Microstructure, mechanical and biological ... - Wiley Online Library

198

DOI 10.1002/mawe.201300113

Mat.-wiss. u. Werkstofftech. 2013, 44, No. 2–3

Microstructure, mechanical and biological properties of zirconium alloyed with niobium after severe plastic deformation Mikrostruktur, mechanische und biologische Eigenschaften von Niob-legiertem Zirkonium nach intensiver plastischer Verformung Y.P. Sharkeev1,2, A.Y. Eroshenko1, K.S. Kulyashova1, S.V. Fortuna3, K.A. Suvorov1, M. Epple4, O. Prymak4, V. Sokolova4, S. Chernousova4 A comparative investigation of microstructure, mechanical and biological properties for zirconium alloyed with niobium in coarse-grained and ultra-fine grained states is presented. The temperature and deformation regimes of multi-stage abc-pressing resulted in ultra-fine grained states with an average size of the structural elements in the range of 0.28 – 0.55 lm, depending on the accumulated strain during pressing. The increase of the accumulated strain at each stage of pressing increased the uniformity of the structure. The microhardness increased by 50% with increased accumulated strain during the severe plastic deformation. Between the microhardness and the average size of the structural elements, a linear dependence was found, indicating a Hall-Petch relationship. The alloy had a good biocompatibility as shown by an MTT test with osteoblasts (MG-63 cell line). The good mechanical properties (microhardness) of zirconium alloyed with niobium in the ultra-fine grained state make it suitable for medical applications, e. g. as implant material. Keywords: plastic deformation / microstructure / mechanical properties / biomaterials / alloys /

Eine vergleichende Untersuchung der Mikrostruktur sowie der mechanischen und biologischen Eigenschaften einer Zirkonium-Niob-Legierung im grobkrnigen und ultrafeinkrnigen Zustand werden vorgestellt. Durch kombinierte Wrmebehandlung und Deformation beim mehrstufigen ABC-Pressen wurde ein ultrafeinkrniges Material mit durchschnittlichen Abmessungen der Strukturbestandteile von 0,28 – 0,55 lm erhalten, je nach akkumulierter mechanischer Spannung. Die Zunahme der akkumulierten mechanischen Spannung bei jedem Pressschritt erhhte die Einheitlichkeit der Struktur. Die Mikrohrte nahm um 50% mit zunehmender akkumulierter Pressspannung zu. Ein linearer Zusammenhang zwischen der Mikrohrte und der Grße der Strukturelemente spricht fr eine Hall-Petch-Beziehung. Die gute biologische Vertrglichkeit wurde durch einen MTT-Test mit Osteoblasten (MG-63 Zelllinie) gezeigt. Die guten mechanischen Eigenschaften (insbesondere die Mikrohrte) sprechen fr eine gute Anwendbarkeit der Zirkonium-Niob-Legierung im ultrafeinkrnigen Zustand, z.B. als Implantatmaterial. Schlsselwrter: Plastische Verformung / Mikrostruktur / mechanische Eigenschaften / Biomaterialien / Legierungen /

1

Institute of Strength Physics and Materials Science of SB RAS, Tomsk, Russia 2 Scientific Educational Center “Biocompatible Materials and Bioengineering” at ISPMS of the Siberian Branch of the Russian Academy of Sciences, NR TPU and SSMU, Tomsk, Russia 3 Public company “Siberian Chemical Combine”, Tomsk region, Seversk, Russia 4 Institute of Inorganic Chemistry and Center for Nanointegration Duisburg-Essen (CeNIDE), Duisburg-Essen University, Essen, Germany

i

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Corresponding author: Y.P. Sharkeev, Institute of Strength Physics and Materials Science of SB RAS, 2/1 Academicheskii pr., 634021, Tomsk, Scientific Educational Center “Biocompatible Materials and Bioengineering” at ISPMS of the Siberian Branch of the Russian Academy of Sciences, NR TPU and SSMU, Tomsk, Russia E-mail: [email protected]

www.wiley-vch.de/home/muw


Microstructure, mechanical and biological properties of zirconium alloyed

1 Introduction Zirconium and its alloys have many applications in materials science, including nuclear materials [1]. Metallic implants and surgical tools are often made of niobium-alloyed zirconium alloys, besides stainless steel, titanium and titanium alloys [2]. The reason is the high corrosion resistance of titanium, zirconium and its alloys due to rapid surface passivation by the corresponding oxides, i. e. titania or zirconia [3]. To improve the mechanical properties of metals and alloys such as microhardness, yield strength, and ultimate strength, severe plastic deformation methods like equal-channel angular pressing, high pressure torsion, multiple forging or pressing (abc-pressing) are applied [4]. A nanostructured or ultra-fine grained state is formed in the bulk of a metal block under severe plastic deformation. As a rule, the nanostructured or ultra-fine grained state in metals and alloys is achieved by combining severe plastic deformation with other methods, e. g. rolling and/or heat treatment [4–7]. In general, the formed structure and its stability depend on the method used and on the temperature and deformation conditions. The investigation of the microstructure and the mechanical and biological properties of zirconium alloyed with niobium in the ultra-fine grained state, obtained by severe plastic deformation, are presented here.

