Abstract
In this paper, a complex porous dental implant with biodegradable magnesium alloy was designed based on selective laser melting (SLM). Finite element analysis (FEA) was used to simulate the stress distribution of dental implant and alveolar bone in two models of preliminary and later stages of implant. The stress concentration area of dental implants was found not in the porous structure, and the weak part of mechanical properties accords with the work requirements. The porous structure of dental implants can promote the function of cancellous bone in the process of conducting the stress of the dental implant, thus improving the bearing capacity of dental implants. In vitro fatigue experiments were carried out on the experimental samples produced by 3D printing. Through the cell contrast experiment, it was proved that the decomposed Mg2+ could reach the titanium surface smoothly through the porous structure and complete the proliferation of osteoblasts.
Keywords
Introduction
With the rapid development of dental implant technology, more and more patients choose this technology to repair. 1 Dental implant restoration involves dentistry, mechanical engineering, computer science and other disciplines. It is a multidisciplinary biomedical engineering problem. 2 The scholars in the field of dentistry lack the experience of deep research on the related knowledge in the field of mechanical engineering, and are limited by the traditional machining methods, the structure of the implants on the market is relatively simple.
SLM is one of the metal three-dimensional printing technology and its unique material-adding manufacturing method is much better than the traditional manufacturing method in the structure of complex parts. 3 At present, SLM technology is also widely used in the manufacture of bone tissue engineering scaffolds and orthopedic implants. Some scholars have studied the microstructure, mechanical properties, corrosion properties and biological characteristics of ferromanganese scaffolds prepared by SLM. The overall results show that the porous Fe-35Mn implant made by SLM shows the application prospect of biodegradable load-bearing bone scaffold. 4 Titanium implants have better mechanical properties and biocompatibility. 5 , 6 The application of porous structure to bone implant materials may enhance the stability between the material and bone through potential growth into the pores. Among them, a porous design is added to the structure of the replacement joints to induce the bone cells to grow into pores to achieve a more powerful carrying capacity.7–9 At the same time, it helps to better integrate bone tissue and implant so as to improve its binding stability. 10 , 11 The study shows that the porous structure helps eliminate the bone resorption and failure of the implant, and the porous structure ensures the circulation of oxygen and nutrients in human skeleton. 12 , 13
Since the stress of dental implants can’t be directly observed in the host, the mechanical properties of dental implants are usually tested by the method of in vitro fatigue test. 14 However, the equivalent stress distribution in dental implants and alveolar bone is generally simulated by finite element analysis (FEA), and the results of equivalent stress are obtained. 15 , 16 According to the results of FEA, the biomechanical properties of dental implants and alveolar bone were evaluated comprehensively. 17 Magnesium alloy has similar mechanical strength and elastic modulus to natural bone. Magnesium alloy not only has the characteristics of biocompatibility and degradability, but also has a bright prospect in implantable medical devices such as bone graft and vascular stent. 18 , 19 Some scholars have designed magnesium implants with coral-like open cell porous inner and outer solid shells, which act as osteoid channels for tissue penetration and cell adhesion, while solid shells provide better structural strength and integrity. The results show that the strength of the new implant material is high, and the porous structure is beneficial to the growth of bone tissue and accelerate the healing process. 20 Magnesium alloy has the ability of self-degradation in human body, and the degradation of Mg2+ can affect the signal transduction mechanism in human bone-marrow stromal cells, then influences the extracellular matrix and transcription factors, thus promoting osteoblast proliferation and enhancing bone regeneration. 21 , 22 Some scholars studied the changes of mechanical properties of degradable magnesium alloy bone substitute made of magnesium alloy during the degradation process, and studied the degradation of porous implants by scanning electron microscope and nano-indentation technology. The results show that the decrease of magnesium concentration will lead to the decrease of average modulus, so that the degraded implant has lower stiffness than the original structure. 23 Therefore, in the process of magnesium alloy degradation, the mechanical properties of the implant will gradually decrease. 24
In this study, the porous structure was introduced into the structural design of dental implants. It was made by SLM technology and magnesium alloy was added to the manufacturing material of dental implants. By designing a special structure, the mechanical properties of dental implants could be ensured to meet the work requirements.
