Study on Mechanical Property and Biocompatibility in Vitro of 3D Printing Tantalum- Niobium Alloy Implants


 tooth defects or deletions. Tantalum (Ta) has been widely used in the biomedical field, but its application in artificial dental implants is rarely reported. In this study, Tantalum-Niobium alloy (TaNb40) implant prepared by 3D printing—SLS was used as the research subject. Its mechanical properties in vitro and the biocompatibility by utilizing human Oral Mucosa Fibroblasts (hOMF) were studied. Results: The mechanical property test results were that the Tensile Strength, Yield Strength, Elongation, and Vickers hardness of TaNb40 implant was 548 ± 50MPa, 420 ± 30MPa, 40%, and 425HV, respectively, manifesting the two indicators met the requirements of dental metal implant, and had good mechanical properties of wear resistance and not easy to brittle fracture. The cytocompatibility test showed that TaNb40 did not inhibit cell proliferation and produce cell cycle arrest. The cytotoxicity grade was 1, there was almost no metal ion release in the culture medium, and the surface had fine cell early adhesion, which met the requirements of biomedical materials. Conclusion: The TaNb40 prepared by domestic 3D printing—SLS has excellent mechanical properties and cytocompatibility in vitro, which is expected to replace the Titanium metal implant prepared by the traditional casting method, having a leading significance for the formation of implants with independent intellectual property rights.

in joint and spinal surgery, etc. [26][27][28][29]. At present, Porous Tantalum Trabecular Metal (PTTM) has been incorporated into Ti alloy dental implants to improve osseointegration in many studies, but Ta as an arti cial root is rarely reported. Solid pure Ta has high mechanical strength and density (16.68g/cm3), therefore, porous structure involving porosity and pore diameter has obvious advantages in this aspect. By adjusting the porosity and pore size of the material, the "stress shielding" is reduced or eliminated, so that facilitating its elastic modulus to match that of human bone tissue. However, high melting temperature, a strong a nity for oxygen, di culty in material preparation, and high cost make it very di cult to prepare Ta through ordinary methods, which limits the wide application of Ta [30]. Ta and Niobium (Nb) are geared to the same category in the periodic table of elements. Nb has a lower melting temperature (2477℃), elastic modulus, and raw material cost than Ta and Nb is a strong β stabilizer [30][31]. Therefore, the Ta-Nb alloy has a lower elastic modulus than the Ta scaffold, without inhibiting its corrosion resistance or introducing any cytotoxic components, and reducing the manufacturing cost.
3D printing is a highly adjustable and complex device matching with human anatomy position, which is suitable for patients and has become a leading manufacturing technology in the eld of health care and medicine [32]. 3D printing, also known as additive manufacturing or additive rapid prototyping, can rapidly and accurately manufacture 3D stacked objects of complex shape under the control of the computer through the precise design of models [33]. At present, Stereolithography (SLA), Fused Deposition Modeling (FDM), Selective Laser Sintering (SLS), and Three-dimensional printing (3DP) are common methods in 3D printing [34]. Among them, SLS is a kind of technology that makes use of powder and computer-control to rapidly separate and produce objects. It uses a laser to combine powder particles, building them layer by layer, and then gain them from under the powder bed [35][36]. Compared with FDM, SLS is a single-step process that does not require hot-melt extrusion to pre-produce suitable laments and produces higher resolution ratio objects due to laser accuracy [37][38][39]. 3D printing is gradually rising in medicine. Brito [40] reported the advantages of additive manufacturing in mandibular fracture surgery. Mafeld [41] carried out the feasibility study of the 3D printing vascular model. Ren [42] used 3D printing to design a calcaneal surgical guide plates to improve the surgical treatment scheme.
Tetsworth [43] applied 3D printing to the reconstruction of complex post-traumatic osteoplasty and achieved ideal results. Personalized biomedical materials can be prepared by 3D printing to meet the needs of different patients. It has broad prospects in terms of tissue scaffold, cell printing, prosthesis.
The 3D printing in the research and development of biomaterials involves the following key technologies: rstly, digital three-dimensional Computer-Aided Design (medical prototype object simulation) system(CAD), which transforms CAD model into "clone" products of real tissues or organs; secondly, medical biomaterials with 3D printing; thirdly, the technology and equipment suitable for biomedical materials printing. These are indispensable.
Although 3D printing has gained the attention of many domestic and foreign scholars, especially in the preparation of biomedical materials, the clinical application of 3D printing still faces many challenges, which are still in the initial stage. The rst problem is CAD software and the selection of 3D printing raw materials including the mechanical properties, biocompatibility and activity retention, etc. Secondly, if 3D printing is applied to living cells or tissues, the survival rate of cells on the surface or inside of the product needs to be maintained during the whole processing. Finally, we need to clarify the mechanism of cell adhesion, growth, and differentiation inside the material.
To develop new-type material implants with independent intellectual property rights, reduce or eliminate   the dependence on imports, and provide Chinese with suitable prices, multiple choices, and better  performance implants, Ta-Nb alloy implant taking Ta as the main component and adding the

