Fatigue behavior of Ti50Zr alloy for dental implant application

Accepted Manuscript Fatigue behavior of Ti50Zr alloy for dental implant application Wenfang Cui, Yaohui Liu PII:

S0925-8388(19)31446-X

DOI:

https://doi.org/10.1016/j.jallcom.2019.04.165

Reference:

JALCOM 50349

To appear in:

Journal of Alloys and Compounds

Received Date: 3 January 2019 Revised Date:

21 March 2019

Accepted Date: 15 April 2019

Please cite this article as: W. Cui, Y. Liu, Fatigue behavior of Ti50Zr alloy for dental implant application, Journal of Alloys and Compounds (2019), doi: https://doi.org/10.1016/j.jallcom.2019.04.165. 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 proof before it is published in its final 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.

ACCEPTED MANUSCRIPT

Fatigue behavior of Ti50Zr alloy for dental implant application Wenfang Cui∗, Yaohui Liu Key Laboratory for Anisotropy and Texture of Materials (Ministry of Education), School of Material Science and Engineering, Northeastern University, Shenyang, 110819, China

Abstract The microstructure, tensile and fatigue behavior of a hot-rolled Ti50Zr alloy applied for dental implant

RI PT

were investigated. The commercial Ti16Zr alloy was taken as a reference material. The load vs. number of cycling curves of the one-piece type Ti16Zr and Ti50Zr dental implants were tested according to ISO 14801 standard testing methods. The results show that the yield strength and fatigue limit of Ti50Zr alloy are up to 1001±8MPa and 500±10MPa, respectively, 23% and 32% higher than Ti16Zr alloy. The enhanced strength of Ti50Zr alloy is

SC

attributed to multiple mechanisms, including acicular martensite strengthening, Zr solid solution strengthening, nano-twins and dislocations strengthening. The maximum cycling loads of Ti50Zr implants in air, artificial saliva solution and with sand blasting plus acid etching (SLA) treated surface reach 400N, 400N and 350N (5×106 cycles

M AN U

without failure), respectively, 33%∼75% higher than Ti16Zr implants. High surface roughness harms the fatigue performance of the Ti-Zr implants owing to the promotion role of notch-type defects to fatigue crack initiation. A corrosion-assisted fatigue process in artificial saliva solution under large load reduces the fatigue life of the Ti-Zr implants. Ti50Zr alloy is an ideal candidate material applied for dental implant with regards to its good biocompatibility, excellent mechanical properties and economic processing route.

1. Introduction

TE D

Keywords: Titanium alloys, Dental implant, Microstructure, Fatigue

Nowadays, millions of the patients with tooth loss in the world have got good treatment through the method of dental implantation. However, it has been reported that the fracture of the center screw and implant parts occasionally occurs due to improper design or material factor [1-3]. Piatteli et al. [4] and Barbosa et al. [5]

EP

reported that implant fracture resulted from fatigue or overload. Hernandez-Rodriguez et al. [6] examined the failed dental implant after 6 months of service and pointed out that the fracture of inner screw was promoted by an overload system with cyclic high level stresses due to bone re-sorption. These results indicate that it is

AC C

indispensable to further improve the mechanical properties of the implant materials. Commercially pure titanium (cp-Ti) has been the first choice for dental implant material with regard to the good corrosion resistance and biocompatibility. But the mechanical strength of cp-Ti is insufficient for the small diameter implants (≤ 3.5 cm) which are preferred to be placed in a narrow edentulous ridge. Ti6Al4V alloy is widely used material for orthopedic implants owing to its high strength and fatigue performance. But corrosion-induced release of V and Al ions has been shown to cause toxic reactions and negative health effects. As a substitute material for Ti6Al4V alloy, Ti6Al7Nb alloy still includes Al element which is a potential harm in human body to man’s memory and osteoporosis [7]. Severe plastic deformation (SPD) techniques, e.g. high-pressure torsion (HPT) [8, 9], equal-channel angular ∗

Corresponding author. E-mail address: [email protected] (W.F. Cui) 1

ACCEPTED MANUSCRIPT pressing (ECAP) [10], multi-pass accumulative roll-bonding [11] etc. were suggested to increase the strength of cp-Ti through the refinement of grains. It has been reported that the ultimate strength of ultrafine-grained titanium (UFG) grade 4 produced by ECAP reached 1240 MPa [12, 13], and the endurance limit of the smooth specimen up to 590MPa after 107 cycles [14]. Although high strength titanium can ensure the long-life service of dental implant, but the complex and inefficient processing route increase the cost of manufacture, which limits the wide application of the material.