Materials and methods
FEA
Structural design and sample preparation of dental implants
The porous structure is introduced into the structural design of dental implants and combined with biodegradable magnesium alloy. In order to achieve this structure design, this study first analyzed the simplified model of dental implants. According to the theory of material mechanics, dental implants in the dental implant stress model were regarded as cantilever structures. Stress was mainly concentrated in the neck position. Therefore, the porous structure was designed below the neck to avoid the stress concentration area, and the central hole connected with the porous structure was designed to fill the magnesium alloy.
As shown in Figure 1(a), the specific structure of dental implants was designed through PRO/E 3D software (Pro/e3.0, Parametric Technology Corporation, USA). It is a segmental dental implant with a total length of 17 mm, a bone graft length of 10 mm, a gingival section of 2 mm, a base height of 5 mm, and a diameter of 4 mm. In the position of the bone graft, the outer thread is designed; the pitch is 1 mm. And in order to reduce the gravity influence on the thread precision when the 3D is printed, the right triangle thread is designed and the suspension angle is less than 45 degrees. The porous structure is distributed between the threads. The pore size is 0.6 mm, which is divided into two parts. The bottom pore is a regular distribution with a spring of 60 degrees, and the upper pore spring is irregularly distributed. At the same time, the upper pore channel is designed by arc to ensure the internal structure has enough mechanical strength. 25 A central hole with a diameter of 1.5 mm and a depth of 6 mm is designed inside the implant. The 3D model is saved as STL(stereolithography) format; the data is introduced into the 3D printer Concept Laser (Mlab 200 R, Concept Laser, Germany, Upper flangan) by adding support slices, and the experimental samples are produced by 3D printing with pure titanium (CP titanium) powders, as shown in Figure 1(b). After spraying sand on the surface of the sample, the magnesium alloy is placed in the central hole of dental implants, and the installation of magnesium alloy in the center hole is achieved through interference fit.

(A) A schematic diagram of the structure of a porous dental implant. (B) The physical map of dental implants produced by 3D printing: (a) An overlook of dental implants with unfilled magnesium alloy; (b) An overlook of dental implants after filling magnesium alloy.
The construction of a finite element model of dental implants
According to the experiment requirements, two models of dental implants were needed to simulate the stress of the dental implants and alveolar bone. 26 They were used to simulate the stress at the preliminary and later stages of an implant, and the later stage of an implant simulates the complete growth of bone tissue into the pores. The model consists of a dental implant system, cortical bone and cancellous bone. A model of the alveolar bone was obtained by intercepting the human natural jaw first, then by repair and stretching. As shown in Figure 2, it was composed of the 2 mm cortical bone wrapped with cancellous bone, and the thickness of the entire alveolar bone was 8 mm. At the same time, the experimental special crown was designed, which consisted of a cylinder and a hemisphere. The cylinder height was 5 mm, and the diameter was 10 mm; the diameter of the hemisphere was also 10 mm. The dental implant system in the preliminary stage of an implant model included a crown, a dental implant, and a magnesium alloy; the dental implant system included a crown and a dental implant in the later stage of an implant model. The alveolar bone model was combined with the dental implant system by the Boolean operation function of Magics software (Materialise, Leuven, Belgium) to obtain the final preliminary and later stages of an implant model. As shown in Figure 2(a) and (b), there were differences in the cancellous bone portions of the two alveolar bone models.

The construction process of finite element static analysis model of a dental implant. (a) A model of alveolar bone at the initial stage of dental implants. (b) A model of alveolar bone in dental implant.
Assuming that cortical bone and cancellous bone were in line with all same-sex requirements, the relevant parameters were set according to the model of A. Dorogoy. 27 The implant and crown were made of pure titanium (CP Ti), and the inner filler was magnesium alloy. 28 The parameters were shown in Table 1. The 3-matic STL software (Materialise, Leuven, Belgium) was used to divide the alveolar bone model with the implant model, in which the contact parts of the alveolar bone and the implant were divided into smaller body grids.
Young’s modulus, Poisson’s ratio and density for the test materials: CP Ti; magnesium alloy; cancellous and cortical bone.
In vitro fatigue test
The job of dental implants is to help patients chew food, and dental implants are mainly subjected to cyclic load in the oral cavity of the host.
