MATERIALS TRANSACTIONS
Online ISSN : 1347-5320
Print ISSN : 1345-9678
ISSN-L : 1345-9678
Effect of Heat Treatment and the Fabrication Process on Mechanical Properties of Zr-14Nb Alloy
Ryota KondoNaoyuki NomuraHisashi DoiHiroaki MatsumotoYusuke TsutsumiTakao Hanawa
Author information
JOURNAL FREE ACCESS FULL-TEXT HTML

2016 Volume 57 Issue 12 Pages 2060-2064

Details
Abstract

The microstructure and mechanical properties of thermomechanically processed Zr-14Nb alloy with low magnetic susceptibility were investigated in this study. The stress-strain curves of Zr-14Nb alloy were classified as higher or lower work-hardening types depending on the thermomechanical process. Serration was observed in curves of the higher work-hardening type. SEM-EBSD analysis indicated that {332}<113> twinning should form at the bands that appear after tensile testing. On the other hand, no serration appeared in the lower hardening rate curves. The mechanical properties of centrifugally cast Zr-14Nb alloy with a lower work-hardening rate possessed moderate tensile strength and high elongation because of the formation of the isothermal ω phase during cooling after centrifugal casting, although the tensile strength was increased and the elongation was decreased with increase of the isothermal ω phase formed after aging. The magnetic susceptibilities of thermomechanically processed Zr-14Nb alloys were still one-third lower than those of Ti-6Al-7Nb. Accordingly, the mechanical properties of Zr-14Nb can be controlled by thermomechanical processing while keeping low magnetic susceptibility.

1. Introduction

Magnetic resonance imaging (MRI) has been widely used to acquire an arbitrary cross-sectional view of the human body. Diagnosis by MRI is non-invasive and is especially used in orthopedics, neurosurgery, and dentistry. MRI diagnostics are usually inhibited when metals are implanted in the body because of the displacement of the implant and/or the generation of artifacts on the images.13) Artifacts are generated by an inhomogeneous magnetic field because the inhomogeneity hampers the detection of MR signals and creates obscure areas in the MR images.46)

To overcome the problem of artifacts, the magnetic susceptibility difference between the implants and surrounding tissue should be decreased.7) From this point of view, Ti-Zr, Zr-Nb, and Zr-Mo alloys with low magnetic susceptibility have been developed, and their magnetic susceptibilities were reduced in Zr-Nb and Zr-Mo alloys up to about one-seventh that of the Co-Cr-Mo alloy and one-third that of Ti and Ti alloys.811) Among these Zr-based alloys, Zr-9Nb and Zr-3Mo possess low magnetic susceptibility due to the contribution of the ω phase. The magnetic susceptibility of the α, β, and ω phases in Zr-based alloys was found to be, from the least to greatest, χω < χα < χβ.8,9) The ω phase effectively decreases magnetic susceptibility and enhances mechanical strength.7,12)

From our previous research, centrifugally cast Zr-14Nb shows good balance between low magnetic susceptibility and appropriate mechanical properties in Zr-Nb binary alloy.12,13) To fabricate medical devices made of Zr-14Nb for MRI applications, thermomechanical processing, such as cold rolling and swaging, would be applied with various heat treatments. Therefore, the mechanical properties and magnetic susceptibility of thermomechanically processed Zr-Nb should be investigated. The ω phase is a key factor in controlling magnetic susceptibility and mechanical properties, so that aging was carried out for Zr-14Nb to form the isothermal ω phase. Therefore, the purpose of this study was to investigate the microstructure, mechanical properties, and magnetic susceptibility of thermomechanically processed Zr-14Nb alloy.