General observation of experimental materials
Tantalum-Niobium alloy (TaNb40) implant material was prepared by SLS. SLS laser parameters were 250W. Its size is 4.5mm in diameter, height 12.2mm, and weight 2.2g (without abutment). The shape is similar to that of Ti implant commonly used in the clinic ( Fig. 1 and Fig. 2). It can be seen from Fig. 1 that the surface of the TaNb40 alloy is bright gray with a certain degree of roughness, which is the original shape that has not been surface treated. To facilitate the co-culture of experimental materials and cells, and the TaNb40 alloy disc prepared by SLS was used to replace the implant for an experiment in vitro. The surface of the substitute was bright silver (Fig. 3). There is little difference in appearance between Ta, Ti, and Ni metal samples prepared by the traditional casting method (Fig. 4).
The mechanical properties of TaNb40 implant TaNb40 implant was tested by a Third-Party Inspection Agency and placed on a tensile tester to detect the tensile property of the material, including TS (Mpa), Yield Strength (YS) (Mpa), and Elongation (%), a routine test item of metal material. The results were shown in Table 1. The TS, YS, and Elongation of TaNb40 implant were 548 ± 50MPa, 420 ± 30MPa, and 40%, respectively, besides, the Vickers hardness was 425HV.

Growth of Cells
The morphology of primary hOMF cultured to the 5 th -6 th generation is shown in , the cell volume increased, and the processes fused into a network, increased cell density and collagen matrix secretion could be observed clearly after HE staining. In this experiment, the 5 th -6 th generation cells as samples grew well, whose morphology was consistent with that of normal broblasts.

Direct effects of materials on cells
As shown in Fig.6, compared with the normal control in Fig Ta groups were densely distributed around the materials, and a small amount of cell apoptotic fragments was observed. There was no cell attachment in the area about 0.1 mm around the Ti sheet, and the cells outside this region grew and arranged well. According to the cytotoxicity scoring standard in vitro, the score of these three materials can be recorded as 0. In the Ni group, about 80% of the cells were round with no protruding pseudopodia, the number of cells decreased as the dissolution of some cells, and the above morphological changes became more obvious over time, there were almost no normal adherent cells on the 5 th day, the cytotoxicity was grade 4, which was material with severe cytotoxicity. (3) The cells at different distances from the same material on the same day showed that there was no signi cant difference between TaNb40 and Ta at three-time points, but the number of cells far away from Ti seemed to be slightly more than that around the material. In the Ni group, it seems that the cells farther away from the material were in better condition at the same time than those near the material.