2 Materials and methods Zirconium alloyed with niobium (produced in Russia) with the composition (by wt%) Zr 96.54, Nb 1.01, Si 0.48, Mo 0.32, W 0.1, Fe 0.29, Ti 0.88 was used. Before mechanical treatment, the zirconium-niobium blocks were annealed in vacuum at 853 K for 3 h. The blocks were rectangular-shaped with a dimension of 23.23.38 mm3. The ultra-fine grained state in the zirconiumniobium blocks was obtained by abc-pressing in air [7]. Three cycle abc-pressing of the blocks was performed on a hydraulic press MIS-6000K (Russia) at temperatures of 773, 723, and 673 K. Each cycle at a preset temperature included double or triple pressing with a change in the deformation axis by turning the block around the longitudinal axis by 90 8. The oxidized surface layer was mechanically removed after every pressing step. The strain rate of the blocks during the pressing process was 103 s1. The total strain at each pressing never exceeded 50%. The total number of pressings was 3, 5 and 9, and the accumulated logarithmic strain was e = 1.5, 2.6, and 4.6 respectively. The total strain e was calculated as the sum of the natural logarithms of initial and final block height ratios at each step of pressing: e¼

n X i¼1

ei ¼

n X i¼1

ln

hi ; h0i

ð1Þ

where hoi is height of the block before the ith pressing step, and hi is the height of the block after the ith pressing step. TEM thin film specimens were prepared as follows. An electric spark device was used to cut the 0.3 mm thick sheets from blocks, which were then mechanically ground to a thickness of 0.1 mm. The sheets were oriented perpendicular to the axes of the 3rd, 5th and 9th pressing. The final TEM thin foils were prepared with an electropolisher, using a mixture of 80% acetic acid

i


(CH3COOH) and 20% perchloric acid (HClO4) solution. The electropolishing voltage was 30 V. The temperature of electrolyte was 273 K. The microstructure was analyzed, and the phases were identified with a JEM-2100 transmission electron microscope at 200 kV. The measurement of the structural size elements, i. e. grains, subgrains and fragments, was carried out using TEM micrographs and optical microscopy; the method of secant lines was used [8]. To identify the zirconium and niobium phases, microdiffraction analysis was carried out using TEM dark-field images [9]. The Vickers microhardness was measured on the surface of 0.3 mm thick sheets, polished with a PMT-3M microhardness tester (Russia) at an indenter load of 0.98 N. X-ray powder diffraction was performed with a Bruker D8 Discover diffractometer. The measurements of intensity profiles were carried out with monochromatic CuKa-radiation. The preferred orientation was evaluated by the texture coefficient (TC) for each (hkl) reflection according to the following formula [10]: TChkl ¼

p c Ihkl =Ihkl ; X 1 p c ðIhkl =Ihkl Þ n

ð2Þ

where TChkl is the probability intensity of the (hkl) plane being perpendicular to the axis of the last deformation direction; Ihkl is the diffraction intensity of the (hkl) peak in the XRD pattern, Ihkl = Hhkl/Hhklmax; Hhkl is the height of the (hkl) peak in the XRD pattern; Hhklmax is the height of the maximum intensity peak in the XRD pattern; n is the number of reflections used in the calculations, c is the index for zirconium; and p is the index for the nontextured standard zirconium sample. Peaks located between 2 = 30–90 8 in the XRD patterns were chosen to calculate TC. The dataset for zirconium of the JCPDS card # 05-0665 was used as non-textured standard. It was assumed that alloying with niobium did not affect the preferred orientation. Two different zirconium-niobium surfaces were used for biological testing, i. e. a polished surface and a sandblasted surface. The samples were sequentially polished with silicon-carbide paper with 120, 480, 600, 1200 grid, respectively. Then, the samples were ultrasonically cleaned for 10 min in distilled water (Elmasonic, Germany). The sandblasted samples were treated similarly and then blasted with corundum particles (Al2O3) of 250–380 lm (Averon, Russia). Finally, the samples were mechanically and ultrasonically cleaned in distilled water and ethanol, respectively. After ultrasonic cleaning, polished and sandblasted zirconium-niobium samples were sterilized by heat treatment at 453 K for 60 min. MG-63 human osteoblast-like cells were cultured in DMEM cell culture medium supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, 100 U mL–1 penicillin and 100 mg mL–1 streptomycin (Life Technology, Germany), 5% CO2 and subcultivated according to standard cell culture protocols. 12 h before the incubation experiments, the cells were trypsinized and seeded in 24 well plates with 5 N 103 cells per well in 0.5 mL DMEM with FCS. The cell viability was analyzed by an MTT assay after 10 days of incubation. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; Sigma, Taufkirchen, Germany) was dissolved in PBS (5 mg mL–1) and then 50 lL of MTT solution was added to the cells and incubated for 1 h at 37 8C under 5% CO2 in humidi-


