Mechanical and electrochemical response in Surface treated low modulus biomedical alloy Ti-Nb-Ta-O

Surface modifica on of metallic biomedical implants are o�en performed using chemical or mechanical methods in order to make them more bio-ac ve or resistant against surface-induced phenomena such as wear, corrosion or corrosion fa gue. In the present study, one such method, known as Surface Mechanical A�ri on Treatment (SMAT), has been studied in terms of its effects on the mechanical and func onal response of a newly developed low modulus metastable β Ti-Nb-Ta-O alloy. The hardness of the surface was found to increase up to a certain dura on of SMAT, due to increased degree of deforma on on the surface. This was also supported by an increase in the peak broadening with respect to SMAT dura on. Apart from surface hardening, SMAT also resulted in improvement of corrosion resistance of the Ti-Nb-Ta-O alloy due to forma on of a more stable passive film.


Introduc on
β-Titanium alloys consis ng of elements like Nb, Ta, Zr are considered to replace conven onal materials such as cp-Ti, Ti-6Al-4V alloys, stainless steel etc. for orthopedic implant applica ons. These alloys have elas c modulii (50-80 GPa) closer to that of the human bone (4-30 GPa) which make them less prone to failure caused by stress shielding effect, resul ng from the modulus mismatch between implant material and human bone (4-30 GPA) [1,2]. On the other hand, these alloying elements (Nb, Ta, Zr) have been reported to be less toxic than V, Al, Cr, Ni etc [3]. The low elas c modulus of these alloys are known to be primarily caused by the stability of β-phase and an electron/atom (e/a) ra o ~4. 2-4.24. Moreover, the elements that are used as in the βtanium alloys, such as Nb, Ta, Zr, have been found to be non-toxic for human body, unlike other elements such as Al, V, Ni which are present in conven onal implants [4]. Therefore, many new alloys pertaining to β-tanium alloy system has been developed in the past few decades, some of the popular composi ons being Ti-13Nb-13Zr [5,6], Ti-35Nb-7Zr-5Ta [7], Ti-29Nb-13Ta-4.6Zr [8,9]. Very recently, a Ti-34Nb-2Ta-0.5O alloy, with low elas c modulus and high strength, has been developed by the present authors [10,11,12].
In addi on to low elas c modulus and bio-compa bility, an implant material needs to possess adequate mechanical strength to impart high fa gue and wear resistance. However, proper es such as fa gue and wear o�en depend upon the strength of the surface rather than the bulk material. Therefore, strengthening methods which involve hardening the surface rather than the whole material can offer promising results for this class of materials. Among different surface hardening methods, Surface mechanical A�ri on Treatment (SMAT) has received much a�en on in the recent mes. This method involves bombarding the surface of a material with hardened steel balls vibra ng, genera ng a hardened casing on the surface [13]. SMAT leads to nano-crystalliza on on the surface by causing severe plas c deforma on [14]. The process have been found to enhance the fa gue life by delaying crack ini a on on the surface due to increase in the hardness and introducing compressive residual stresses [15]. In addi on, there are reports on improvement in cell a�achment and prolifera on on several systems [16,17].
However, most of the literature regarding effect of SMAT on mechanical, biological and corrosion behaviour has been centred on stainless steel and cp-Ti. Similar systema c studies on β tanium alloys are few in literature, though a posi ve influence of SMAT on corrosion behaviour of β alloys has been reported [18]. diffrac on and microstructural characteriza on. The electrochemical behaviour of the material was analysed using the polariza on study.

Material and Processing
Ti-34Nb-2Ta-0.5O alloys sheets of ≈ 5 mm thickness, obtained in hot rolled (above β-transus) condi on, were cold rolled up to 70% thickness reduc on and then annealed at 900°C for 0.5h to obtain a uniform microstructure [10]. These annealed samples were then subjected to SMAT treatment at a frequency of 25 Hz for different dura ons of me, using a custom-designed SMAT unit (Cosmic Industrial Laboratories Limited, Bengaluru, India). The number of hardened steel balls used for the SMAT was 500 with each ball having a diameter of 4.75mm each. The working distance between base of the chamber and sample surface was 20 mm. Prior to SMAT treatment, the sample surfaces were mechanically polished up to P3000 paper followed by electropolishing at 40 V for 20 s using A3 solu on in a Struers Lectropol-5 machine.

Microstructural characteriza on
The structural characteriza on of the samples was done primarily using X-ray diffrac on. The X-ray diffrac on pa�erns were recorded, before and a�er SMAT, on the normal plane of the samples. A PanAnaly cal X'pert Pro X-ray diffractometer with Cu-Kα radia on was employed for this, where the pa�erns were recorded at a scan speed of 0.011º/s.The microstructures were examined and recorded using both op cal microscope (Zeiss) as well as a scanning electron microscope (ESEM, Quanta FEI). Transverse cross-sec ons of the samples were polished and then etched by Kroll's reagent (2% HF, 6% HNO 3 , 92% H 2 O) for 30 s, for microstructural examina ons.

Hardness measurement
The surface hardness values of the annealed and SMAT-processes samples were measured on the by Micro-vickers indenter at an indenta on load of 25 gf with a dwell me 10 s. The centre-to-centre distance between two indenta ons were kept more than five mes the diagonal of each indenta on.