Effects of materials on cell early adhesion
The micrographs of adherent cells after AO uorescence staining were shown in Fig.7. In the TaNb40 and Ta group, the cell showed a large amount of clear green uorescence, which was evenly distributed on the surface of the material, and the number of adhesion cells increased signi cantly with time. In the Ti group, most showed red and yellow uorescence, which represented the difference of binding capacity between staining agent and DNA. The cells were scattered in different visual elds, and a small number of cell fragments and apoptotic bodies could be seen, the amount at the 12 th hour was signi cantly more than that at the 4 th and 6 th hour. In the Ni group, the quantity was signi cantly less than that of the other three groups at each time point. It can be seen that the chromatin shrinks and breaks into patches of varying sizes, showing the shape of green fragments, and only a few cells adhered at three-time points.
According to the quantitative analysis in Table 2 and Fig.8, the comparison of cell adhesion rate was successively Ti > Ta > TaNb40 > Ni and the difference between Ti and Ta, Ti and TaNb40 was statistically signi cant (P < 0.05). At the 6 th and 12 th hour, the cell adhesion rate successively was Ta > TaNb40 > Ti > Ni, there were signi cant differences between Ta and Ti, TaNb40 and Ti (P < 0.01). The adhesion rate increased with time in TaNb40, Ta and Ti group, and the highest at the 12 th hour. No signi cant difference between the 4 th and 6 th hour in the Ti group (P > 0.05), but it increased signi cantly between the 6 th and 12 th hour (P < 0.01). However, The adhesion rate of the Ni group was very low at all time points, and the mean value of the TaNb40 and Ta group exceeded 100% at the 12 th hour. It can be seen that both TaNb40 and Ta have no effect on cell adhesion, and the initial adhesion process of surface cells was even better than that of Ti, and can quickly adhere to the surface in the early stage.

Effects of materials on cell proliferation
To evaluate the effect of different material extracting liquid on cell proliferation, using CCK-8 reagent and enzyme labeling to detection. According to the quantitative analysis results in Fig.9, on the 1 st day, the cell proliferation activity (OD value) was as follows: blank control > Ta > Ti > TaNb40 > Ni, Ta and Ti group (P < 0.05), TaNb40 and Ti group (P < 0.01). On the 3 rd day, the comparison was blank control > Ta > Ti > TaNb40 > Ni, there was statistical difference between Ti and Ta Group (P < 0.05), but there is not between Ti and TaNb40 group (P > 0.05). On the 5 th day, the order successively was blank control group > Ti > Ta > TaNb40 > Ni, Ti and TaNb40 group (P < 0.01), Ti and Ta Group (P > 0.05). The OD value of the blank control group at each time point showed the highest cell activity (P < 0.01). There was no statistical difference in the OD value of the Ni group at three-time points. As can be seen from Fig.9, the number of cells did not increase but also decreased with time.
The RGR and cytotoxicity grade as shown in Table 3 were calculated from the OD value. The Ta, TaNb40, and Ti group was grade 1, which met the requirements of medical biomaterials, while the Ni group was grade 3, which did not conform to the speci cation of medical biomaterials.

Effects of materials on cell cycle
As shown in Fig.10, the cell cycle distribution was detected by ow cytometry after 48 hours of treatment with different extracting liquid, and the data were listed in Table 4. The one-way ANOVA analysis on the percentage of cells in G1, G2, and S phase among the ve groups in Table 4 indicated there were signi cant differences among the phase of each cycle between groups(P < 0.01). However, compared with the blank control group, the percentage of G1 / S phase cells tended to increase and the G2 phase decreased in the Ta Group. The TaNb40 group increased in the G1 phase and decreased in S / G2 phase, although the cell cycle percentage of the TaNb40 group was statistically different compared with a blank control group, it tended to be similar in numerical value, and there was no signi cant increase or decrease. In the Ti group, the G1 phase decreased and S / G2 phase increased. In the Ni group, the percentage of S-phase cells increased signi cantly, and Ni induced S phase arrest accompanied by a decrease in the percentage of G1 phase cells. Compared with 3.11% of the blank control group, the cells treated with Ni had the highest apoptosis percentage of 3.91% in the sub-G1 phase.

Detection of metal ion content
The content of metal ions in the four kinds of material extracting liquid was listed in Table 5. The amount of ions released by Ni was the highest, up to 264 mg/kg. The metal ions were hardly detected in other groups. In the reference data of detection limit value, the amount of the ions of TaNb40 and Ta Group were less than 0.1mg/kg, and that of Ti group were less than 0.2mg/kg.