RI PT

Recently, a Ti-(13~17) % Zr alloy used for narrow diameter implant was developed (Roxolid; Institute Straumann AG, Switzerland). The new material behaves good biocompatibility and improved mechanical strength compared to cp-Ti grade 4 [15-17]. The clinical reports display that narrow diameter Ti-15Zr implant shows high survival and success rate (>95%) in the period of 36 month [18]. Nevertheless, Kobayashi et al [19] found out that hardness and tensile strength of Ti-50% Zr alloy could be increased to a maximum of about 2.5 times cp-Ti. And

SC

Ti-50%Zr alloy showed more cells attachment than Ti, higher ALP activity and OC expression than TiNb alloy [20]. Without question, Ti-50% Zr alloy is also a good candidate material for dental implant application. Up to

M AN U

date, the current study is limited to the tensile properties of as-cast Ti-50% Zr alloy [21, 22]. Few investigations evaluate the tensile and fatigue performances of the hot-worked Ti-50% Zr alloy. The present paper aimed to systematically study the mechanical behavior of the hot-rolled Ti-50%Zr alloy, including the effects of the oral environment and SLA surface treatment (sand blasting + acid erosion, which is used for improving bone binding of dental implant) on the fatigue performances of dental implant specimens. The strengthening and fracture mechanisms of Ti-50%Zr alloy were clarified. Since the commercial Ti-(13~17) % Zr alloy is used under cold working condition, for comparison, a cold forged Ti-16% Zr alloy was investigated simultaneously in the same

TE D

testing conditions as Ti-50% Zr alloy.

2. Materials and experimental procedures

The ingots of Ti-16 wt.%Zr and Ti-50 wt.%Zr alloy (abbreviated as Ti16Zr and Ti50Zr) were re-melted for three times using vacuum consumable electrode arc furnace, and then hot forged into φ15mm and φ25mm bar at ∼900 , respectively. Ti16Zr bar was further rotary forged into φ11 mm rod at room temperature. Ti50Zr

EP

850

bar was hot rolled into φ11 mm rod at 800

followed by air cooling.

AC C

The phase constitute of Ti50Zr alloy was analyzed by using X-ray diffractometer (Smart Lab, Japan) with Cu Kα radiation at the scanning speed of 8°/min. The microstructures of the as-received Ti16Zr and Ti50Zr alloy were observed by optical microscope (Zeiss, Germany). The thin foils for TEM observation were mechanically ground on waterproof abrasive paper to 50 µm thickness and then electrolyticly polished by twin jet at −30°C in an electrolyte consisting of 10% perchloric acid and 90% ethanol (vol.%). TEM observation was performed by using transmission electron microscope (JEM-2100, Japan) operating at 200 kV accelerating voltage. A one-piece type dental implant specimen was designed with φ 4.1 mm in diameter and 14 mm in total length, as shown in Fig. 1a. The radius of the notch at the thread was 0.12 mm. The thread parts of some specimens were subjected to SLA surface treatment. The implants were firstly sand blasted with φ 0.2mm Al2O3 grit under 0.6 MPa pressure for 90 seconds. After cleaned in ethanol and acetone, the implants were etched in the solution of H2SO4: HCl: H2O=1: 1: 2 (vol.) at 100

for 20 min. The surface morphology of the SLA treated implant was

observed by scanning electron microscopy (JSM-7001F), as shown in Fig. 1b. It can be seen that a lot of 2

ACCEPTED MANUSCRIPT micro-pits with the size of (2~10) µm covered with the thread surface. The surface roughness (Ra) was measured to be 3.88±0.41 µm by laser confocal microscope (LSM 800, Germany). The monotonic tensile tests were performed by using an Instron (AG-Xplus) servohydraulic machine with the specimen of 25 mm in gauge length and 6 mm in gauge diameter. The cross-head velocity was 0.005 min-1. Fatigue tests were performed in air by using MTS 810 Hydraulic servo fatigue testing machine. The smooth fatigue specimens have gauge diameter of 5 mm and gauge length of 10 mm. The testing frequency f was 15 Hz,

RI PT

and the stress ratio R was −1. The endurance limit was determined as the maximum stress under 5×106 cycles without fracture.

Fatigue tests of the dental implants were performed by using electromagnetic dynamic fatigue testing machine (CARE M-3000, China), as shown in Fig. 2. The installation and testing procedures of the implant specimens conformed to ISO 14801-2007 standard, which specifies that the force must be applied at an angle of

SC

30 degrees from the axis of the implant specimen, the compressive load ratio R = 0.1, frequency f = 15 Hz in air at room temperature and frequency f = 10 Hz in artificial saliva solution at 37°C. The implant specimens were fixed

M AN U

in an epoxy resin inlay with an analogous elastic modulus to human bone and sufficient toughness for cyclic testing. Some machined and SLA treated specimens were tested in air for evaluating the effect of surface roughness on the fatigue performance. The other machined specimens were tested in artificial saliva solution to examine the effect of the aggressive oral environment on the fatigue performance. The endurance limit of the implant specimen was determined as the maximum compressive force under 5×106 cycles without fracture. The fracture surfaces of the specimens were observed by SEM.

