Author's Accepted Manuscript

Influence of Si addition on the microstructure and mechanical properties of Ti-35Nb alloy for applications in orthopedic implants A.M.G. Tavares, W.S. Ramos, J.C.G. de Blas, E.S.N. Lopes, R. Caram, W.W. Batista, S.A. Souza www.elsevier.com/locate/jmbbm

PII: DOI: Reference:

S1751-6161(15)00242-8 http://dx.doi.org/10.1016/j.jmbbm.2015.06.035 JMBBM1531

To appear in: Journal of the Mechanical Behavior of Biomedical Materials

Received date:10 March 2015 Revised date: 22 June 2015 Accepted date: 26 June 2015 Cite this article as: A.M.G. Tavares, W.S. Ramos, J.C.G. de Blas, E.S.N. Lopes, R. Caram, W.W. Batista, S.A. Souza, Influence of Si addition on the microstructure and mechanical properties of Ti-35Nb alloy for applications in orthopedic implants, Journal of the Mechanical Behavior of Biomedical Materials, http://dx.doi.org/10.1016/j.jmbbm.2015.06.035 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Influence of Si addition on the microstructure and mechanical properties of Ti35Nb alloy for applications in orthopedic implants

A. M. G. Tavaresa, W. S. Ramosa, J. C. G. de Blasb, E. S. N. Lopesc, R. Caramc, W. W. Batistaa, S. A. Souzaa*, a

Department of Materials Science and Engineering, Federal University of Sergipe, 49100-000, São Cristóvão, SE, Brazil

b

Metallurgical Engineering Program - PEMM / COPPE, Federal University of Rio de Janeiro, C.P.68505, 21945970 Rio de Janeiro, RJ, Brazil

c

Department of Materials Engineering, Faculty of Mechanical Engineering, State University of Campinas, 13083-860 Campinas, SP, Brazil

* Corresponding author: Tel.: +55 79 21056972 E-mail address: [email protected] (S.A. Souza)

In the development of new materials for orthopedic implants, special attention has been given to Ti alloys that show biocompatible alloy elements and that are capable of reducing the elastic modulus. Accordingly, Ti-Nb-Si alloys show great potential for application. Thus, this is a study on the microstructures and properties of Ti-35Nb-xSi alloys (x = 0, 0.15, 0.35, 0.55) (wt. %) which were thermally treated and cooled under the following conditions: furnace cooling (FC), air cooling (AC), and water quenching (WQ). The results showed that Si addition is effective to reduce the density of omega precipitates making beta more stable, and to produce grain refinement. Silicides, referred as (Ti,Nb)3Si, were formed for alloys containing 0.55% Si, and its formation presumably occurred during the heating at 1000°C. In all cooling conditions, the hardness values increased with the increasing of Si content, as a result from the strong Si solid solution strengthening effect, while the elastic modulus underwent a continuous reduction due to the reduction of omega 1 

precipitates in beta matrix. Lower elastic moduli were observed in water-quenched alloys, which concentration of 0.15% Si was more effective in their reduction, with value around 65 GPa. Regarding Ti-35Nb-xSi alloys (x = 0, 0.15, 0.35), the "double yield point" phenomenon, which is typical of alloys with shape memory effect, was observed. The increase in Si concentration also produced an increase from 382 MPa to 540 MPa in the alloys' mechanical strength. Ti-35Nb-0.55Si alloy, however, showed brittle mechanical behavior which was related to the presence of silicides at the grain boundary. Keywords: titanium alloys, microstructure, mechanical properties

1. Introduction

Over the past few years, titanium and its alloys have become the main class of materials intended for biomedical applications, especially in dental and orthopedic fields, because of their higher biocompatibility in comparison to the main types of metallic biomaterials, such as stainless steel and Co-Cr alloys (Niinomi et al., 2012; Geetha et al., 2009). In terms of orthopedic applications, titanium alloys must also provide a unique combination of properties such as low elastic modulus, good mechanical strength, and excellent corrosion resistance caused by body fluid (Banerjee et al., 2005; Zhou et al., 2004; Song et al., 1999). However, the titanium alloys currently in use also exhibit a high elastic modulus (110 GPa) in comparison to that of the human bone elastic modulus (10-30 GPa), fact that may cause the stress-shielding phenomenon and result in the premature failure of the implant; This factor, in turn, is considered to be a major cause of revision surgeries (Zhou and Luo, 2011; Zhou and Niinomi, 2009; Zhentao and Lian, 2006; Hao et al., 2006). 2 

