile α1 phase into the brittle β/β grain boundaries.

Introduction

Shape memory alloys (SMAs), which are widely utilized as sensors and actuators in these decades for their shape memory effect (SME) and/or superelastic (SE) behavior [1]. The SME is known to originate from the thermoelastic martensitic transformation, which is the diffusionless first–order phase transition from the high–temperature austenite phase to the low–temperature martensite phase [2]. The shape deformation and their shape recovery strain are manipulated via controlling the applied temperature and/or stress. Among the SMAs, Ti–Ni alloys (Nitinol) are conventional materials for practical uses in virtue of their large shape recovery strain [3]. However, there is an issue of hypersensitivity of Ni for the uses in the human body [[4], [5], [6]], the Au–based SMAs, which are considered as biocompatible alloys, were thus chosen in this work.

Among the Au–based SMAs, Au–Cu–Al alloys, promising SMAs, are known as one of the Hume–Rothery β electron compounds, whose crystal structure is determined by the electron–to–atom ratio (e/a) [7,8]. Au–Cu–Al SMAs have been employed as functional materials by taking advantage of their SME and SE, which correspond to B2 (cubic) – doubly extended B19 (orthorhombic) and/or B2 – doubly extended B19’ (monoclinic) martensitic transformation [9,10]. Besides, excellent biocompatibility [11], good X–ray contrast [12], and comparable magnetic susceptibility to the human body [13] (i.e., MRI artifacts–less) of Au–Cu–Al alloys are especially favored for the applications towards biomedical materials. Therefore, the Au–Cu–Al SMAs, which were considered superior to the aforementioned Nitinols for the human body uses, have become potential candidates.

Concerning the mechanical properties, micro–Vickers hardness [14], elastic modulus and nanoindentation hardness [15], and fracture strains of microwires of the Au–Cu–Al SMAs [16] have been reported. Besides the aforementioned mechanical related literature, the influences of the chemical composition on the martensitic transformation start temperature (Ms) and the mechanical properties of the Au–Cu–Al ternary SMAs, which have been studied by our group, are shown as follows [17]. This study revealed that Ms, ultimate tensile stress (σUTS), and fracture strain (εf) strongly depended on the Al concentration. It is crucial to manipulate the Ms of Au–Cu–Al SMAs to around body temperature to disclose its functionality (i.e., superelasticity) for practicing in biomedical applications for the human body. The alloys with high Al concentration (i.e., 20–25 mol.% Al) meet the aforementioned requirements of the manipulation of Ms; however, a high amount of Al ended up with fractures of these alloys during the elastic deformation stage. It has been reported that such embrittlement might be caused by the decrement of activated independent slip systems [18] and/or the increment of Al–Al bonds at the grain boundaries (GBs) [19,20] in our preliminary works. Solutions to the embrittlement of the functional Au–Cu–Al alloys have thus become a crucial issue among the Au–Cu–Al SMAs.

In order to alleviate its brittleness, ductile phase toughening (DPT) [21] is often put into practice. Ishida et al. [[22], [23], [24]] stated that the introduction of the ductile γ phase (A1) greatly improved the hot–workability and the ductility at room temperature (RT) in the case of the Ni–Al based alloys, which are also intermetallic compounds. They asserted that the enhancement was due to the GB toughening by the ductile γ phase. In addition to the Ni–Al based alloys, DPT was also executed on many intermetallic compounds, such as Ti–Al based alloys [25,26] and Nb–Si based alloys [27], and has been proven to be practical.