3.1 Phase and microstructure

TE D

3. Results and discussion

Based on Ti-Zr binary phase diagram, β→α phase transformation temperature (Tβ) of Ti50Zr alloy is down to 630°C. It is possible that small amount of remained β phase was retained to room temperature after hot working in

EP

β phase field [23, 24]. In order to confirm this, the phase composition of Ti50Zr alloy was analyzed by XRD pattern, as shown in Fig. 3. Only α-Ti (Zr) phase with HCP crystal structure exists in Ti50Zr alloy, proving the complete transformation of β phase into α-Ti (Zr) phase. It is noted that the diffraction angles (2θ) of the crystal

AC C

planes shift c.a. 2° towards low angle compared to standard α-Ti crystal diffraction, indicating a severe lattice distortion in Ti50Zr due to the atomic radius difference between Ti and Zr (rTi=0.147 nm, rZr=0.162 nm). Fig. 4 shows the optical micrographs of the as-received Ti16Zr and Ti50Zr alloy. In Ti16Zr alloy, α platelets are elongated towards deformation direction. The platelet interfaces become short and curve (Fig. 4a). Ti50Zr alloy displays an acicular microstructure with different size and orientations. No primary β grain boundaries were found (Fig. 4b). Combining the microstructure feature with XRD analysis, the acicular microstructure in Ti50Zr alloy can be identified as acicular martensite ( α ' ). It has the same crystal structure and similar lattice constants with stable α-Ti (Zr) phase. TEM observation gives the microstructure details of the as-received Ti16Zr and Ti50Zr alloy. In Ti16Zr alloy, high-density dislocations exhibit "U" or "Y" type distribution (Fig. 5a). Small amount of short and narrow α platelets can be observed (Fig. 5b). Since rotary forge processing produces multidirectional deformation, the 3

ACCEPTED MANUSCRIPT movement direction of dislocations is constantly changing, and the interaction between dislocations is extremely strengthened. Various dislocation behaviors, e.g. tangle, intersection and cross slip etc., are intertwined, leading to heterogeneous dislocation distribution. Different from the cold worked Ti16Zr alloy, there are large amount of multi-scale α ' platelets in the hot rolled Ti50Zr alloy. Fig. 6a shows the extremely fine platelets with 0.05∼0.07 µm in width and 0.5∼0.8 µm in length. Besides, the parallel slip bands and nano-twin structures within platelets

RI PT

were frequently observed (Fig. 6b and 6c). Fig. 6d shows that the intersection of twins and slip bands causes the misplacement or kink of slip bands, as indicated by yellow arrow. It can be seen that Ti50Zr alloy has diversified microstructure features, which associate with hot deformation and phase transformation process. Because of low Tβ temperature of Ti50Zr alloy, the hot rolling process was allowed to be performed at lower temperature (700~800°C) in β phase field. A large number of crystal defects (dislocations, slip bands etc.) were thus produced

SC

within β grains owing to the difficult occurrence of recover and recrystallization. These defects became preferential nucleation sites for martensite and accelerated martensitic phase transformation process. This explains

M AN U

why 100% martensitic microstructure can be obtained under air cooling condition after rolling. The formation of fine twins was accompanied by acicular martensitic phase transformation. Since martensite is a non-diffusive shearing transformation, the rapid growth of martensite nucleation usually causes great impact force between platelets with different orientations. The formation of the fine twins is favorable to release stress concentration at interfaces [25]. Therefore, the fine twins can be considered as a substructure of acicular martensite.

TE D

3.2 Strengthening mechanisms

Fig. 7 shows the engineering stress-strain curves and S-N curves of Ti16Zr and Ti50Zr alloy. The tensile properties and fatigue limits are shown in Table 1. In comparison with Ti16Zr alloy, Ti50Zr alloy exhibits 23% higher yield strength (Rp0.2 ), 23% higher ultimate tensile strength (Rm) and 32% higher fatigue limit (σ-1). The

EP

yield strength of Ti50Zr alloy is up to 1001±8 MPa without drastic reduction of elongation. The endurance limit (σ-1) of Ti50Zr alloy reaches 500±10 MPa, even better than annealed Ti6Al4V alloy with basket-weave microstructure [26]. Since high stress concentration usually exists at the variable cross-section of implant

AC C

abutment or screw, the high strength material can greatly increase the resistance to accidental overloading or fatigue fracture for those unreasonably designed or improperly installed dental implant systems [27]. Table 1 Tensile and fatigue properties of the as-received Ti16Zr and Ti50Zr alloy.

