The effect of bioinert electroexplosive coatings on stress distribution near the dental implant-bone interface

In this study, the first time a 2d finite element models of the titanium dental implant with Ti–Zr or Ti–Nb coating sprayed by electro explosive method and bone tissue located near were constructed. The present models simulate small surface implant section and bone located near. Three models with or without bioinert coating were studied in two configurations with cortical or cancellous bone tissue. All materials used in this study were assumed to be linearly elastic, homogenous, and isotropic to simplify the calculation. The stress distribution in the implant and bone tissue located near is uniform. The largest von Mises stress was obtained near the bone-implant interface in the implant area. It has shown that the stress pattern changed in the models with bioinert coatings. The second stress maximum appeared on the boundary between titanium subtract and the coating layer. The most significant changes in stress distribution were reached in the model with Ti–Zr coating. The electro explosive bioinert coatings help to reduce the stress shielding effect and implant failure probability because of bone strength loss. It also was found shear stress changes in the bone tissue.


Introduction
Currently, α-β-titanium-based alloy Ti6Al4V, also known as Grade 5 Ti64, is one of the most commonly used materials for the major part of implants on the medical market [1,2]. Aluminium and vanadium applying as alloying constituents and high Young's module are two significant disadvantages of this alloy. Ions release from the implant's surface because of corrosion can cause pathologic health issues. There are lots of studies [3,4] supporting cytotoxicity and carcinogenicity of aluminium and vanadium ions. Excessive aluminium intake increases the risk of breast cancer [5] and neurological conditions, such as Alzheimer's disease [6]. Oral or inhalation vanadium and vanadium compounds intake affect negatively on the respiratory system, blood parameters, liver, neurological system, and other organs, and also increase malignant processes [7,8]. There are lots of in vivo and in vitro studies supporting high safety qualities, corrosion resistance and biocompatibility of titanium alloys with niobium and zirconium additions [9,10]. J Ureña and co-authors [11] show the high biocompatibility of niobium coating spread by different methods. It was archived positive cell viability and cellmaterial interaction of the osteoblast-like cells (MG-63). In the study [12] it has been shown that binary Ti-Zr alloy with different Zr concentrations (5, 10 и 15 mass%) has better electrochemical behaviour in medium simulated body fluid properties.
The high Young's module of titanium alloy (≈110 GPa [13,14]) compared to that of human bone (≈10-30 GPa of cortical bone; ≈0.01-2 GPa of cancellous bone [15,16]) is much higher. It is another disadvantage of Ti6Al4V. The difference between Young's modulus of bone and implant leads to a change in force distribution between metal and bone. According to Wolff's Law, bone tissue near implant remodels in response to stress. This effect, known as stress shielding, increases the risk of implants failure [17].
The results achieved in the article [18] shown that Ti-Zr coating, sprayed by ionic-plasma deposition and containing 11 mas% and 22% of zirconium, 5 μm thick has Young's modulus from 77 to 98 GPa instead of titanium substrate Young's modulus equal 110 GPa. The same results were obtained in the study [19]. Young's modulus of Ti-Nb coating ranges around 53-64 GPa. Some studies show low Young's module coating changing force distribution in bone tissue located near the implant. For example, applying the Poiyactive ® coating on the implant, described in the article [20], reduces the compressive radial stress at the bone-implant interface around the neck of the implant by a factor by 6.6 times and the tensile radial stress by a factor by 3.6 times.
Electroexplosive method, developing nowadays intensively, is used for spraying different coatings [21,22] and also can be appeared for Ti-Zr Ti-Nb coatings development. The result archived by studying electro explosive Ti-Zr, Ti-Nb coating, shows that Young's modulus of these coatings is lower as compared to Ti6Al4V titanium alloy or commercially pure titanium Young's modulus [23]. However, the stress distribution between implant with electro explosive bioinert coatings and the bone study does not exist, because of it, this investigation is important. Finite element analysis (FEA) is one of the most effective and informative methods to studied problems associated with biomechanics. It supposes to avoid problems, occurring of analytical techniques, and also get higher-precision results [24,25].
The aim of this work is the determinate effect of bioinert electro-explosive coatings on stress distribution near the dental implant-bone interface by means of FEA.