Corrosion studies
The electrochemical performance of the Ti-Nb-Ta-O alloy before and SMAT processing was inves gated in terms of polariza on studies in a simulated body fluid (SBF) solu on using a standard three electrode poten ostat (CHI604E, C.H. Instruments).
Pla num wire was used as counter electrode and saturated calomel electrode (SCE) was used as reference . The SMAT-proccessed samples and non-SMAT electropolished samples were first cleaned by ultrasonic method and then were immersed in the solu on for 3 h to obtain the open circuit poten al. This was followed by polariza on studies using tafel extrapola on at a scan rate of 2 x 10 -4 Vs -1 with the voltage range between -0.6 V to +0.4 V.

Microstructural features of the star ng material
The X-ray diffrac on (XRD) pa�ern of the Ti-Nb-Ta-O alloy in as-received (AR) condi on is shown in Fig.1, respec vely. The XRD pa�ern consists of peaks from β-phase only. The AR sample was water quenched from 900°C a�er annealing, causing the reten on of β phase. Figure 2 shows op cal microstructure of the AR sample which consists of equiaxed β-grains. The presence of 0.5 wt% oxygen leads to preven on of forma on of α" or ω, which has been discussed in our previous study [10]. The average diameter of the β-grains were found to be 53±24 µm.

Mirostructural evolu on a�er surface treatment
The XRD pa�erns of the material a�er being subjected to SMAT for different dwell mes, i.e. 15 min., 30 min., 60 min. and 120 min., are shown in Fig.3. All the four pa�erns reveal the presence of β phase only, with minor broadening observed in the There is also an indica on of satura on in the FWHM values as the me reaches 120 min. from 60 min., at least for β {110} , β {200} peaks. The peak broadening in plas cally deformed materials is manifested due to reduced crystallite size, and increased microstrain. The increase in the FWHM values as a func on of me, clearly indicates the existence of plas c deforma on on the surface. Peak broadening in SMAT-processed materials have previously been reported in literature and have been a�ributed to the forma on of nanocrystallized grains as well as a significant increase in the disloca on densi es a�er SMAT [15,19]. In the present study, the fine features on the deformed SMAT surface could not be resolved with op cal microscopy, and therefore wee characterized using scanning and transmission electron microscopy.  Figure 5 shows the cross-sec on SEM images of samples undergoing SMAT for 60 min. The region near the SMAT-affected side, marked by arrows, shows a contrast difference with the rest of the sample. The deforma on-affected region has also been marked with the dashed lines. It also shows the presence of band-like features, as indicated by the blue arrow-lines. These could be perceived as microbands formed by SMAT. In metastable β-alloys, forma on of microbands due to heavy plas c deforma on have been reported in several literature [20,21]. Their forma on during SMAT has been a�ributed to spli ng of dense disloca on walls, which separate the cell blocks within the grains, leads to the forma on of these microbands [17].

Effect of surface treatment on hardness
The surface of AR sample shows a hardness of 290 + 7 HV in the annealed condi on, which is similar to that of the oxygen-added similar alloys, in annealed or β solu on treated condi on. The surface hardness a�er SMAT for different dwell mes is shown in Fig. 6. The hardness can be seen to increase as a func on of dwell me up to 60 min, beyond which no significant change was observed. The increase in the hardness a�er SMAT on the surface results from the deforma on induced hardening mechanisms such as grain size strengthening and disloca on strengthening [15,22]. The broadening in the XRD peaks, as discussed earlier, also confirm this. The microbands observed in the microstructures are also perceived to be a consequence of the heavy deforma on process associated with SMAT. These microbands have been a�ributed to lead to the eventual forma on of nanocrystallized grains on the surface. The microstructure does not reveal any other features dis nguishing the deformed layer produced by SMAT, indica ng that the microstructural features that dis nguish the SMAT affected layer are fine enough.

Corrosion Studies
The corrosion current density (I corr ) and the corrosion poten al (E corr ) values were measured by extrapola ng the the cathodic and anodic regions of the Tafel plot. Average Icorr values of the AR and SMAT samples were found to be 47 x 10 -3 µA/cm 2 and 11 x 10 -3 µA/cm 2 , while the average E corr values were found to be -0.28 V and -0.12 V. The reduc on in I corr values and increase in E corr values clearly show an improvement in corrosion resistance a�er SMAT. This type of behaviour had been reported previously for SMAT-processed cp-Ti [15], Ti-6Al-4V alloy [16] and β Ti-25Nb-3Mo-3Zr-2Sn alloy [18]. It has been proposed in the literature that severe plas c deforma on methods like SMAT leads to genera on of large number of la ce defects, as also seen from the microstructure of SMAT sample in the present alloy. The presence of these defects leads to forma on of a much denser and more stable passive film on the surface of these samples. A similar mechanism can also be expected to act in the present case. Though the passive oxide film behaviour needs to be further studied in order to validate this, the Tafel plots indicate that SMAT is a favourable method in terms of electrochemical performance of the β-Ti-Nb-Ta-O alloy.
In the present study, the effects of surface mechanical a�ri on treatment on the mechanical and electrochemical behaviour of a Ti-Nb-Ta-O alloy have been studied. The surface treatment resulted in surface hardening and the me dura on of SMAT was found to affect the surface hardness. The microstructure of SMAT materials exhibited the forma on of micro-bands near SMATprocessed surface. This in effect resulted in hardening of the surface region. The corrosion behaviour also improved a�er SMAT possibly due to forma on of more stable and dense passive oxide film by this process.