Alloys

Rp0.2 (MPa)

Rm (MPa)

A (%)

σ-1 (MPa)

Ti16Zr

813±10

852±8

14.1±0.4

380±10

Ti50Zr

1001±8

1045±5

12.2±0.1

500±10

The differences in the mechanical properties of Ti16Zr and Ti50Zr alloy are attributed to their different strengthening mechanisms. Since no compound exists in Ti-Zr binary alloy, the solid solution strengthening is one of the important strengthening ways. Supposing ignoring shear modulus difference between Ti and Zr, the formula for calculating solution strengthening is given in Eq. (1) [28]:

∆τ 1 = αGε b c1 / 2 3/ 2

(1) 4

ACCEPTED MANUSCRIPT where

α −constant, 2.97×10-3, ε b − mismatch of Ti and Zr atom radius, 0.23, G −shear modulus, 41.4GPa [29],

c −Zr concentration in Ti matrix (in mass percent). The shear stress increment ∆τ 1 for Ti16Zr and Ti50Zr alloy are estimated

to

be

18.4MPa

and

32.6MPa,

respectively.

It can be seen that the contribution of the solid solution strengthening in Ti-Zr binary alloys is very limited. The main strengthening mechanism of Ti16Zr alloy is attributed to work hardening. High-density dislocations formed in the course of cold working become strong obstacles to dislocations movement. Based on the famous

RI PT

Pellet-Hirsh relationship (2) describing the flow stress increment ∆τ with dislocation density ρ [28]: ∆τ 2 = αGb ρ

(2)

the shear stress increment ∆τ 2 in Ti16Zr alloy is about 160MPa, assuming α= 0.4, dislocation density ρ = 10 /cm2 11

and Burgers vector b =0.295 nm. The value accounts for 60% of the yield shear stress, indicating the important contribution of working hardening to the strength of Ti16Zr alloy. Moreover, cold working also reduces the size of

SC

primary β grains and α platelets, which limits dislocation slip distance and dislocations movement. To summarize, multiple mechanisms jointly produce strengthening effects to Ti16Zr alloy. By comparison, Ti50Zr alloy exhibits

M AN U

different strengthening mechanisms. Solid solution strengthening only accounts for 10% of the yield shear stress of Ti50Zr alloy. Thus, the dominated strengthening mechanism originates from multi-scale and multi-orientation martensite platelets and defect structures including nano-twins and slip bands etc. This indicates that increasing Zr content in Ti-Zr binary alloy can obtain greatly improved mechanical properties by decreasing thermomechanical processing temperature and promoting martensitic transformation. Furthermore, the low temperature thermomechanical processing route of Ti50Zr alloy has greater advantage in the aspect of mass production

TE D

efficiency in comparison with ECAP + annealing processing of cp-Ti [30]. 3.3 Fatigue behavior of implant specimens

Fig. 8 shows the load-number of cycling (L-N) curves of the one-piece type Ti16Zr and Ti50Zr implants with smooth and SLA treated surfaces tested in air and artificial saliva solution, respectively. It can be seen that in the various conditions, Ti50Zr implants are able to endure higher load than Ti16Zr implant. The wet environment has

EP

no influence on the fatigue limit of Ti16Zr and Ti50Zr implants (300N and 400N under 5×106 cycles without failure, respectively). But with the increase of the load, the effect of corrosion-assisted fatigue fracture is more and

AC C

more obvious, as seen in Fig.8a and 8b. Compared to smooth surface, high roughness surfaces cause harmful effect on the fatigue performances of Ti16Zr and Ti50Zr implants (Fig. 8c). The fatigue limits decrease to 200N and 350N, respectively. The results indicate that although sand-blast treatment produces surface hardened layer, which is beneficial to the improvement of the fatigue performance, but micro-dimple defects formed by acid erosion after sand-blast treatment accelerate fatigue cracking. Nevertheless, the fatigue limit of the SLA treated Ti50Zr implant is still higher than Ti16Zr implant with smooth surface (300N) and cp-Ti Grade 4 processed by ECAP (160~180N) [31,32], showing the higher resistance level of Ti50Zr alloy to corrosion-fatigue and notch-fatigue than cold-worked Ti16Zr alloy and extruded pure titanium. 3.4 Fracture surface observation and analysis

By observing the fracture location of the implant specimens, it was found that all the SLA treated specimens were broken in thread part. Most of the specimens tested in dry and wet environment were broken in thread part, 5

ACCEPTED MANUSCRIPT and only a few ones in narrow neck part. This indicates that the root of the thread close to the support bears maximum tensile stress [33]. The fracture surface morphologies of the Ti50Zr implants were observed by SEM, as shown in Fig. 9. When the maximum cycling load was 500 N, the numbers of cycling to fracture in air, artificial saliva solution and with SLA surface were 356540, 65248 and 14224 cycles, respectively. The fracture surfaces of three specimens consist of crack initiation, propagation and overload regions. The paths of the crack initiation and propagation are indicated by yellow arrows in Fig. 9a, 9c and 9e. Under different testing conditions, the