29
According to the commonly used fatigue test method of dental implants, combined with the finite element model of dental implants, the fatigue test of the porous dental implants with the worst mechanical properties after the complete degradation of magnesium alloy was carried out. The experimental method, in Figure 3, showed the dental implant system was placed into a rigid clamp, and had a cyclic load with a vertical down size of 200 N and a loading frequency of 15 Hz.
30
After fixing the fixture, the central axis of the dental implant and the force loading direction of were in the angle of 30 degrees. Therefore, in Figure 3(b), the gingival section was 2 mm, and the height of the denture combined with the abutment was 5.2 mm. The horizontal distance between the central axis of the implant and the direction of stress is y =

In vitro fatigue test. (a) A physical fatigue test in vitro. (b) A schematic diagram of in vitro fatigue test: 1. Loading device; 2. Gingival segment; 3. Abutment; 4. Hemispherical loading member; 5. Bone graft; 6. Specimen holder.
Osteoblast proliferation experiment
The existing theory is sufficient to prove that magnesium ions can promote bone cell proliferation. However, it is necessary to further study whether the porous implants filled with magnesium alloys in this design can successfully deliver the decomposed magnesium ions through the complex holes to the implant surface and promote bone proliferation. Therefore, the porous titanium tablets used for the experiment were designed for the specific size of the cell test orifice, which was made by SLM technology, and the comparison experiment of bone cell proliferation was carried out. As shown in Figure 6(a), the diameter of the titanium plate is 8 mm and the thickness is 2.5 mm. The inner cavity is connected to the cell attachment surface, and the diameter of the cavity is 1.5 mm. The pore size is 0.6 mm, and the porous distribution is circumferential, with a normal spacing of 1 mm. The magnesium alloy was filled into the inner cavity through the hole on the circumference side of the titanium sheet, and the hole was sealed by using cyanoacrylate. The porous titanium slices filled with magnesium alloy and the porous titanium slices of the unfilled magnesium alloy were aseptic treated in the orifice plate respectively. The 1 ml mouse anterior osteoblast (MC3T3-E1) (ATCC, Xingzhi Biotechnology Co., LTD, China) was inoculated on the porous surface of the orifice plate with the initial concentration of 10,000/ml. After the first, fourth, seventh days, the samples were taken out and the floating cells were washed away, leaving only the cells adhered to the porous surface. The absorbance of each sample was measured at 450 nm using an enzyme calibrator after two hours of exposure to light by adding 10% Cell Counting Kit-8 (CCK-8) (DOJINDO, Japan) reagent 0.3 ml to each sample.
Results
FEA
As shown in Figure 4, two experimental models for dividing the good body mesh were introduced into the ANSYS FEA software, and the static load of 200 N with the vertical axis of 30 degrees was added to the static analysis, and their equivalent stress analysis was obtained.

Results of equivalent stress analysis of dental implant models. (a) The equivalent stress analysis result of implant model. (b) The result of the equivalent stress analysis of the alveolar bone. (c) The result of the equivalent stress analysis of dental implants.
As shown in Figure 4(a), the equivalent stress distribution diagram of the dental implant model can be seen that the two models had a quite similar equivalent stress distribution. The stress was mainly distributed within the dental implants, and the force direction along the static load was relatively distributed on both sides of the implant neck. The equivalent stress peak value in the preliminary stage of implant model was 150.18 MPa, and the later stage of implant model has an equivalent stress peak value of 137.03 MPa. As shown in Figure 4(b), the equivalent stress of the alveolar bone model showed that the equivalent stress distribution of the two models was similar, and the stress distribution was in the cortical bone near the position of the dental implant neck.
The equivalent stress diagram for dental implants is shown in Figure 4(c). It can be seen that the equivalent stress distribution of the two models was more similar, and the peak stress appeared in the position of the implant neck close to the cortical bone. The stress spread up and down from the peak stress point to form a stress concentration area, which was mainly compressive stress. And there was a smaller stress concentration zone in its relative position, which was mainly tensile stress. The edge of the stress concentration area of the two models extended to the porous section of the implant, and the stress range of the porous segment in the preliminary stage of implant model was obviously larger than that of the later stage of implant model. And the equivalent stress of the two models in the porous segment was about 20–0o MPa. The equivalent stress of the compressive and tensile stress concentration areas in the preliminary stage of implant model was about 65–5ou MPa, and the equivalent stress of the compressive and tensile stress concentration areas was about 60–0ou MPa.