Research materials and methods
A dental implant made of Ti6Al4V titanium alloy was used as the substrate. The explosive spraying of Ti-Zr and Ti-Nb coatings were carried out on the electro electroexplosive installation EVU 60/10M by an electric explosion of zirconium or niobium foils. The power density was 2.0 GW/m 2 . The weight of zirconium or niobium foils were 850 mg. The structure and morphology of sprayed coating and layer located near were analysed by means of scanning electron microscopy (Carl Zeiss EVO50 SEM).
The coatings' thickness was studied on the cross-sections by digital solutions Leica Application Suite. The thickness of the obtained coating is about 63 μm. The Young's module investigation of Ti-Zr and Ti-Nb coating was provided at low indenter load 50 mN by NHT-S-AX-000X Nano Indentation Tester. Nanoindentation was carried out on three lines at 5 μm from the surface-coating interface. The distance between indention is 10 μm. The next indention series consisting of 3 lines starts at 10 μm from the surface-coating interface. Thus, the average distance between indention is 5 μm.
The Young's module measure was provided by the Oliver-Pharr method by Nano Indentation Tester software. Under this method, the coating's elastic modulus is obtained from indentation load-displacement data found during one cycle of loading and unloading.
The Yong's module E of the studied material is evaluated as: where β in the range from 1.02 to 1.08 for different variants of indentation. The founded values S and A с allows to determinate effective Yong's module in contact E r . This parameter is related to elastic constants as: where E and E i are Yong's modulus; v and vi-Poisson coefficients of studied the material. Thus, in order to determine the elastic modulus and hardness using the Oliver-Pharr method, it is sufficient to find the contact stiffness S from the loading diagram.
where P is indention loading, and H is hardness. Equations don't depend on indentation depth, contact area, presence of the material roughness and allow to obtain elastic module Е. Obtained values were shown in table 1.
The Hounsfield values of cancellous and cortical bone density are 1362.94 and 472.21, respectively [27]. The relationship between HA and kg m −3 is · r = + a b H [28], where = a 527, = b 0.44 are coefficients, H is Hounsfield units. All materials properties are shown in table 1.
The three 2d models were created to investigate the effect of the electro-explosive bioinert coating on the stress distribution near the implant-bone interface. In the third model, also known as the reference model, the intermediate layer, as well as the substrate, is Ti-6Al-4V alloy. This model was used to evaluate the stress distribution near the implant-bone interface without electro-explosive bioinert coating.
The length of studying models is 1000 μm, the thickness is 300 μm, while the titanium surface thickness is 87 μm, the intermediate layer thickness is 63 μm as shown on figure 1, and bone layer thickness is 150 μm. All dimensions are shown in table 2. The boundary BE is fixed. The compressing force F1 parallel to the x-axis and equal 114.6 N. The bending strength F2 is parallel to the y-axis and equal 29N. The general force F is the geometric sum of F1 and F2, which value is 118.2 N [29]. Two forces are directed to the JF boundary on the other side of the models. The groups of boundaries AB, JC, HD, FE and AJ is free (figure 2). All models were developed and performed in COMSOL Multiphysics ® 5.5. Meshing plays a major role in the accuracy of the results. There are 745634 mesh elements. All materials were assumed to be homogenous, isotropic, and linearly elastic. Models specification are shown in table 2.
All calculations were carried out according to the theory of elasticity for the stationary case. Therefore, Newton's second law, which serves as the equilibrium equation, in tensor form has the form: The generalised Hooke's Law equation related stress tensor σ with deformation ε is given bellow: The Young's module and Passion ratio relate to Lame parameters as follows: The last required equation is the kinematic link between the displacement u and the deformation ε in the tensor form (Cauchy-Green strain tensor).
T is a transposition operation.