2. Experimental Procedure

2.1 Specimen preparation

Pure Zr buttons were prepared by arc-melting from Zr sponges (99.6mass%). Zr-14Nb was arc-melted as buttons from pure Zr buttons and Nb shots (99.9mass%) under an Ar atmosphere on a water-cooled copper hearth. The button was flipped and remelted at least 10 times to homogenize its composition. The button was encapsulated in a quartz tube in a vacuum of less than 10−3 Pa and solution-treated at 1173 K for 3.6 ks, followed by quenching in iced water and the simultaneous fracture of the tube (STQ). The STQed specimen was cold-rolled up to ca. 1.5 mm in thickness. After removing the oxide scales formed at quenching, the specimen was encapsulated and STQed again (hereafter abbreviated as CR-STQ). The CR-STQ was encapsulated again in a quartz tube and aged at 623 K for 120 s, 240 s, and 360 s (hereafter abbreviated as AG2, AG4, and AG6, respectively), followed by quenching in ice water. A temperature of 623 K was selected for heat treatment because the formation of the isothermal ω phase was confirmed in a range from 573 K to 673 K in Zr-12Nb and Zr-20Nb.14,15) An electrical discharge machine was used to cut dog-bone-type specimens with gauge dimensions of 3 mm × 1 mm × 18 mm for tensile testing. Specimen surfaces were mirror polished with #1000 SiC emery paper, a 9 μm diamond suspension, and a 0.04 μm colloidal silica suspension. A centrifugally cast Zr-14Nb and water quenched the cast Zr-14Nb were prepared for comparison (hereafter abbreviated as CC and CC-STQ, respectively). The fabrication method of the cast alloy was the same as our previous work.12)

2.2 Microstructural characterization

The microstructures of variously processed Zr-14Nb were observed at the gauge part of tensile test specimens. Against the tensile direction, the normal direction of the observed plane for CR-STQ, AG2, AG4, and AG6 was perpendicular, while that for CC and CC-STQ was parallel. The constituent phase was analyzed using X-ray diffraction (XRD) with CuKα radiation under 40 kV and 40 mA. The microstructure was characterized using an optical microscope (OM), a scanning electron microscope with an electron backscattering diffraction analyzer (SEM-EBSD). Specimens for the above-described observations were mirror polished and electrically polished in a solution of 5% HClO4 and 95% CH3OH with 26 V at 223 K for ca. 10 s.

2.3 Evaluation of magnetic susceptibility and mechanical properties

Magnetic susceptibility was measured using a magnetic susceptibility balance with a magnetic field of 0.35 T at room temperature. The magnetic susceptibility was measured six times for each specimen. A uniaxial tensile test was performed at an initial strain rate of 1.1 × 10−3 s−1 using an Instron-type universal testing machine at room temperature. The strain was measured using a non-contact optical strain gauge. The Vickers hardness was measured at least 10 times with a load of 4.9 N for 30 s.

3. Results

Typical stress-strain curves and mechanical properties of CR-STQ, AG2, AG4, AG6, CC, and CC-STQ are shown in Fig. 1 and Table 1, respectively. The ultimate tensile strength (UTS) increased and elongation decreased as aging time increased (AG2, AG4, and AG6).

Fig. 1

Stress-strain curves of variously processed Zr-14Nbs: (a) CR-STQ, (b) AG2, (c) AG4, (d) AG6, (e) CA, and (f) CA-STQ.

Table 1 Mechanical properties of variously processed Zr-14Nbs.
  0.2% proof
strength, σ0.2/MPa
Ultimate tensile
strength, σUTS/MPa
Elongation,
δ (%)
CR-STQ 575 770 13
AG2 580 750 7
AG4 785 810 7.5
AG6 870 925 1.5
CC 685 780 12
CC-STQ 600 790 13

The strain-hardening exponents of CR-STQ, AG2, and CC-STQ are 0.21, 0.17, and 0.19, respectively, and those of AG4, AG6, and CC are 0.05, 0.07, and 0.09, respectively. From the strain-hardening exponent, these stress-strain curves can be divided into two types: a higher work-hardening type (CR-STQ, AG2, and CC-STQ) and a lower work-hardening type (AG4, AG6, and CC). Serration was observed in the curves of CR-STQ and AG2. In addition, a distinctive sound was heard during these tensile tests.

Figure 2 shows OM images of CR-STQ and AG4 before tensile testing and after plastic deformation at 1.5%. The same area, which was marked by a Vickers indenter, was observed before and after deformation. Needle-like structures were observed in CR-STQ and AG in both cases. Similar needle-like structures were observed in β-type Zr-Nb alloys,16) although the crystal structure and mechanical properties of the needle are still obscure12,17) and under investigation. After the deformation, band-like structures were observed in CR-STQ (white arrows). On the other hand, there was no clear difference before and after plastic deformation in AG4. Figure 3 shows OM images of CC and CC-STQ before tensile testing and after plastic deformation at 1.5%. The needle-like structures were also observed in CC and CC-STQ. However, the bands were observed in CC-STQ (white arrows), as observed in CR-STQ. These results suggest that a band-like structure should form during plastic deformation in CC-STQ.