RI PT

specimens exhibit different fracture features. The dry specimen has small fatigue region (Fig. 9a), in which the plastic deformation traces with many ductile dimples and tearing ridges are clearly seen (Fig.9b). It indicates that the dry specimen has experienced large plastic strain accumulation before overload fracture. Contrast to this, the fracture surfaces of the wet and SLA treated specimens are relative flat. The area of the fatigue regions becomes large. The secondary cracks perpendicular to the main crack are seen to propagate along platelet interfaces (Fig.

SC

9d and 9f). These features show that the fatigue life of the dry specimen is mainly cost in the stage of crack nucleation. The wet environment and high surface roughness accelerate crack nucleation, and thus reduce fatigue

M AN U

life of the implant specimen.

As mentioned above, the fracture features of the implant specimens are similar to the retrieved broken screw in clinical application [34], suggesting that the loading mode basically conforms to the practical force condition of dental implant. Due to the 30° inclination angle along the implant axis from compression loading direction, a tension stress was produced close to epoxy resin level on one side of the specimen [35]. In this case, the aggressive environment and the surface roughness of the specimen have great influence on fatigue crack initiation and growth rate. In artificial saliva solution, the natural surface passive film of the implant is easily damaged

TE D

under cyclic tensile loading. Local corrosion at the rupture position of the passive film will gradually develop into fatigue crack. It is a corrosion-assisted fatigue process, depending on load and the properties of passive film. Ti50Zr alloy exhibits thicker and more stable passive film than Ti16Zr alloy because more zirconia incorporates in titanium dioxide (The relevant findings will be published in a separate paper). Therefore, the wet environment

EP

produces less effect on the fatigue crack initiation of Ti50Zr implant particularly under moderate loading. Fatigue crack growth is also possibly affected by corrosive environment. The combined effects of the stress corrosion and true fatigue will increase crack growth rate [36]. However, it was not found the corroded traces in the fatigue

AC C

region of the implant specimen. This may be because the high loading frequency shortens the crack opening time. It is difficult for corrosive solution to enter the crack tip. SLA treatment produces a large number of notch-shaped defects (micro-pits). They become potential crack sources [37]. The stress concentration at the bottom of micro-pits strongly promotes the early formation of fatigue crack. Since fatigue crack growth is not affected by surface defects, thus the reduced fatigue life of the SLA treated Ti-Zr implants is mainly attributed by the premature fatigue cracking. Compared to machined specimen, the endurance limits of the SLA treated Ti16Zr and Ti50Zr specimens decrease 33% and 12.5%, respectively, indicating a lower notch-fatigue sensitivity of Ti50Zr alloy. Under whatever conditions, martensite platelets in Ti50Zr alloy play a role of the obstacles to crack growth. As crack tip encounters the platelet interfaces, the secondary cracks are helpful to release the stress concentration there and consume main crack propagation energy. This is one of the reasons why fatigue properties of Ti50Zr alloy is better than Ti16Zr alloy. 6

ACCEPTED MANUSCRIPT 4. Conclusions 1. The yield strength (Rp0.2) and fatigue strength (σ-1) of the hot-rolled Ti50Zr alloy reach 1001MPa and 500MPa, respectively, increasing 23% and 32% than the cold-forged Ti16Zr alloy. Multi-mechanisms including solid solution strengthening, multi-scale and multi-orientation acicular martensite strengthening and defect structure strengthening jointly make contribution to increasing the mechanical strength of Ti50Zr alloy.

RI PT

2. The fatigue endurances in air and artificial saliva solution of the as-received Ti50Zr implants reach 400N, and the fatigue endurance of SLA treated Ti50Zr implant reaches 350N, increased by 33% and 75%, respectively, compared to Ti16Zr implant. The greatly improved fatigue properties of Ti50Zr implants are attributed to the enhanced yield strength and the inhabitation role of acicular martensite platelets to fatigue crack propagation.

SC

3. SLA surface treatment is harmful to the fatigue performances of Ti50Zr implants. The notch-type surface defects constitute potential crack sources, which leads to the reduction of fatigue strength. The surface passive

M AN U

films of the Ti50Zr implants in artificial saliva solution are easily damaged by high cyclic loading. Local corrosion promotes the formation of fatigue crack.

4. In comparison with Ti16Zr implant, the fatigue performances of Ti50Zr implant are less affected by oral saliva and high surface roughness. Improving the mechanical properties and corrosion resistance of the material are helpful to increase the capability of anti corrosion-fatigue and notch-fatigue of dental implant. Ti50Zr alloy is an ideal material applied for small diameter dental implant with regards to its good biocompatibility, excellent

Acknowledgement

TE D

mechanical properties and economic processing method.