Research results and discussion
The obtained result may numerically vary from stress in implants and located near bone tissue, because all materials were assumed to be linear elastic and isotropic to simplify calculations. The difference between isotropic and anisotropic materials was shown in the article [26]. However, as in the articles [30,31], it was shown the unequal stress distribution between implant and bone. The higher von Mises stress locate in the implant area, rather than in bone tissue layer (figure 3). This difference can be described by higher Ti6Al4V Yong's module, using as a construction material of implants. Maximum von Mises stress locates on bone-implant interphase (boundary JC) and boundary FE nearly fixed boundary BE. The second region with maximum stress value exists because of the studied model geometry.
In the bone layer, the maximum von Mises stress values were obtained near point B on the fixed boundary BE. Among the models' variants with cortical tissue, the least von Mises stress in the bone layer was obtained in the reference models, in which the average and the maximum stress values are 0.3691 MPa and the 3. In the models' variants with cancellous bone tissue, the lowest von Mises stress was obtained in the reference models, in which the minimum stress value is 4.9102·10 −7 MPa, the average stress value is 0.1032 MPa, and the maximum stress value is 1.0927 MPa. The von Mises stress is larger in the model with Ti-Nb coating. The minimum stress value is 5.8782·10 −7 MPa, the average stress value is 0.1198 MPa, and the maximum stress  The minimum, average and maximum von Mises stress values in models with cancellous tissue are less than in models with the cortical bone. The minimum von Mises stress values differ between variants with tissues in 2.1846 times for the reference model, 2.0726 times for models with Ti-Nb coating and 1.8600 times for models Ti-Zr coating. The differences between average stress values in variants with cortical bone tissue and cancellous bone tissue are 3.5761 times for reference models, 3.2906 times for models with Ti-Nb middleware, and 3.1498 times for models with Ti-Zr middleware. The maximum stress values differ in 3.6292 times, 3.3366 times and 3.1930 times, thoroughly.
In  It also was found that stress in cortical bone tissue is higher than in cancellous tissue, as it was shown in the articles [32,33]. Instead of it, the interface stress is higher in the models with cancellous bone tissue. It can be described by higher cortical tissue Young's module. The studying of the model with electroexploive bioinert coating was shown that Von Mises stress increase in the bone tissue and decrease in the implant in models' variants with cortical tissue. This overview indicates that electroexplosive coating transfer stress into the bone tissue. The most significant effect was shown in the model with Ti-Zr coating.
So, the electroexplosive coating supposed to decreasing stress shielding effect and possibility of implant failure because of bone loss. As Ti alloys [34] with low Young's module, this coating decrease interface stress between bone and implant.
However, it is too early to state categorically, that the electroexploive coating decrease stress shielding. It needs to create more complex models taking into account bone and implant geometry and anisotropic properties of bone tissue.
It should be noted that, because of the different coating and subtract Young's module, the second stress maximum was found at the coating-subtract interface ( figure 5).
Among    However, the second stress maximum cannot cause coating lamination or cracking formation on the interface, since the Ti-Nb and Ti-Zr coatings possess high adhesion to Ti substrate because of the coating particle penetration into the substrate material during electroexplosive straying [35].
Using of bioinert coating have a direct impact on the shear stress on the interface between bone and implant, that is key to the mechanical transduction mechanism of osteoblast proliferation and differentiation [36,37].
In the studied model, on the major part of the interface between implant and bone, except the micrometres near the fixed board, shear stress is higher in variants with bioinert coating. However, near the fixed boundary, it was found that shear stress rapidly increases and the greatest values were found in the reference variant without bioinert coating ( figure 7).  The minimum shear stress values are larger in the variants with cortical bone tissue layer. In the reference models, the difference of the values between variants with cortical and cancellous bone layer is 2.9755 times. Values reached in models with electro explosive Ti-Nb, and Ti-Zr coatings differ in 23.5158 times and 4.3436 times, thoroughly. The average results are also larger in variants with a cortical bone layer. The average stress values differ in 3.4385 times for reference models, 3.1650 times in models with Ti-Nb coating, and 3.0313 times in models with Ti-Zr coating. The maximum values are higher in variants with cancellous bone tissue and differ in 0.3926 times 0.3498 times and 0.3283 times in the reference models, in the model with Ti-Nb and Ti-Zr coating, thoroughly.
The shear stress values decrease from the implant-bone boundary. At the same time, in the bone layer, the maximum shear stress is higher in models with bioinert coating. In the model's variants with cortical bone tissue, the largest maximum stress values were obtained in the model with Ti-Nb model. At the distance of 1 μm from the implant-bone boundary, for the variants with cortical the largest maximum shear stress is 0.2894 MPa ( figure 8). In the model with Ti-Zr coating, maximum shear stress decrease to 0.2861 MPa at the same distance. The least value of 0.2739 MPa was reached in the reference model. In view of the foregoing, it can be affirmed, that studied bioinert coating can supposed osseointegration mechanism by shear stress increasing.

Conclusion
In this study was found that the von Mises stress distributes in studied model unequal. The maximum values located in the implant area near the bone-implant interface (JC) and the boundary FE. The layer of bioinert Ti-Zr or Ti-Nb coating changes the stress distribution in the model. The Stress values decrease in implant area grow in bone tissue. These changes contribute to reducing stress shielding. It should be noted that vale increase near the coating-substrate interface. Ti-Zr coating has the most significant impact on stress distribution because of its lower Yong's module. The average shear stress values are higher in variants with bioinert coating. However, maximum stress values are higher in the reference model. These results were achieved for both bone tissue types. Despite the simplicity of the studied model, it possible to conclude, that studied coating can favourably change the life cycle of the implant.