Discussion
At present, Ta is widely used in the biomedical eld, especially in orthopedic metal implants, which attracts the attention of researchers in the eld of biomaterials. In recent years, PTTM has been incorporated into Ti alloy dental implants to improve osseointegration. Compared with traditional bone implant materials, Ta has high corrosion resistance, hardness, and wear resistance, which can avoid the adverse biological effects to the most extent caused by the release of metal ions in vivo. On the other hand, Ta has strong osteoinductive and osseointegration abilities in vivo, even better than Ti dental implants, which can improve the stability of implants at the early stage [44][45][46]. However, Ta is a relatively active element, the powder particle size is small, and it is easy to oxidize at high temperature, so it is di cult to prepare, at the same time, the mechanical strength and density of solid pure Ta are very high. Adding Nb element can improve the mechanical properties of the material, reduce the sintering temperature and manufacturing cost, and promote the sintering of Ta at high temperature, to meet the clinical requirements of dental implant material. Also, the traditional processing technology is di cult to prepare Ta, therefore, SLS was used to meet the personalized requirements in the early stage of this study. The surface of TaNb40 alloy prepared by SLS is brightly gray with a certain roughness, moreover, the oxygen content is lower than 40ppm in the degreasing environment, and there is no oxidation phenomenon. As a load-bearing part of the body, the bio-metal orthopedic implant must have good mechanical properties. In addition to evaluating the stability of implant through bone integration, the matched mechanical properties between the alloy and bone tissue are conducive to the uniform distribution of force transmission to the alloy and surrounding bone tissue, and prevent the generation of "stress shielding" [47]. Mechanical properties and biocompatibility are the primary factors to be considered in the design of new materials [48]. Especially in the dental implant, they will be exposed to the alternating stress of load and fatigue when performing their functions [49]. The strength of the alloy shall be high enough to withstand external forces including tension, compression, bending and torsion [48,50]. TS and YS are the basic properties of hard tissue replacement materials, which can prevent the plastic deformation of materials in the process of implantation, to ensure stability in bone tissue [51]. The technical standard ASTM F67 divided pure Ti implant into four grades: G1-G4, with YS range of 170-483 MPa and TS of 240-550 Mpa [51]. The YS of Ti alloy (Ti6Al4V) implant is signi cantly higher than that of pure Ti, and the YS of Ti alloy prepared by different processes is in the range of 360-3267 MPa [52]. The YS and TS of TaNb40 alloy detected in this study were 420 ± 30 MPa and 548 ± 50 MPa respectively, compared with the pure Ti and Ti alloy recorded in the literature, these two aspects of TaNb40 implant are within the range of mechanical performance indicators, and have good mechanical properties. The Elongation of the TaNb40 alloy is 40%, which is higher than pure Ti and Ti alloy implant based on the relevant literature [51], indicating that the TaNb40 implant has the characteristic of not easy brittle fracture. Besides, compared with Ti material, the Vickers hardness value (425HV) of TaNb40 was higher than that of pure Ti and Ti alloy implant [51]. Hardness refers to the resistance of the material to permanent distortion. It is not too much to say Hardness is an important indicator of wear resistance. Besides, the elastic modulus is also an important property, if it is far greater than that of bone tissue, the difference may produce greater stress at the bone-implant interface in the process of load transmission, resulting in bone loss and implant loosening failure. In our previous studies, through a series of operations such as the design of computer-aided software, the preparation of SLS raw material powder, and the control of processing conditions, the TaNb40 implant was successfully prepared. In this study, the YS and Hardness were preliminarily determined. However, as a medical biomaterial implant, it is still necessary to further detect compressive stress, shear force, elastic modulus, and fatigue strength in the future because the stress in the jaw is very complex, which could provide the basis for adjusting the parameters of 3D printing properly so that all the mechanical properties of the implant can enough to match the bone tissue in vivo.
Good biocompatibility is the prerequisite and foundation of Ta application in the biomedical eld. At present, the evaluation of biocompatibility of biomaterials mainly refers to International Organization for Standardization (ISO) 10993 and national standard GB / T16886, including the utilization of different experimental methods in vitro and in vivo. Researchers can choose some of methods to achieve their own research objectives. It can be concluded whether the material conforms to the biological safety and functionality through a series of evaluations, in other words, the material has no toxic effect on the human body and does not cause host heterologous recognition reaction. Besides, the biomaterials should be able to perform the corresponding functions in the speci c parts, not be rejected and destroyed, maintain their original physicochemical, mechanical, and biological properties, and have a long-term good combination with the host. Therefore, the other focus of this study is the cytotoxicity of metal materials in vitro. The cytotoxicity test is one important index to evaluate the biocompatibility, which is the most basic experimental method in biological performance evaluation of material. In this study, we observed the effects of TaNb40 and Ta on the morphology, early adhesion, proliferation, and cell cycle of hOMF in vitro, moreover, widely recognized metal pure Ti with good biocompatibility as positive control and pure Ni as negative control were used. After initial implant stability, it is also very necessary to close the soft tissue wound in the second stage of implant surgery, which can promote the healing of soft tissue and prevent the infection caused by a microorganism and other exogenous substances. The healing of oral mucosa soft tissue had to do with the broblast proliferation and collagen deposition [53]. Due to the poor antigenicity of broblasts and the nonspeci city of antigens, some studies tried to identify broblasts based on immunohistochemical staining, but the rate of success rate is very low [54][55][56]. Therefore, no immunohistochemical method was used to identify broblasts in this study.
