Bio-functionalization of a novel biocompatible high entropy alloy used for bone implants

In this paper, a novel biocompatible alloy defined as FeMoTaTiZr was obtained and functionalized by hydroxyapatite-based coatings (HAP) in order to increase their biocompatibility, bioactivity, and resistance to corrosion for to be used as bone implants. To obtain the surface with antibacterial properties, the HAP coatings were doped with small amount of Zn. The alloy was prepared using the VAR (Vacuum Arc Remelting) equipment, while the coatings by RF magnetron sputtering method. The EDS analysis confirmed the presence of Ca and P in the case of all developed coatings, having Ca/P or Ca/(P+Zn) ratio of about 1.70 and 1.66, respectively. The XRD and ATR-FTIR investigations confirmed the presence of calcium phosphate phases. The roughness of uncoated substrates increased after coating with HAP, and it was considerably increased by the Zn addition. The electrochemical tests showed that the un-doped HAP exhibited good corrosion behavior, while Zn doped HAP coatings have a high dissolution rate in fetal bovine serum, being more proper as a biodegradable material.


Introduction
High entropy alloys (HEA) proposed by Yeh and Miracle [1,2] are a new class of metallic materials characterized by chemical composition that contains at least 5 different elements, in equiatomic percent. In order to be appropriate for biomedical applications, the high entropy alloys must contain just elements that are considered biocompatible, such as: Ti, Zr, Ta, Mg, Ca, Zn, Si etc. Other elements with low toxicity such as Mo and Fe, are currently used for to obtain biocompatible alloys [3][4][5][6].
Currently, there is a wide variety of high entropy alloys that can be used in different fields, due to their special and specific properties, including industrial, aeronautical, military, medical applications and so on [7][8][9]. The high entropy alloys from the FeMoTaTiZr alloying system can be used for the manufacture of medical devices, human prosthesis, and implants. Iron plays a central role in the hemoglobin molecules formation and in the transport of oxygen. Human tissues contain 50 ppm of bound iron, especially hemoglobin, myoglobin, and various cytochrome enzyme systems. Iron is toxic only for high level contents. It is released through oxidation processes and it does not accumulate in tissues being metabolized [10]. Molybdenum is an element with a low toxicity and it is essential for the human body. It is added in metallic alloys to grains refinement, to improve the mechanical and chemical properties of solid solutions [11][12][13]. Metallic tantalum is a promising biomaterial whose applications have been limited for reasons of technical performance and not biological ones. Recent cell culture studies have related the good osteoblastic behavior of commercial pure tantalum. It can be used for a wide variety of clinical applications such as: joint replacement, reconstruction after tumor resection, necrotic treatment of the femoral head and spine [14][15][16]. Titanium is a non-toxic element for the human body, even in larger quantities. Titanium is used in medical applications for multiple specific fields, such as: orthopedics, cardiovascular surgery, ophthalmology, dentistry, urology, cosmetic surgery, neurology, suture material for wound healing, controlled drug delivery systems. The excellent biological behavior of titanium, its remarkable corrosion resistance and elasticity similar to human bone justify its increased use as a biomaterial and as a component of medical and dental devices [17,18]. Zirconium and zirconium alloys are used for medical applications, such as devices for the total replacement of the knee and hip, dentures, due to its excellent resistance to corrosion and wear provided by its dense and superficial adherent oxide layers [19,20].
Metallic surfaces used in clinical practice for preparation of orthopaedic implants (hip, knees, shoulder, etc.), dental implants (orthodontic braces) or osteosynthesis devices (nails, screws, plates, rods etc.) or other medical tools used for long or short implantation (e.g. stents, pacemaker). One way or another, the metallic implants were enormously used in medical applications for many years, nonetheless these devices still have some disadvantages in clinical practice and they should be removed and replaced by a new implant. This process is stressful for patients due to the physical and emotional trauma. For example, in the case of dental and orthopaedic implants, the main disadvantage is related to the low bioactivity and osseoconductivity abilities [21,22]. The most common metallic materials used in biomedical applications are Ti based alloy (Ti6Al4V), stainless steel (316L), CoCr alloys [23]. It is worth mentioning that the challenge is to find proper material, which should combine various qualities such as: hardness, toughness, strength, biocompatibility, and wear or corrosion resistance, as well as good antibacterial capabilities [21,22,[24][25][26][27][28][29]]. These requirements are difficult to find in one material. Thus, the research was turned to two directions: (a) to find an alloy with proper mechanical, strength, corrosion, and wear and fatigue resistant as well as nontoxic; and (b) to improve the alloy surface by coating in order to obtain the proper osseointegration and bioactive surfaces. Regarding the first point, there are a huge number of proposed alloys, most of these based on Ti [24,25,[27][28][29]. However, the research community proposed many medical implants made of novel obtained alloys, but none of these are approved by FDA yet (United States Food and Drug Administration), mainly caused by the concerns linked to biocompatibility [24,30].
The goal of the present paper is to obtain novel biocompatible alloy with functionalized surfaces used for bone implants. For this purpose, FeMoTaTiZr high entropy alloy was obtained by VAR method, and then the functionalization of their surfaces, by depositing of two types of coatings based on hydroxyapatite (HAP), in order to increase their biocompatibility, bioactivity and resistance to corrosion. In order to also have the surface with antibacterial properties, the HAP coatings were doped with a small amount of Zn. The content of Zn was selected based on the results reported by other researchers [31][32][33][34], who proved that a small amount of Zn is enough to assure the antibacterial properties.