In vitro fatigue test
Subsequently, an in vitro fatigue cycle test was carried out. After 11,844,266 cycles, the dental implant was completely fractured. As shown in Figure 5, the schematic diagram after fracture shows that there was a fracture near the junction between porous and solid dental implants. However, compared with conventional fatigue fracture, three fragments were generated, resulting in two cracks. Among them, the small internal debris was generated in the compressive stress concentration zone. At the same time, the highest point of the fracture section appeared in the tensile stress concentration zone, and the crack extended into the solid position. From the position of the abutment, the crack was observed downward. It can be seen that the crack at the junction of the tensile and compressive stress concentration zones was divided into two, then into the pressure stress concentration zone. As shown in Figure 5(a), the typical fatigue fracture morphology was observed in the section of No. 2 fragment with No. 1 fragments through the SEM, in which the crack source appeared at the position of the dental implants near the center hole. The shear head indicates the fatigue fracture process of the No. 2 fragment. Figure 5(d) is a section of No. 3 fragment and No. 1 fragment, which is adjacent to the crack propagation area of No. 2 fragment. It can be seen that the section shows brittle fracture morphology, and many irregular cracks can be observed.

The dental implant was broken into three segments: No. 1, No. 2 and No. 3. (A) The equivalent stress distribution of dental implants. (B) The analysis of fracture surface morphology: (a) The source of crack; (b) The crack propagation zone; (c) The final break zone; (d) The irregular crack zone.
Osteoblast proliferation experiment
As shown in Figure 6(b), the cell proliferation rate of the porous titanium sheet filled with magnesium alloy was significantly faster than that of the porous titanium sheet filled with unfilled magnesium alloy (P < 0.05). And it had reached its peak value on the fourth day of the experiment.

(a) A schematic diagram of the structure of a porous titanium sheet. (b) The absorbance values corresponding to the number of osteoblasts were compared on the first, fourth, seventh day of experiment.
Discussion
The results of the FEA showed that the stress is mainly concentrated in the neck position of the dental implants, which is in accordance with the results obtained by simplifying the force model to the cantilever beam. Therefore, the stress of the porous segment in the structural design proposed in this study was less, and the mechanical strength of the whole dental implant was guaranteed in theory. 31 Suppose that a person chews for 30 minutes a day, and the frequency of mastication is once per second. Then the chewing rate of this person is about 6.0 out05 a year. Therefore, according to the results of the fatigue experiment, it can be found that the dental implant proposed in this study can guarantee the continuous use of a patient for 20 years. Therefore, the mechanical properties of dental implants proposed in this study are in line with the requirements of dental implants (only the patients without bruxism). 31 , 32
The peak value of the alveolar bone and dental implants in the preliminary stage of implant was compared with the peak value of the dental implants in the later stage of implant. As shown in Figure 7, the peak stress of the alveolar bone and dental implants in the later stage of implant model was less than that in the preliminary stage of implant model. The only difference between the two models is that the cancellous bone in the later stage of implant model grows into the pores of the dental implants. This shows that although the FEA pointed out that the stress of the alveolar bone in the two models almost all appears in the cortical bone, the cancellous bone also plays a vital role. Therefore, the porous structure of dental implants can promote the effect of the cancellous bone in the stress of the dental implant, thus improving the force carrying capacity of the dental implants and prolonging its service life.

Comparison of equivalent stress peaks between alveolar bone and dental implants in the preliminary and later stages of implant models.
According to the equivalent stress attempt of the preliminary and later stages of implant models, it can be seen that there were mainly two stress concentration areas inside the dental implants which bear a static load. That is, the compressive and tensile stress concentration areas, were relatively distributed, and the action range of the compressive stress concentration area was larger than the tensile stress concentration area. Because the vertical component of static load was compressive stress on dental implants, the compressive stress in dental implant was greater than tensile stress. At the same time, the compressive yield strength of 3D printing workpiece is less than tensile yield strength. It indicates that the compressive stress concentration area of dental implants is the most prone to fatigue area, in which it is consistent with the results of in vitro fatigue test. 33 The working principle of 3D printing is stacking up layer by layer, and the combination defects between layers will accumulate, resulting in a poor mechanical performance in the vertical direction after forming. Therefore, in the in vitro fatigue test, when the dental implant was fatigue fracture, there were small fragments in the compressive stress concentration area. In the results of in vitro fatigue test, the crack extension in the tensile stress concentration zone entered the solid part, indicating that the tensile stress expanded the damage to the dental implants by extending the crack in the fatigue fracture. In the tensile stress concentration zone, the fracture surface and the horizontal plane showed a 45 degrees angle, and it broke along the direction of the maximum shear force, indicating that the shear force was the main cause of the damage.