199

200

Y.P. Sharkeev et al.


Figure 1. Microstructure of zirconium-niobium in the initial state: (a) optical image, (b) bright-field transmission electron microscope micrograph with SAED pattern: (c) histogram of the grains on sizes. Bild 1. Mikrostruktur der Zirkonium-Niob-Legierung im Ausgangszustand: Optische Mikroskopie (a); Hellfeld-Transmissionselektronenmikroskopie mit Elektronenbeugungsmuster (b) und Histogramm der Korngrßenverteilung (c).

fied atmosphere. During incubation, the cells metabolize MTT to a dark blue formazane salt which is soluble in organic solvents. 300 lL DMSO were added to the cells. After 30 min, a 100 lL aliquot was taken for spectrophotometric analysis with a Multiscan FC instrument (Thermo Fisher scientific, Vantaa, Finland) at k = 570 nm. The absorption of cells on the tested surface was normalized to that of control cells on a glass surface, thereby indicating the relative level of cell death.

3 Results and discussion Fig. 1 shows optical and transmission electron micrographs of the zirconium-niobium microstructure in the initial recrystallized state, i. e. after annealing at 853 K for 3 h, and a histogram of the grain size distribution. In the structure, the majority (65%) of the grains had a size between 2 and 3 lm. However, 10% of the grains were smaller than 2 lm and 25% of the grains were larger than 3 lm. The distribution was single-modal with an average grain size of 2.8 lm. According to electron microscopy, the microstructure of zirconium-niobium in the initial state consisted of a-Zr equiaxed grains and niobium precipitates located at the grain boundaries and within the grains. Selected area electron diffraction (SAED) showed reflections of a-Zr and b-Zr. The average size of the niobium precipitates was 0.4 lm. During severe plastic deformation, the grain size decreased, subgrains were formed, and fragmentation occurred, both leading to metal hardening. The average size of the zirconium grains after severe plastic deformation was approximated as the average size of all structural elements (grains, subgrains and fragments). Fig. 2 shows the microstructure and the histograms of the grain size distribution of zirconium-niobium subjected to abc-pressing up to deformations of e = 1.5, 2.6, and 4.6 after 3, 5, and 9 pressings, respectively.

i


Fig. 2a shows a TEM bright-field image with a SAED pattern of zirconium-niobium after the first cycle of abc-pressing, including triple-step pressing. The processing temperature was 773 K, and the accumulated strain was e = 1.5. The bright-field image shows elongated grains. Inside the deformed grains, there were small subgrains with poorly discernible boundaries. Extinction contours were observed in bright-field images (marked by arrows), indicating a high level of residual stress formed in the material during severe plastic deformation [9]. The average size of the structural elements was 0.6 lm, which corresponds to an ultra-fine grained state according to ref. [11]. Moreover, the bright-field images showed niobium precipitates. Elemental microanalysis performed by energy dispersive X-ray spectrometry confirmed the presence of niobium precipitates, Fig. 3. The average grain size of the niobium precipitates during this stage decreased to 0.28 lm in comparison with the initial state, but the concentration of the niobium particles did not change. This may indicate a partial dissolution of the niobium precipitates under severe plastic deformation. The basic features of zirconium structure did not change after five pressing steps, Fig. 2c. The microstructure shows a decreasing average size of the extended subgrains and fragments. The average size after five pressings was 0.44 lm for the a-Zr matrix structural elements and 0.20 lm for the niobium precipitates. An increase of the number of pressing steps up to nine led to a further refinement of structure and the formation of a more homogeneous structure, Figs. 2 e, f. The average size of matrix structure elements was 0.28 lm, and the average size of niobium precipitates was 0.2 lm. The analysis of the SAED patterns after abc-pressing showed the reflections from the base phase a-Zr, but also from b-Zr and the Nb precipitates. Zirconium oxide, ZrO2, was also found. The a-Zr reflections had a high intensity while the Nb and ZrO2 reflections had a low intensity, indicating a low volume fraction