As a result, during the last decades, increasing attention has been given to the development of new type-ȕ titanium alloys, they are assumed to be possible substitutes for the most widely used alloy in orthopedic applications - Ti-6Al-4V alloy (Niinomi et al., 2012; Gabriel et al., 2012; Kim et al., 2006a; Guo et al., 2010; Lopes et al., 2011). In this context, Nb addition to titanium alloys has become common because of its biological passivity and capability to reduce the elastic modulus (Lopes et al., 2011; Cremasco et al., 2011; Afonso et al., 2007). Recent studies also demonstrate the feasibility of adding Si to Ti alloys, due to the possibility of combining the ability to reduce the elastic modulus and the increasing of alloys’ mechanical strength (Zhang et al., 2013a; Li et al., 2012; YunQing et al., 2011; Kim et al., 2006a). Given that titanium alloys properties are strongly dependent on the composition and distribution of the phases, it becomes relevant to study phase transformations and the correlation between microstructure and its properties in the development of new alloys for biomedical applications. Thus, the current study aims to investigate the influence of Si addition on the microstructure and mechanical properties of Ti-35Nb alloys, in order to assess the potential application of these materials in orthopedic implants.

2. Experimental

Ingots of Ti–35Nb alloys with different Si compositions (0; 0.15; 0.35; 0.55) (wt. %) were arc melted from Ti (99.84%), Nb (99.99%) and Si (99.9999%) using a water cooled copper hearth and non-consumable tungsten electrode under a high purity argon atmosphere; the obtained ingots were melted at least eight times, being flipped between each melt. A homogenizing heat treatment was applied at 1000 °C/8 h, and the ingots were then hot-rolled at 1000 °C in order to obtain 4 mm thick plates; several rolling operations 3 

were necessary and the ingots were reheated before each one of these operations. Afterwards, they were mechanically and chemically cleaned to remove oxide layers and were cut to obtain three samples for each composition. The dimensions of each sample were (25 x 20 x 3) mm. Finally, the samples were heated at 1000 ºC / 1 h and then cooled under the following conditions: furnace cooling (FC), air cooling (AC), and water quenching (WQ). Other ingots were obtained for tension testing according to the homogenization and hot rolling procedures described above. Five tensile specimens from each composition were machined from 5 mm thick plates; the tensile specimens’ dimensions were: 40 mm total length, 10 mm gauge length, 12.50 mm resection radius and 4 x 4 mm gauge area. They were water quenched after heat treatment at 1000 °C for 1 h, and tested in an INSTRON 3367 machine. The loading rate was 0.5 mm/min. Differential thermal analysis (DTA) was also employed in water-quenched alloys. Netzsch, model STA 449F3 was the DTA equipment used and the runs were performed under heating and cooling rates of 10 ºC/min in an Ar atmosphere in order to predict the phase transformation behavior. Triplicate tests for thermal analysis experiments were carried out. Vickers hardness was obtained in all the samples with loads of 2 kgf for 15 s; the reported values are the average obtained from five measurements. Elastic moduli were determined by ultrasonic methods, by measuring density, and longitudinal and transversal wave velocities. A piezoelectric transducer (10 MHz), in contact with the sample via coupling gel, was used for these measurements. Samples’ bulk density was measured by Archimedes method. Optical metallography was carried out on the samples, which were mounted and ground using a series of SiC sandpapers from 220 to 1500; they were then polished with 6 4 

to 3 ȝm diamond and with 1ȝm alumina suspensions. The samples were etched with Kroll’s solution which consisted of 6 mL HNO3, 3 ml HF and 91 ml H2O. Alloys fracture surfaces were observed by scanning electron microscope (SEM), by using a JEOL JCM – 5700 microscope. Energy dispersive X-ray spectroscopy (EDS) was used to map Si distribution in polished samples; Ti, Nb and Si compositions were also investigated via EDS. Transmission electron microscopy (TEM) images were obtained from FEI/Philips CM-200T (200 kV LaB6) using thin foils prepared in the FEI Helios NanoLab™ 600 DualBeam (FIB/SEM). Samples characterization by X-ray diffraction was performed using a Shimadzu XRD - 6000 equipment (40 kV, 30 mA), in Bragg–Brentano reflection geometry with Cu KĮ radiation (Ȝ = 1.5418 Å); data were obtained between 30º and 90º 2ș in steps of 0.02º with counting time of 2 s.