According to the SEM in Figure 5, the fatigue fracture was found in the compressive stress concentration area, and the specific crack source area was far from the peak value of the equivalent stress. It indicates that the fatigue life of the actual fracture position was the lowest. The fatigue fracture location was suspended in the 3D printing station. Although the support was added in the manufacture, its mechanical properties must be reduced. At the same time, in the later period of in vitro fatigue test, the dental implants were also fretting in the process of cyclic loading. This micro movement caused the interface stress of between the compressive and tensile stress concentration zones in the dental implant, which was constantly changing between the compressive stress and the tensile stress, then increasing the stress amplitude directly, thus shortening the fatigue life of the position. Therefore, the crack source of fatigue fracture in theory should be in the boundary area between the internal compressive and tensile stress concentration zones. It was unified with the results obtained from the section analysis. The load of dental implants will transmit the load into the alveolar bone after the load is loaded in the host, and the strain produced by the alveolar bone makes the dental implants fretting, which is consistent with the micro movement in the in vitro fatigue test.
According to Figure 5(d), it is inferred from the multiple cracks in the section of the No. 3 fragment that it was subjected to a larger load impact before fracture. Through the process diagram of the cyclic experiment, we can see that the fatigue test cycle was about 5,000,000 times, and the dental implants had a large displacement in a short time and then tended to be stable. It is inferred that this is the fatigue crack source in the dental implants and the first fracture, and the fracture occurred between No. 1 and 2 fragment. At about 5,000,000 to 10,000,000 cycles, the displacement of dental implants increased slowly, indicating that the cracks in dental implants gradually expanded. In the process of crack propagation, the loads on the implants and adjacent uncracked areas will gradually increase. At about 10,000,000 cycles, dental implants had the small displacement of 0.1 mm occurred in a short time. It is inferred that after the increase of the great extreme value of the internal load of dental implants, there are second cracks in the interior of the implant. The fracture occurs between No. 1 and 2. Large load impact accelerated the fracture process of dental implants. After 11,844,266 cycles, the final fracture of dental implants occurred and three fragments were formed. From the above analysis, we can infer that the last fracture is mainly the larger shear force, resulting in the brittle fracture between the No. 1 and No. 3 fragments.
Compared with the unfilled titanium sheet, the proliferation rate of the cells on the titanium sheet filled with magnesium alloy was significantly faster, which indicated that the decomposed Mg2+ reached the surface of titanium sheet smoothly through the porous structure, and completed the proliferation of osteocytes. Initial stability is the key to the success of dental implants, and the prerequisite to ensure initial stability is the osseointegration of dental implants with alveolar bone. 34 Therefore, Mg2+ can accelerate the process of osseointegration by promoting the proliferation of osteoblasts, thus Mg ions are very beneficial to improve the early secondary stability of implants. On the seventh day, the cells on the titanium sheet filled with magnesium alloy almost stopped proliferation, which was affected by the limitation of the surface area of titanium sheet, so that the number of cells on the surface reached the peak.
Conclusion
The mechanical properties of the complex porous dental implants in this paper met the life requirements of patients (only the patients without bruxism). The porous structure has good biocompatibility and can effectively avoid the stress shielding phenomenon. The magnesium ions produced by the magnesium alloy can promote the proliferation of bone cells, and the process of osseointegration can be accelerated in theory. At the same time, when the bone cells grow into the porous structure of dental implants, the equivalent stress of alveolar bone decreases, and indirectly increases the load bearing capacity of dental implants. Based on the fatigue fracture of the dental implants produced by the SLM technology, the crack source appears at the junction area of the compressive and tensile stress concentration areas, and is very easy to form small fragments in its compressive stress concentration area.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is supported by Science and Technology Planning Project of Guangdong province, China (No.2017B090901039); Science & Technology Plan Project of Guangzhou (201710010193).