Figure 2. Evolution of the zirconium-niobium microstructure with the accumulated strain value: e = 1.5 (a, b), e = 2.6 (c, d), e = 4.6 (e, f, g); (a, c, e) bright-field TEM micrographs with SAED patterns; (f) dark field electron micrograph, extinction contours are indicated with arrows; (b, d, g) histograms of the size of the structural elements. Bild 2. Entwicklung der Mikrostruktur der Zirkonium-Niob-Legierung mit zunehmender akkumulierter Spannung: e = 1,5 (a, b), e = 2,6 (c, d), e = 4,6 (e, f, g); (a, c, e) HellfeldTransmissionselektronenmikroskopische Abbildungen mit SAED-Beugungsmustern; (f) Dunkelfeld-Transmissionselektronenmikroskopische Abbildung; die Extinktionsgrenzen sind durch Pfeile gekennzeichnet; (b, d, g) Histogramm der Korngrßenverteilung.

i



201

202



Figure 3. Bright-field micrographs of niobium precipitates (a, b) and the corresponding energy-dispersive X-ray spectrum in keV (c). Bild 3. Elektronenmikroskopische Abbildung (a, b) und energiedispersives Rntgenspektrum der Niob-Ausscheidungen in keV (c). Table 1. Texture coefficients of zirconium-niobium deformed by abcpressing. Tabelle 1. Texturkoeffizienten von Zirkonium-Niob nach dem abcPressen.

hkl

100 002 101 102 110 103 112

Figure 4. The dependence of the average size of the structural elements on the accumulated strain after abc-pressing. Bild 4. Abhngigkeit der mittleren Grße der Strukturelemente von der akkumulierten Spannung beim abc-Pressen.

of the Nb and ZrO2 phases. Zirconium oxide was formed by oxidation during the thermal and mechanical treatments of the alloy. Fig. 2b, 2d and 2g show the change of the size distribution of structural elements (grains, subgrain and fragments) with increasing accumulated strain. The initial state is characterized by an asymmetric grain size distribution with a maximum in the range of 2.0 – 2.5 lm, Fig. 1a. A small fraction of the grains was larger than 5 lm. The increase in accumulated strain led to a shift towards a smaller grain size and in general to a decrease of the size of the structural elements, Fig. 4. The strongest refinement of structure was observed after the first deformation steps. Fig. 5 shows the X-ray diffraction pattern of zirconiumniobium in the initial recrystallized state and after abc-pressing at different stages. The X-ray diffraction pattern of the initial state showed the peaks of a-Zr, Fig. 5a. The peaks of b-Zr and b-Nb were not detectable because of the low content of these phases. There was also a redistribution of peak intensities due to texture

i


Initial state

0.06 3.125 1.225 0.55 0.55 0.55 0.75

Accumulated strain 1.4

2.6

4.6

2.42 1.02 0.19

0.48 2.875 0.48 0.53 1.43 1.1 0.5

0.44 0.25 0.9 0.27 0.55 0.21 0.66

0.97 0.38

formation during severe plastic deformation as expressed by the texture coefficient (TC). A high value of the TC indicates a high degree of preferred orientation. Texture coefficients of zirconium-niobium deformed at different deformation steps are presented in Table 1. The multistep deformation during abc-pressing clearly changed the texture of the Zr blocks. The texture was less pronounced in the Zr block after nine abc-pressing steps, i. e. the structure became more isotropic. Fig. 6 shows the dependence of the zirconium-niobium microhardness on the accumulated strain after multistep pressing. The increase of microhardness after nine steps of abc-pressing in comparison to the initial state was about 600 MPa. Thus, after nine abc-pressing steps, the microhardness reached 2200 MPa. The value corresponds to the microhardness of titanium alloyed with oxygen (Grade 4). As mentioned above, in order to achieve higher levels of microhardness, and thus a yield strength and ultimate strength, it is necessary to use combined methods for severe plastic deformation, such as abc-pressing and rolling. Note that the zirconiumniobium alloy has good corrosion resistance and does not contain alloying elements which are harmful to the body, in contrast to the common biomedical alloy Ti6Al4V. The data on the average size of the structural elements and microhardness allow to obtain the relationship between the microhardness and the average size of the structural elements,




Figure 5. X-ray diffraction patterns of zirconium-niobium after abc-pressing: The accumulated strain e was 0 (a; initial state), 1.4 (b), 2.6 (c), and 4.6 (d). Bild 5. Rntgendiffraktogramm von Zirkonium-Niob nach dem abc-Pressen mit akkumulierten Spannungswerten e von 0 (a, Ausgangszustand), 1,4 (b, 2,6 (c) und 4,6 (d).