Ultrastructurally, broblasts are identi ed by their stellate appearance with slender branching pseudopods and have marked rough endoplasmic reticulum(RER) and Golgi complex [57][58]. We observed the hOMF by light microscope, and the results were consistent with the morphology of normal broblasts.
To direct observation of the number and morphological changes of the cells in contact with the materials, the direct contact method was used to detect the cytotoxicity. On the 5th day, a small number of apoptotic fragments were observed around the TaNb40 Ta Ti group, which was the normal apoptosis during the growth of cells. In the Ti group, there was no cell attachment in the area about 0.1 mm around the metal sheet, because the weight of Ti (0.2756g) was less than that of Ta (0.6964g) and TaNb40 (0.7007g) when inoculated with the same volume of culture medium with the same number of cells, Ti would not adhere rmly to the plate due to buoyancy, which affected the surrounding cells to a certain extent. According to the cytotoxicity scoring in vitro, TaNb40, Ta, and Ti were recorded as 0, which did not affect cell growth and had no toxicity to cells. In the Ni group, about 80% of cells were suspended dead cells on the 5th day, the cytotoxicity score was 4, which was considered as a serious toxicity material.
Due to the different number and morphology of cells in different distances from the same material, it can be inferred that the cell is affected by some metal elements released from the material, which makes the metal ions distribute unevenly in the culture medium, thus affecting the growth of surrounding cells. Toxicity test is to detect the effects of small molecular substances on cells when materials are degraded or decomposed, the adverse reactions in local tissues are related to the metal ions released by material [59]. Therefore, the cytocompatibility in vitro can also be analyzed by the indirect contact, that is, the extracting liquid method. By ISO 10993-5:2009, the effects of TaNb40, Ta, Ti, and Ni extracting liquid on cell proliferation were tested. CCK-8 can react with enzymes in mitochondria of living cells and display orange-yellow in solution, therefore, the more living cells, the greater the degree of reaction with reagent, the deeper the orange color. The higher the OD value indicating that the cytotoxicity of the material is smaller. The OD value measured by Microplate Reader can re ect the cell activity and cell proliferation. The constant uctuation of OD values of TaNb40, Ta, and Ti at three-time points may be the unintentional damage to cells caused by the gun head and the error of the testing instrument. The statistical difference between each of the three group and the blank control group indicates that either these three materials do have a slight impact on cells, or the initial number of cells in the blank control group is more than that in the TaNb40, Ta and Ti group due to the counting error, so go a step further analysis the cell cycle distribution. Although there was the statistical difference, the cytotoxicity analysis showed that TaNb40, Ta, and Ti did not produce cytotoxicity, and the grade was 1, which met the requirements of medical biomaterials. However, the OD value of the Ni group was lower than 0.5 at the outset, and the cell proliferation was inhibited over time, which was consistent with the results of direct contact morphological analysis. The cytotoxicity of the Ni group was grade 3, which did not conform to the standard of medical biomaterials.
The surface properties of the material, such as surface energy, hydrophobicity/hydrophilicity, net charge, and roughness, affect the adsorption of cells and macromolecules on the surface after implantation [60].
To observe the effect of the surface properties on the early adhesion of hOMF, an AO reagent was used to stain cells and observe the early cell adhesion at the 4 th , 6 th , and 12 th hour. AO is a uorescent dye used to observe the number of adhesion cells on the surface of opaque material, the reagents emit different colors of uorescence by binding with nuclear DNA and RNA, so the adhesion ability of cells can be evaluated. The cell adhesion rate of TaNb40, Ta, and Ti group increased with time, and the maximum was found at the 12 th hour, the TaNb40 and Ta increased signi cantly at three-time points (P < 0.01), indicating that they did not affect the cell adhesion. Compared with the other three groups, the initial adhesion rate of the Ni group was very low, and as time goes on, not only no cells attached, but also the cells adhering on the surface of the material also showed lysis apoptosis, and the adhesion rate was lower. At the 4 th hour, the cell adhesion rate of the Ti group was higher than that of TaNb40 and Ta group (P < 0.05), at the 6 th and 12 th hour, the TaNb40 and Ta were higher than that of the Ti group, and the difference was statistically signi cant (P < 0.01), which make clear that the initial adhesion of cells of the TaNb40 and Ta is better than that of Ti, and can quickly stick to the surface in the early stage, this result may be related to the roughness surface prepared by SLS technology. Studies have shown that the surface topography ( at, rough, nanometer) and crystal size of biomaterials affect the interaction between cell and material interface, and then have an effect on their biological characteristics [61]. The low level of cell differentiation in the at morphology might be due to its more inert surface, resulting in less cellular reaction [62]. The roughness of the nanoparticle size can intervene in the behavior of the cells. Some scholars [63][64] prepared