The study was financially supported by National Natural Science Foundation of China (No. 51525101). References

EP

[1] K. Shemtov-Yona, D. Rittel, Identification of failure mechanisms in retrieved fractured dental implants, Eng. Failure Analysis 38 (2014) 58–65.

AC C

[2] V. Zatkalíková, P. Palček, L. Markovičová, M. Chalupová, Analysis of fractured screw shaped Ti6Al4V dental implant, Mater. Today: Proceed. 3 ( 2016 ) 1216 – 1219. [3] S.M. Croitorua, I.A. Popovici, R&D on dental implants breakage, Appl. Surf. Sci. 417 (2017) 262–268. [4] A. Piattelli, M. Piatelli, A. Scarano, L. Montesani, Light and scanning electron microscopic report of four fractures implants, Int. J. Oral Maxillofac. Implants (1998) 561–564. [5] C. Barbosa, J.L. Nascimento, R.O. Centeno, I.M.V. Caminha, I.C. Abud, Failure Analysis of Titanium-Based Dental Implant, J. Fail. Anal. Preven. 10 (2010) 138–142. [6] M.A.L. Hernandez-Rodriguez, G.R. Contreras-Hernandez, A. Juarez-Hernandez, B. Beltran-Ramirez, E. Garcia-Sanchez, Failure analysis in a dental implant, Engin. Fail. Analysis 57 (2015) 236–242. [7] J. Karbowniczek, F. Muhaffel, G. Cempura, H. Cimenoglu, A. Czyrska-Filemonowicz, Influence of electrolyte 7

ACCEPTED MANUSCRIPT composition on microstructure, adhesion and bioactivity of micro-arc oxidation coatings produced on biomedical Ti6Al7Nb alloy, Surf. Coat. Technol. 321 (2017) 97-107. [8] H.Y. Um, B.H. Park, D.H. Ahn, M.I.A.E. Aal, J. Park, H.S. Kim, Mechanical and biological behavior of ultra fine-grained Ti alloy aneurysm clip processed using high-pressure torsion, J. Mechan. Behav. Biomed. Mater. 68 (2017) 203–20.

RI PT

[9] A.P. Zhilyaev, T.G. Langdon, Using high-pressure torsion for metal processing: Fundamentals and applications , Prog. Mater. Sci. 53 (2008) 893-899.

[10] R.Z. Valiev, T.G. Langdon, Principles of equal-channel angular pressing as a processing tool for grain refinement, Prog. Mater. Sci. 51 (2006) 881-889.

accumulative roll bonding, Scripta Mater. 62 (2010) 321-324.

SC

[11] D.K. Yang, P. Cizek, P. Hodgson, E.W. Cui, Ultrafine equiaxed-grain Ti/Al composite produced by

[12] C.N. Elias, M.A. Meyers, R.Z. Valiev, S.N. Monteiro, Ultrafine grained titanium for biomedical applications:

M AN U

An overview of performance, J. Mater. Res. Technol. 2 (2013) 340–350.

[13] I.P. Semenova, R.Z. Valiev, E.B. Yakushina, G.H. Salimgareeva, T.C. Lowe, Strength and fatigue properties enhancement in ultrafine-grained Ti produced by severe plastic deformation. J. Mater. Sci. 43 (2008) 7354–7359.

[14] I.P. Semenova, G.Kh. Salimgareeva, V.V. Latysh, T. Lowe, R.Z. Valiev, Enhanced fatigue strength of commercially pure Ti processed by severe plastic deformation, Mate. Sci. Engin. A 503 (2009) 92–95.

TE D

[15] H. Michelle Grandin, S. Berner, M. Dard, A review of titanium zirconium (TiZr) alloy for use in endosseous dental implants, Mater. 5 (2012) 1348-1360.

[16] J. Gottlow, M. Dard, F. Kjellson, M. Obrecht, L. Sennerby, Evaluation of n new titanium-zirconium dental

14 (2012) 538-545.

EP

implant: A biomechanical and histological comparative study in the mini pig, Clin. Implant Dent. Relat. Res.

[17] N. Bernhard, S. Berner, M. De Wild, M. Wieland, The binary TiZr alloy- A newly developed Ti alloy for use

AC C

in dental implants, Forum Implantol 5 (2009) 30-39. [18] S. Barter, P. Stone, U. Bragger, A pilot study to evaluate the success and survival rate of titanium-zirconium implants in partially edentulous patients: Results after 24 months of follow-up, Clin. Oral Implants Res. 23 (2012) 873-881.