Fig. 2

Optical micrographs of non-deformed (a) CR-STQ and (b) AG4 and 1.5% plastic-deformed (c) CR-STQ and (d) AG4.

Fig. 3

Optical micrographs of non-deformed (a) CC and (b) CC-STQ and 1.5% plastic-deformed (c) CC and (d) CC-STQ.

Figure 4 shows the XRD profiles of CR-STQ, AG2, AG4, AG6, CC, and CC-STQ. The lattice parameters and site in ω phase could be changed with increase of Nb concentration in quenched Zr-Nb alloy. Two types of theoretical peak patterns of ω phase (ω-Zr*1 and ω-Zr*2) were also shown in Fig. 4. The peaks of the ω-Zr*2 were in good agreement with those appeared in Zr-14Nb.18) Peaks originating from the β and ω phases were observed in all specimens. The intensity from (0002) and ($11\bar{2}2$) of the ω phase was slightly increased as aging time increased, suggesting that the amount of the ω phase increased with the aging time. When microstructure of CR-STQ was compared to that of CC-STQ (Figs. 2 and 3), the densities and distributions of needles were different from each other. Also, peak intensities and angles of $11\bar{2}1$ and $11\bar{2}2$ originating from ω phase were different between CR-STQ and CC-STQ, as shown in Fig. 4(a) and (f). According to previous our work, the needles were changed to the spheroids with decrease of ω phase.12) However, further investigation about the needles should be needed to clarify the relationship between the needles and ω phase formation.

Fig. 4

XRD profiles of variously-processed Zr-14Nbs: (a) CR-STQ, (b) AG2, (c) AG4, (d) AG6, (e) CC, and (f) CC-STQ with theoretical peak patterns of β-Zr, and two ω-Zr. *1 was calculated by lattice parameter of a = 0.5039 nm and c = 0.3136 nm and site of (1/3, 2/3, 1/2) and (2/3, 1/3, 1/2) and *2 was calculated by lattice parameter of a = 0.5001 nm and c = 0.3085 nm and site of (1/3, 2/3, 0.4433) and (2/3, 1/3, 0.5567).18)

Figure 5 shows the Vickers hardness of CR-STQ, AG2, AG4, AG6, CC, and CC-STQ. The hardness values of CR-STQ, AG2, and CC-STQ were almost identical. However, the hardness of aged specimens increased with the aging time. This hardness tendency was similar to that of the UTS.

Fig. 5

Vickers hardness of variously processed Zr-14Nbs.

Figure 6 shows the SEM-EBSD analysis of non-deformed and 1.5% plastically deformed CR-STQ. From the image quality (IQ) map and the inverse pole figure (IPF) map, the needle-like and band-like structures were determined to be β-Zr. The bands in CR-STQ after 1.5% plastic deformation orientated to the <110> direction normal to the specimen surface, which is the same crystallographic orientation of the matrix.

Fig. 6

EBSD images of non-deformed CR-STQ: (a) SE image, (b) IQ map, and (c) IPF map; and 1.5% plastically deformed CR-STQ: (d) SE image, (e) IQ map, and (e) IPF map.

Figure 7 shows misorientation angles between point S and the points on the lines 1 and 2. The IPF maps of (a) and (b) are the same as those in Fig. 6. Misorientation between the needles and matrix was not observed. However, misorientation between the bands and matrix was confirmed, and the angle was about 50°, indicating that the direction in the bands should have an angle of about 50° against <110> of β-Zr.

Fig. 7

Misorientation angles from point S in <110> β grain.

Figure 8 shows the mass magnetic susceptibilities of CR-STQ, AG2, AG4, AG6, CC, and CC-STQ. These specimens still possessed low magnetic susceptibilities below one-third of that of Ti-6Al-7Nb.9) Since Zr-14Nb dominantly constituted with β phase and small amount of ω phase, the difference of magnetic susceptibilities could be small among them. This result indicates that variously processed Zr-14Nb in this study will suppresses artifacts in MRI.