bulk ultra ne-grained pure Ta by Equal Channel Angular Pressure (ECAP), after treatment, the number of grain boundaries on the surface increased, providing more cell adhesion sites, showing better cell activity and biocompatibility. The nano pits can change the surface roughness, wettability, and the adsorption capacity of bronectin, it could also be recognized by broblasts, especially when the size of nano pits is 50 nm and 60 nm respectively, which can signi cantly enhance the attachment and proliferation of broblasts [65]. The TaNb40 surface showed a certain roughness whose pit size could not be determined due to without measurement of the surface microstructure. It can be seen that the early adhesion rate of cells on the rough surface of TaNb40 prepared by SLS is higher than that on the Ti surface from our experimental results, but the error in the operation is also not ruled out, which is related to the inherent chemical properties of Ta surface. Some scholars have studied the effect of inherent chemical properties of Ti and Ta on Bone Marrow Mesenchymal Stem Cells (BMSCs), the results showed that the expression of integrin α5 and β1 on Ta surface was higher. Integrin plays an important role in the formation of focal adhesion complex, mediates intracellular signal transduction, and regulates the process of cell differentiation [66]. Other studies have shown that [44][45][46] Ta has better osteogenic differentiation than Ti in prosthesis and implant coating. The surface characteristics of materials, such as surface structure and physicochemical properties, can signi cantly affect the behavior of cells and subsequently bone induction and osseointegration. At the 12 th hour, the average number of adherent cells on the Ta and TaNb40 group was more than 100%, indicating that not only the number of adherent cells on the surface of the material increased signi cantly but also some cells were in the early stage of proliferation at this time point. In a word, the TaNb40 implant or disk-shaped material made by SLS has no surface treatment, which does has better cell adhesion, even better than Ti. Therefore, it can be inferred that it may not require further surface treatment, and the relatively rough surface is conducive to cell adhesion and growth.
To determine whether the cell cycle was blocked by different materials and which cell cycle stage is inhibited by Ni, the cells were stained with Propidium Iodide (PI), so the DNA content was detected by ow cytometry. Various cell cycle stages have different DNA ploidy numbers, according to the distribution of cells in each phase of the cell cycle, we can judge whether the materials have no, promote or inhibit proliferation effect on cells. The cells were treated with the extracting liquid after 48 hours. Compared with the blank control group, the percentage of G1 / S phase cells in the Ta group tended to increase, the G2 phase decreased, indicating that DNA synthesis accelerated, thus promoting DNA replication and cell proliferation. In the TaNb40 group, the results indicated that the cells were in normal growth and proliferation stage. In the Ti group, the G1 phase decreased and S / G2 phase increased, the cells were in G1 to S phase transformation stage combined with cell proliferation results, which promoted cell proliferation. The cell proliferation results between the TaNb40, Ta, Ti, and blank control at three-time points were statistically different (P < 0.01), but there was no cell cycle arrest. The possible reason is that the initial number of cells in the blank control group was more than that in these three groups due to the error of cell count, thereby displaying the statistical difference of OD value in each group. Compared with 3.11% of the blank control group, the percentage of S-phase cells increased signi cantly in the Ni group, and 3.91% of cells in the sub-G1 phase were apoptotic. Therefore, Ni induced cell cycle arrest in the S phase and promoted cell apoptosis in this study. The ion content of Ni extracting liquid was 264mg/kg.
Qiao et al. [67] found that 10 um Ni could cause DNA damage in vitro. Besides, it also was found that [68] Ni nanoparticles (NiNPs, size: 28nm) decreased the survival rate of human liver (HepG2) cells in a dosedependent manner in the concentration range of 25-100μg / ml. DNA damaged cells accumulate at Gap1 (G1), DNA synthesis (s), or Gap2 / Mitosis (G2/M). Studies have shown that [68][69] Nickel chloride (NiCl2) induces G2 / M arrest of the liver. A large number of metal ions accumulated in the body can induce a series of pro-in ammatory reactions. A kind of metal ion can activate a variety of intracellular signaling pathways to mediate the release of cytokines, promote the formation of the local in ammatory response and initiate cell apoptosis mediated by mitochondria. If this immune-in ammatory reaction persists, osteolysis will eventually occur, resulting in implant loosening and fracture. The long-term wear of arti cial joints and the particles produced by dissociation in the humoral environment will lead to the above phenomenon. In this study, the ion contents in the different extracting liquids were further measured. Except for Ni ions, almost no metal ions were detected in other groups. The amount of metal ions precipitated in TaNb40 and Ta group was less than 0.1mg/kg, and the Ti was less than 0.2mg/kg. The number of the released metal elements is closely related to the properties of surface passivation oxide lm and corrosion resistance. Many references [70][71][72] have proved that the passivation oxide lm on the surface of Ta is very stable, and no corrosion sign has been found in physiological solution in vitro. Two kinds of passive oxide lms, Ta2O5 and Nb2O5, exist on the surface of TaNb40, which has excellent corrosion resistance [72]. The corrosion resistance of Ni is very poor, the stability of surface lm NiO is low, and a large amount of Ni ion [73][74] can be released in a short time. Therefore, Ta and TaNb40 have excellent corrosion resistance and biocompatibility. At least there is almost no metal ion release in vitro. The poor biocompatibility of Ni is related to the instability of surface passive lm and the release of Ni.