[19] E. Kobayashi, S. Matsumoto, H. Doi, T. Yoneyama, H. Hamanaka, Mechanical properties of the binary titanium-zirconium alloys and their potential for biomedical application, J. Biomed. Mater. Res. 29 (1995) 943-950. [20] S. Sista, C. Wen, P.D. Hodgson, G. Pande, The influence of surface energy of titanium-zirconium alloy on osteoblast cell function in vitro, J. Biomed. Mater. Res. 74A (2005) 49-58. [21] H.C. Hsu, S.C. Wu, Y.C. Sung, W.F. Ho, The structure and mechanical properties of as-cast Zr–Ti alloys, J. 8

ACCEPTED MANUSCRIPT Alloys Compounds 488 (2009) 279–283. [22] Y. Li, Y. Cui, F. Zhang and H. Xu, Shape memory behavior in Ti–Zr alloys, Scripta Mater. 64 (2011) 584–587. [23] J.M.C. Moreno, M. Popa, S. Ivanescu, C. Vasilescu, S. I. Drob, E.I. Neacsu1 and M.V. Popa, Microstructure, Mechanical Properties, and Corrosion Resistance of Ti-20Zr Alloy in Undoped and NaF Doped Artificial

RI PT

Saliva, Met. Mater. Int. 20 (2014) 177-187. [24] H.C. Hsu, S.C. Wu, S.K. Hsu, et al. Effects of heat treatments on the structure and mechanical properties of Zr-30Ti alloys, Mater. Character. 62 (2011) 157-163.

[25] S. Banerjee, R. Krishnan, Martensitic Transformation in Zr-Ti alloys. Metall. Trans. 4 (1973) 1811.

SC

[26] J.H. Zuo, Z.G. Wang, E.H. Han, Ultra-high cycle fatigue behavior of Ti-6Al-4V alloy, Acta Metallurgica Sinica 43 (2007) 705-709.

[27] M. Baldassarri, Y. Zhang, Van P. Thompson, E.D. Rekow, C.F.J. Stappert, Reliability and failure modes of

M AN U

implant-supported zirconium-oxide fixed dental prostheses related to veneering techniques, J. dentistry 39 (2011) 489-498.

[28] T.H. Courtney, Mechanical Behavior of Materials, McGraw-Hill Companies, Inc. publication, USA 2000. [29] A.E. Medvedev, A. Molotnikov, R. Lapovok, R. Zeller, S. Berner, P. Habersetzer, F.D. Torre, Microstructure and mechanical properties of Ti-15Zr alloy used as dental implant material, J. Mech. Behav. Biomed. Mater. 62 (2016) 384-398.

TE D

[30] R.B. Figueiredo, E.R. de C. Barbosa, X.C. Zhao, X.R. Yang, X.Y. Liu, P.R. Cetlin, T.G. Langdon, Improving the fatigue behavior of dental implants through processing commercial purity titanium by equal-channel angular pressing, Mater. Sci. Engin. A, 619 (2014) 312-318. [31] J.M. Ayllón, C. Navarro, J. Vázquez, J. Domínguez, Fatigue life estimation in dental implants, Engin.

EP

Fracture Mechanics 123 (2014) 34-43.

[32] R.B. Figueiredo, E.R. Barbosa, X.C. Zhao, X.R. Yang, X.Y. Liu, P.R. Cetlin, T.G. Langdon, Improving the

AC C

fatigue behavior of dental implants through processing commercial purity titanium by equal-channel angular pressing, Mater. Sci. Eng. A 619 (2014) 312–318. [33] T.T. Wu, H.Y. Fan, R.Y. Ma, H.Y. Chen, Z. Li, H.Y. Yu, Effect of lubricant on the reliability of dental implant abutment screw joint: An in vitro laboratory and three-dimension finite element analysis, Mater. Sci. Eng. C 75 (2017) 297-304.

[34] K. Yokoyama, T. Ichikawa, Hiroki Murakami, Y. Miyamoto, K. Asaoka. Fracture mechanisms of retrieved titanium screw thread in dental implant, Biomater. 23 (2002) 2459–2465. [35] J.M. Ayllón, C. Navarro, J. Vázquez, J. Domínguez, Fatigue life estimation in dental implants, Engin. Fracture Mechanics 123 (2014) 34–43. [36] R.A. Antunes, M.C. L. Oliveira, Corrosion fatigue of biomedical metallic alloys: Mechanisms and mitigation, 9

ACCEPTED MANUSCRIPT Acta Biomaterialia 8 (2012) 937–962. [37] A. Haghshenas, M.M. Khonsari, Damage accumulation and crack initiation detection based on the evolution

AC C

EP

TE D

M AN U

SC

RI PT

of surface roughness parameters, Inter. J. Fatigue, 107 (2018) 130-144.