Fig. 8

Mass magnetic susceptibility of variously processed Zr-14Nb and Ti-6Al-7Nb.

4. Discussion

As shown in Fig. 1, higher and lower work-hardening types of stress-strain curves were observed, depending on the fabrication process and heat treatment. The higher work-hardening type was characterized by serration in stress-strain curves, distinct sounds during tensile testing, and the formation of band-like structures. These characteristic behaviors were reported to occur in the process of mechanical twins.19) {112}<111> and {332}<113> mechanical twinning were observed in bcc, and these angles of orientation were 70.90° and 50.48° along the [110] direction.2023) The angle obtained in this study is in good agreement with the latter case, so that the bands would be formed as a result of {332}<113> mechanical twinning.

On the other hand, stress-strain curves showing lower work hardening were confirmed without the bands in AG4, AG6, and CC. From the XRD profiles, the ω phase was observed in AG4 and AG6 (Fig. 4). Therefore, the twinning was suppressed as a result of the formation of the ω phase, which is presumed to be the isothermal ω phase. Changes in stress-strain curves in accordance with several heat treatments were reported in Ti-15Mo, which is in good agreement with the present study.24) The stress-strain curve of STQed Ti-14Mo was of the higher work-hardening type, and the curve changed to the lower work-hardening type after isothermal ω phase formation with appropriate heat treatment because twin formation was suppressed and the {112}<111> slip was dominantly activated.24) The {332}<113> mechanical twinning was observed in metastable β-Ti alloys, and, eventually, the stress-induced ω phase formed after the twins were formed.25,26) The phenomena are also in good agreement with those of STQed Zr-14Nb. Therefore, the stress-strain curves of Zr-14Nb also changed from a higher to a lower work-hardening type by the suppression of twin formation and by the formation of the isothermal ω phase. It was reported that the formation of the twin and strain-induced ω phase in Ti alloys was suppressed by the increased oxygen concentration.24) Centrifugally cast Zr-14Nb may contain a higher oxygen content, as compared to the Zr-14Nb arc-melted water-cooled copper hearth because the molten alloy was cast and cooled in a mold mainly consisting of alumina. The oxygen content of centrifugally cast Zr-Nb alloys ranged from 0.08 to 0.16mass%.12) However, the stress-strain curve of CC-STQ was of a higher work-hardening type. Therefore, the amount of oxygen in CC and CC-STQ could be less than that suppressing twin formation. A possible reason a lower work-hardening type is shown for CC is the formation of the isothermal ω phase during cooling after centrifugal casting. For the above-stated reasons, the microstructure of CC was similar to that of AG4, coincidentally. However, the reason why CC containing isothermal ω phase showed high elongation and CC-STQ with no obvious serration exhibited higher work hardening should be clarified by intensive study in future.

5. Conclusion

Low magnetic Zr-14Nb was thermomechanically processed, and its microstructure and tensile and magnetic properties were investigated. Two types of stress-strain curves, showing higher and lower work-hardening exponents, were observed in the stress-strain curves. In the case of higher work-hardening exponents of CR-STQ, AG2, and CC-STQ, the band structure was confirmed after tensile testing. SEM-EBSD analysis indicated the existence of {332}<113> mechanical twins at the bands. However, in the case of lower work-hardening rates of AG4, AG6, and CC, the formation of the isothermal ω phase was suggested during aging or cooling after centrifugal casting.

Eventually, moderate strength and high elongation were obtained at CC among the conducted thermomechanical conditions. The magnetic susceptibilities of thermomechanically processed Zr-14Nb kept low magnetic susceptibility that was comparable to that of Ti-6Al-7Nb. The mechanical properties of Zr-14Nb can change, depending on the heat treatment and the fabrication process; thus, Zr-14Nb could be a promising alloy for use in medical devices for MRI.

Acknowledgments

This work was partially supported by a Grant-in-Aid for Fundamental Scientific Research (Kiban B: Nos. 22360287 and 15H04140) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. This work was also partially supported by an S-innovation grant from the Japan Agency of Medical Research and Development.

REFERENCES
 
© 2016 The Japan Institute of Metals and Materials
feedback
Top