Conclusions
TaNb40 implant was prepared by SLS technology of 3D printing in our previous studies, its surface is bright gray, showing a certain degree of roughness. (1) The TS, YS, Elongation, and Vickers hardness was 548 ± 50MPa, 420 ± 30MPa, 40%, and 425HV respectively, which indicate that the TaNb40 implant has good mechanical properties, wear-resistance, and not easy to brittle fracture. (2) TaNb40 and Ta were tested in vitro by direct contact, AO uorescence staining, CCK-8, and ow cytometry, it was found that the two materials did not inhibit cell proliferation and cell cycle arrest, the cytotoxicity grade in vitro was 1.
There was almost no metal ion release in the culture medium, and both of them have good early cell adhesion ability. All above these, it meets the requirements of medical biomaterials.  Table 6, and the totality weight ratio is less than 1.5%, the particle size of Ta and Nb powder is 10μm-60μm, and the SLS laser power is 250W. To facilitate the co-culture of the cells with materials in the same culture dish, the Ta-Nb alloy sheet, or TaNb40 for short (9 mm in diameter and 1 mm in thickness) was prepared by SLS. At the same time, Ta sheet with the same size was prepared by traditional casting method as another experimental group, and Ti and Ni sheet (Yudingda Metal, China) was used as positive and negative control materials respectively. The materials were sterilized by high temperature and high pressure. (set up by the author's team) were established as experimental cells. The third generation of hOMF was taken out from liquid nitrogen and the cells after resuscitation were cultured in an incubator (ThermoFisher) containing 5% CO2 and 37 ℃ / 95% humidity until it grows to 80% -90%, then going down to the future generation.
The cell growth observed by the direct contact method The experiment was divided into ve groups: TaNb40, Ta, Ti, Ni, and blank control group (only cells without material). It was proper in the density of 3×10 4 cell/ml, each group of materials setting up four parallel groups was placed in a 24-pore plate (Excell Bio, China) to co-culture with the cells. On the 1 st , 3 rd , and 5 th day, the cell growth around the material was observed by an inverted microscope(Leica). According to the national standard GB / T 16886.5, the cytotoxicity in vitro score was recorded taking 100% of the number of cells in the blank control group as the benchmark. Such as the following: 0. No cytotoxicity, no effect on cell growth; 1. Slight cytotoxicity, the number of round cells ≤ 20%; 2. 20% < the number of round cells ≤ 50%, with slight cell growth inhibition; 3. 50% < the number of rounds or dissolved cells ≤ 70%; 4. Severe cytotoxicity, resulting in serious cell damage.
The early cell adhesion observed by AO uorescence staining The experiment was divided into four groups: TaNb40, Ta, Ti, and Ni group. Acridine Orange (AO) uorescence staining (BestBio, China) was used to observe the early cell adhesion on the surface of the material (within 12 hours). Collecting the sample cells, counting, the cell density was about 2 × 10 5 /ml, and the 50ul cell suspension was added to the different material surfaces. After 30 minutes of cell sedimentation, adding 0.5ml culture medium again. The 50 ul of dye solution was added to the surface after the 1 st day and 4 th day respectively and incubated for 10-15 minutes in dark at 4 ℃. The maximum excitation wavelength was 488 nm.
Using Image J software, ve visual elds were randomly selected for each sample under the microscope, nally set up the database, perform statistical analysis.