10

ACCEPTED MANUSCRIPT Figure captions Fig. 1 One-piece type dental implant specimen (a), its surface morphology after SLA treatment in the thread part (under low and high magnification) (b). Fig. 2 Dynamic loading apparatus for the fatigue test of dental implant specimen (a) and the illustration diagram of fixed implant and loading direction (b).

RI PT

Fig. 3 XRD pattern for analyzing the phase composition of Ti50Zr alloy.

Fig. 4 Optical microstructures of the as-received Ti16Zr (a) and Ti50Zr alloy (b).

Fig. 5 TEM images of Ti16Zr alloy showing high-density dislocations (a) and α platelets (b).

and intersection between slip bands and twins (d).

SC

Fig. 6 TEM images of Ti50Zr alloy showing fine α ' platelets (a), slip bands (b), nano-twins (c)

M AN U

Fig. 7 The engineering stress-strain curves and S-N curves of Ti16Zr and Ti50Zr alloy. Fig. 8 Load−number of cycling curves of the Ti16Zr and Ti50Zr dental implants. The specimens were tested in air (a), in artificial saliva solution (b), and in air with SLA treated surface (c), R=0.1.

Fig. 9 SEM images of the fracture surfaces and the corresponding high magnification images in the fatigue regions for the dry (a, b), wet (c, d) and SLA treated (e, f) Ti50Zr implants under 500 N

AC C

EP

TE D

maximum cycling load. The arrows designate the direction of crack initiation and propagation.

ACCEPTED MANUSCRIPT

(b)

AC C

EP

TE D

M AN U

Fig. 1

SC

10µm

RI PT

(a)

500 µm

ACCEPTED MANUSCRIPT

(b)

Load-bearing cap Dental implant

AC C

EP

TE D

M AN U

Fig. 2

SC

Epoxy resin inlay

RI PT

(a)

ACCEPTED MANUSCRIPT

(002)α

(100)α

30

40

(110)α

50

60

(112)α

70

M AN U

2θ /deg.

(103)α

SC

(102)α

RI PT

Intensity/cps

(101)α

AC C

EP

TE D

Fig. 3

80

ACCEPTED MANUSCRIPT (a)

50 µm

AC C

EP

TE D

M AN U

SC

Fig. 4

RI PT

(b)

50 µm

ACCEPTED MANUSCRIPT

(b)

(a)

α platelet

0.5µm 0.5 µm

AC C

EP

TE D

M AN U

Fig. 5

SC

0.5µm 0.5 µm

RI PT

dislocations

ACCEPTED MANUSCRIPT

(b)

(a)

acicular martensite

RI PT

slip bands

0.5 µm 0.5µm

(c)

(d)

SC

11 µm µm

2

M AN U

twins

slip band kink

0.5 µm 0.5µm

0.5 µm 0.5µm

AC C

EP

TE D

Fig. 6

twins

1

ACCEPTED MANUSCRIPT

1200

700

(a) 650

600 400 200

0.05

0.10

0.15

0.20

600 550 500 σ−1=500ΜPa

450 400

σ−1=380ΜPa

350

0.25

Strain

300

4

10

5

10

EP

TE D

M AN U

Fig. 7

6

10

Number of cycles

SC

0 0.00

Ti16Zr Ti50Zr

AC C

Stress/MPa

800

Ti50Zr Ti16Zr

(b)

RI PT

Stress amplitude σa/MPa

1000

10

7

ACCEPTED MANUSCRIPT

(a)

Ti16Zr Ti50Zr

800 700 600 500 400

400N

300

300N

200 1000

10000

100000

1000000

1E7

SC

Number of cycling

900

(b)

Ti16Zr Ti50Zr

800 700

M AN U

Maximum compression load /N

RI PT

Maximum compression load/N

900

600 500 400

400N

300

300N

200 1000

10000

100000

1000000

1E7

TE D

Number of cycling

800

Ti16Zr Ti50Zr

(c)

700

EP

Maximum compression load /N

900

600

AC C

500 400 300

350N

200 100 1000

200N

10000

100000

1000000

Number of cycling

Fig. 8

1E7

ACCEPTED MANUSCRIPT

(a)

(e)

(c)

Overload region

Overload region

Overload region

Fatigue region 1 mm

1 mm

(b)

10µm

M AN U

10µm

EP

TE D

Fig.9

AC C

1 mm

(f)

SC

(d)

RI PT

Fatigue region Fatigue region

10µm

ACCEPTED MANUSCRIPT Highlights

EP

TE D

M AN U

SC

RI PT

Commercial Ti16Zr alloy and new Ti50Zr alloy used for dental implant. Microstructures and strengthening mechanisms of Ti16Zr and Ti50Zr alloy. Fatigue behavior of Ti16Zr and Ti50Zr alloy under different conditions. Ti50Zr implant has better fatigue property than Ti16Zr implant.

AC C

1. 2. 3. 4.