Figure 6. Dependence of the microhardness on the accumulated strain after abc-pressing. Bild 6. Die Abhngigkeit der Mikrohrte von der akkumulierten Spannung nach dem abc-Pressen.

Fig. 7. There is a linear dependence between the microhardness and the value of d–1/2 (d is the average size of the structural elements) which indicates a Hall-Petch relationship [12]. The HallPetch relationship gives a quantitative description of the growth of the yield strength or microhardness of a polycrystalline mate-

i


Figure 7. The relationship between microhardness and average size of the structural elements. Bild 7. Die Abhngigkeit der Mikrohrte von der Grße der Strukturelemente.

rial with decreasing grain size. It is based on the dislocation mechanisms of plastic deformation and the grain boundary barriers for dislocation movement.


203

204



tion. A linear dependence between microhardness and d–1/2 was found which indicates a Hall-Petch relationship. The MTT viability data showed a good biocompatibility of the zirconiumniobium samples.

Acknowledgments The work was partially supported by the Program of the Presidium of the Russian Academy of Sciences, project No. 5.27, the Russian Foundation for Basic Research, grant No. 12-03-00903-a, the Federal Program “Kadry”, contract No. 8036, and the German Federal Ministry of Education and Research, BMBF project No. RUS 11/024.

5 References Figure 8. MTT-test of zirconium-niobium after incubation with MG63 cells (osteoblasts) for 10 days. Control: glass (* p f 0.001, ** p f 0.05). Bild 8. MTT-Test von Zirkonium-Niob nach 10 Tagen Zellkultur (MG63 Zellen; Osteoblasten). Kontrolle: Glas (* p f 0,001, ** p f 0,05).

MTT data represents the viability of cells after 10 days culture. Glass samples were used as a control and the cells were proliferating well on the zirconium-niobium surfaces as shown by the MTT test, Fig. 8. The cell viability of the polished zirconiumniobium surface was increased compared with the sandblasted surface and the control sample. This suggests that polished zirconium-niobium is well suited for cell growth.

4 Conclusions Abc-pressing of niobium-alloyed zirconium blocks resulted in transformation of coarse-grained structure in the ultra-fine grained structure with an average size of the structural elements (grains, subgrains and fragments) in the range of 0.6–0.28 lm depending on the value of the accumulated strain. An increase in the number of pressing steps from three to nine did not change the character of the formed ultra-fine grained structure, but led to structural uniformity. The microhardness increased with the increase of the accumulated strain during severe plastic deforma-

i


[1] H.R. Higgy, F.H. Hammad, J. Nucl. Mater. 1972, 44, 215. [2] Ph. Hernigon, G. Mathieu, A. Poignard, O. Manisom, P. Filippini, A. Demoura, Eur. J. Orthop. Surg. Traumatol. 2007, 17, 243. [3] S.A. Brown, J.E. Lemons, Medical Applications of Titanium and its Alloys: the Material and Biological Issues, ASTM special technical publication, New York 1996. [4] R.Z. Valiev, J. Mater. Sci. 2007, 42, 1483. [5] V.M. Segal, Mater. Sci. Eng. A-Struct. 1999, 271, 322. [6] Y.R. Kolobov, R.Z. Valiev, G.P. Grabovetskaya, A.P. Zhilyaev, E.F. Dudarev, K.V. Ivanov, M.B. Ivanov, O.A. Kashin, E.V. Naydenkin, Grain Boundary Diffusion and Properties of Nanostructured Materials, Cambridge International Science Publishing, London 2007. [7] Y.P. Sharkeev, E.V. Legostaeva, A.Y. Eroshenko, I.A. Khlusov, O.A. Kashin, Compos. Interface 2009, 16, 535. [8] Standard Test Methods for Determining Average Grain Size Using Semiautomatic and Automatic Image Analysis. ASTM E1382 – 97 2010. [9] P.B. Hirsch, Electron Microscopy of Thin Crystals, Butterworth, London 1965. [10] A.J. Perry, Thin Solid Films 1986, 135, 73. [11] E.V. Kozlov, N.A. Koneva, L.I. Trishkina, A.N. Zhdanov, Russ. Metall. Metally. 2010, 4, 264. [12] J.T. Al-Haidary, N.J. Petch, E.R. Rios, Philos. Mag. A 1983, 47, 869. Received in final form: December 20th 2012 T 113