3. Results and Discussion

3.1 Microstructural Characterization Micrographs of the FC samples are shown in Fig. 1. The corresponding X-ray diffraction results are shown in Table 1, which displays the ȕ, Ȧ and Į phases. It was verified in all FC samples, except for the one containing 0.55% Si, that the microstructures are comprised by the ȕ phase (Fig. 1); some dark spots are found in the surroundings of the grain boundaries (Fig. 1a-c) and they appear to result from etch pits. Although no distinct change was noticed over the microstructures (Fig. 1a-c), silicon is considered to work as a ȕ-stabilizer in Ti-Nb alloys (Kim et al., 2006a). However, it is worth mentioning that silicon has an important role in hindering the occurrence of the Ȧ-phase (Kim et al., 2006a), i.e., stability of the ȕ-phase would be enhanced with increasing Si content. In TiNb alloys, Ȧ particles in a beta matrix are developed in the composition range from 26 to 5 

35% Nb (Souza et al., 2010; Zhou et al., 2004). In this study, the Ȧ phase was detected by X-ray diffraction in all FC samples (Fig. 2a), but there are evidences that the continuous increase in the Si content reduces the Ȧ-precipitate density; this fact is discussed in further detail in Section 3.2. Regarding the Ti-35Nb-0.55Si sample, peaks of Į phase were identified by X-ray diffraction (Fig. 3b), and its occurrence is indicated in dark etched areas by optical microscopy, as it can be seen in Fig. 1d. The SEM image (Fig. 3a) of this sample with greater magnification shows that the microstructure was strongly etched by Kroll’s solution and it is not possible to distinguish the etch pits and the cavities corresponding to Į phase, which was selectively corroded. However, a small volume fraction of Į phase seems to be formed due to low intensity peaks, as it can be seen in Fig. 3b. The difference between the morphology of this sample and that of the samples containing 0, 0.15 and 0.35% Si can be reasonably explained through equilibrium phase diagrams. Thus, in the 1000 ºC isothermal section proposed by Xu et al. (2005) for the Ti-Nb-Si system, it is noticed that the ȕ-Ti field is narrow, and the continuous increase of Nb content gradually reduces the solubility of Si in ȕ-Ti. In this study, all the samples were heated at 1000 ºC / 1 h and then cooled. Therefore, the phase transformation for a Ti-35Nb-0.55Si composition sample is done through the ȕ-Ti + Ti3Si two-phase field based on this isothermal section. Moreover, by taking the Ti-Si phase diagram (Massalski, 1990) under consideration, the presence of the Į phase can be ascribed to the result of a partial transformation originated from the eutectoid decomposition ȕ-Ti ĺ Į-Ti + Ti3Si at temperature near 865ºC; in relatively high concentration of ȕ-stabilizing elements, this decomposition reaction could be sluggish, even during furnace cooling (Narayana and Archbold, 1970). According to these diagrams, the occurrence of Ti3Si silicide is also feasible. Actually, there is limited information about its existence in the literature (Ramos et al., 2006). Masumoto et al. 6 

(2006), who worked with Ti-Nb-Al-Si alloys, identified Ti3Si-type particles, which were considered to be formed during the homogenization and/or the solution treatment. Therefore, in order to investigate a possible occurrence of titanium silicides, a Si distribution map was carried out by using energy dispersive spectroscopy (EDS). Figures 4a-c show scanning electron secondary images from the polished sample surfaces, while those in Fig. 4d-f are found in the respective mappings of Si distribution in the alloys. Through the analysis of Fig. 4(d and e), it was verified that Si distribution in the samples with 0.15% and 0.35% Si, respectively, was more homogeneous. However, for the sample containing 0.55% Si (Fig. 4f), some regions with higher concentrations of this alloying element have been observed. The composition of these samples obtained by EDS is shown in Table 2, in which it can be seen that data are in accordance with the nominal composition proposed in the study. For the sample containing 0.55% Si, a spot analysis of the composition by EDS was performed in a region in which there is higher concentration of this alloying element (point 1 in Fig. 4f) as well as in the matrix (point 2 in Fig. 4f). The composition of the two analyzed regions is shown in Table 2, which provides evidence of the existence of particles with Si composition close to 12 wt. %, whereas the matrix shows composition near to that of the sample containing 0.35% Si. As mentioned above, these particles are suggested to be Ti3Si-type compounds. Furthermore, Nb–Si and Ti–Si systems are analogous, and their Ti3Si and Nb3Si phases are isomorphous (Zhan et al., 2009; Xu et al., 2005). However their existence temperatures are different; Nb3Si is stable in the temperature range from 1700ºC to 1980ºC, whereas Ti3Si is stable at lower temperatures (

Influence of Si addition on the microstructure and mechanical properties of Ti-35Nb alloy for applications in orthopedic implants.

In the development of new materials for orthopedic implants, special attention has been given to Ti alloys that show biocompatible alloy elements and ...
1KB Sizes 0 Downloads 17 Views