Prepare the extracting liquid
According to the national standard ISO 10993-5:2009, the material extracting liquid was prepared according to the ratio of the extraction medium and the surface area (or weight) of the biomaterial. In this experiment, the sample diameter is 9mm and the thickness is 1mm, the extraction ratio is 3 / cm 2 · ml -1 based on Table 7. It is calculated that 0.53ml culture medium should be added to the 24-well plate containing materials, and placed in the refrigerator(ThermoFisher) at 4 ℃ as the extracting liquid.
The cell proliferation determined by CCK-8 The experiment was divided into ve groups: the extracting liquid of TaNb40, Ta, Ti, Ni, and the blank control group. The cell density was about 3×10 4 cells/ml. Each group of materials was put into a 96-well plate (Excell Bio, China), dropping 100ul cell suspension, after 24 hours, discarding the old culture medium, and adding 100μl per experiment group. After the 1 st , 3 rd , and 5 th day, add 10ul CCK-8 reagent (BestBio, China).
The OD value of each pore was measured by enzyme-labeled instrument (Thermo Scienti c Multiskan FC) at 450 nm. The OD value of the blank control group was taken as the standard, and the relative growth rate (RGR) was calculated by the ratio of OD mean value of experiment groups to that of the blank control group. RGR (%) = ODsample / ODblank × 100%. The cytotoxicity evaluation grading is shown in Table 8.

Cell cycle determined by ow cytometry
The experimental groups were the extracting liquid of TaNb40 and Ta, the control groups were the extracting liquid of Ti, Ni, and the blank control group(the complete medium). The number of cells in each group was about (5-10) × 10 6

Consent for publication
Not applicable.

Competing interests
The author(s) declared no potential con icts of interest concerning the research, authorship, and/or publication of this article.