In order to apply the DPT to the Au–Cu–Al ternary system, the 4th element addition is necessary because Al was constrained at low concentration (i.e., the α phase at low Al regime) when the α phase and the β phase coexist according to the isothermal Au–Cu–Al ternary phase diagram at 773 K [28]. For solving the aforementioned dilemma, the introduction of Ag into the Au–Cu–Al ternary system was carried out in this study. Based on the ternary isothermal phase diagram of Ag–Au–Cu at 773 K [29], the disordered–fcc phase (α1 + α2), which was presumed to be a ductile phase, was found in a broad composition range. In addition, since the Au–Ag binary is an isomorphous system [30] and Ag is highly soluble in both Cu and Al [30], it is possible that precipitations of other phases, which might affect the Ms, could be limited. Apart from the solubility issue, the effects of the e/a ratio were also taken into accounts. According to the Hume–Rothery theory, Ag is defined as a monovalent metal ion, which is exactly the same as Au and Cu. Therefore, it is supposed that the crystal structure remained the same after the inclusion of Ag. According to the aforementioned inferences, the 4th element, Ag, was introduced for realizing the DPT.

In this study, Au–28 mol.% Cu–16 mol.% Al–28 mol.% Ag, which possessed medium Al concentration, was determined as the alloying composition. As stated in the previous sections, the ductile Ag–rich disordered–fcc α1 phase was assumed to be precipitated in the Au–28Cu–16Al–28Ag alloy, which was composed of 2 phases (i.e., brittle β phase + ductile α1 phase). The Al concentration in the β phase would be raised accordingly and the high Al concentration, which lowered the Ms, is predicted to be realized while the ductile α1 phase coexists with the β phase. The mechanical properties of the dual–phase alloy and each single–phase alloys have been clarified by the microstructures, compositional analysis, phase constituents, and phase diagrams. To sum up, the purpose of this work is to introduce the ductile phase into the Au–Cu–Al SMAs by the Ag addition to solve the embrittlement issue of the β phase. In addition, a strategy for the fabrication of the promising Au–Cu–Al–Ag SMAs was constructed for future works.

Section snippets

Experimental

The Au–28 mol.% Cu–16 mol.% Al–28 mol.% Ag alloy was fabricated from the as–received Au, Cu, Al, and Ag raw materials with their purities higher than 99.99%. The alloy is abbreviated to Au–28Cu–16Al–28Ag unless mentioned otherwise. In addition to the Au–28Cu–16Al–28Ag alloy, the Au–37Cu–22Al–4Ag alloy and the Au–4Cu–1Al–80Ag alloy, which were also manufactured and evaluated for the purpose of the comparisons among these 3 alloys. The reasons for the selection of these 2 more alloys are

Microstructures and phase constitutions

Fig. 1 shows the microstructures of the 3 alloys. The chemical compositions and the phase fractions of these alloys were listed in Table 1. Regarding the solidification path and the evolution of the microstructures, details are shown in section 3.6. In Fig. 1(a), the primary solidification phase (large dark dendritic phase pointed by a red arrow) and the eutectic matrix phase (lamellar fine dark phase and fine gray phase surrounded by green hollow square) showing the eutectic structure,

Conclusions

The effect of ductile phase toughening (DPT) was fulfilled by the introduction of Ag into the ternary Au–Cu–Al shape memory alloy. The Au–28Cu–16Al–28Ag alloy was chosen based on the three–dimensional diagram of the quaternary Au–Cu–Al–Ag system, which was constructed in this work. The enhancements of the mechanical properties and pseudoelastic effect of the Au–28Cu–16Al–28Ag alloy by taking the advantages of the precipitation of the ductile disordered–fcc Ag–rich α1 phase are elucidated in the

CRediT authorship contribution statement

Ayano Toriyabe: Conceptualization, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Visualization, Funding acquisition. Wan-Ting Chiu: Conceptualization, Writing – review & editing, Validation. Akira Umise: Methodology, Data curation, Supervision. Masaki Tahara: Supervision. Kenji Goto: Methodology, Supply of material. Hiroyasu Kanetaka: Conceptualization, Writing – review & editing. Takao Hanawa: Writing – review & editing. Hideki Hosoda:

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This study is supported by Japan Society for the Promotion of Science (JSPS) (KAKENHI 19J23030, KAKENHI 20K20544, and KAKENHI 18K13655), and MEXT funds of Research Center for Biomedical Engineering and Design & Engineering by Joint Inverse Innovation for Materials Architecture.

References (41)

Cited by (0)

View full text