Alloy preparation
The recipe for the alloy in the FeMoTaTiZr system was designed by choosing the chemical elements that have the lowest possible bio-toxicity. Currently, some of these elements are used as an alloying base for the manufacture of medical devices. The new high entropy alloys based on FeMoTaTiZr system were obtained using the MRF ABJ 900 vacuum arc remelting equipment. The technological calculation for obtaining the alloy was performed for a weight of about 40 g/load, using raw materials with high purity (99.5%).
The first step of the technological process was to obtain a pressure of 5x10 -3 mbar in the vacuum chamber, followed by purging with argon (Ar 5.3 quality), 3 times, to reduce the oxygen concentration up to 60 ppm. Then, the smelting of the metallic materials was performed under an argon atmosphere (pressure of 1.2 bars), using an electric arc, with an intensity current of 450 A. The melting was performed on both sides by the successive rotation of the mini-ingots, 8 times on each part, for a better homogenization of the chemical composition, due to the high melting temperature of the chemical elements used, as follows: Fe -1538 o C; Mo -2623°C; Ta -3017 o C; Ti -1668°C; Zr -1855°C. Figure 1 shows the mini-ingots of high entropy alloy and the samples prepared for metallurgical investigations and depositions (sanding with abrasive papers -Grit 320 -1000 and polishing with abrasive alpha alumina powders in suspension). EDX chemical analysis showed the following chemical compositions for FeMoTaTiZr alloy (masic %): Fe = 15.46%; Mo = 20.45%; Ta = 32.45%; Ti = 12.67%; Zr = 18.97%.

Coating's preparation
The coatings were carried out using a RF magnetron sputtering technique with a system from Aja International Co (SUA). For the preparation of HAP, one cathode was used, made of HAP with 99.98% purity. For Zn based HAP coatings, two cathodes made of HAP and ZnO (99.98% purity) were used. Because we intended to achieve coatings with slight quantities of Zn (considered as dopant), we used the cathode made of oxide (ZnO), selected due to the comparable deposition rate with that of HAP.
Prior to the deposition, substrates were pretreated as described in Refs. [22,35].
Deposition conditions of the coatings were listed in Table 1. These parameters were judiciously selected based on our previous results published in Refs. [35][36][37][38][39][40][41][42]. We also intended to obtain coatings with a thickness around 300 nm.  (Bruker, Germany), from 10° to 100°, with a 1D LynxEye detector and a scan step size of 1s per step and 0.040° step size. After the indexing, the diffraction data was subjected to Rietveld refinement using TOPAS program. Phase composition of coatings was evidenced by grazing-incidence XRD (X-ray diffraction) using a CuKα radiation (SmartLab, Rigaku) from 20° to 80° with a step size of 0.02°/min and an incident angle of 2°.
The IR spectra of coatings were obtained using a FT-IR spectrometer (Jasco 6300), equipped with an ATR unit (Pike Technologies). A Si/ZnSe crystal was mounted in order to collect spectra in the 4000-550 cm -1 range at a spectral resolution of 4 cm -1 .
Before collecting any data, the ATR system was purged with a constant flow of nitrogen gas while the background and sample spectra were acquired using 150 average scans. The samples were placed directly on the surface of the Si/ZnSe crystal and a high-pressure clamp was further used to ensure a good spectral quality. After every single IR measurement, the crystal's surface was cleaned with isopropyl alcohol and the background spectrum was collected again.
The hardness and elastic modulus of the coatings were analysed using a Hysitron Premier TI nanoindentation system, equipped with a Berkovich indenter tip (100 nm radius). The Z-axis calibration was executed using a standard calibration sample The electrochemical measurements were performed with a VoltaLab 40 potentiostat/galvanostat, using a standard electrochemical cell, with 3 electrodes (reference electrode: Ag/AgCl; counter electrode: platinum) and a thermostat jacket.
Because the samples had irregular shapes, they were trapped in system with a "platinum wire" and all surfaces, except the working one, were insulated with varnish.
The electrochemical determinations were performed in fetal bovine serum to reproduce in vivo study conditions. All measurements were performed at room

Microstructures and composition of uncoated high entry alloy
The SEM image of the FeMoTaTiZr alloy shown in Figure 2 reveals its structural inhomogeneity and it captures the demarcation area between the two regions: one, with a dendritic structure and the second region, with a fine textured structure, in which, even if there are still dendrites, they are of significantly reduced dimensions.
The inhomogeneous structures in terms of composition, as well as the appearance of microsegregations are phenomena that can occur in case of melting and solidification of high entropy alloys, due to the high content of main alloy components.
The HEAs require to be several times re-melted in order to achieve ingots with chemical uniformity. Another difficulty consists in the differences of melting temperatures for the constituent elements that increase with the growing of the number of the alloy components. In order to avoid these negative aspects, for future experimental work, the manufacturing of the FeMoTaTiZr alloy will be performed starting from binary intermediate alloys, prepared by selection and association of the appropriate elements (considering the melting point criterion) [45].
The compositional chemical mapping of the analyzed area allows the understanding of the association of elements in the different investigated regions.
Correlating these images with the EDS analysis (Figures 3 and 4   For the same region of the FeMoTaTiZr alloy, a quantitative analysis of the component elements was performed ( Figures. 5a and 5b). The crystalline structure of FeMoTaTiZr alloy sample was analyzed through X-ray diffraction ( Figure 6), the crystalline phases identification from the XRD pattern being done using the specific data base. The recorded peaks ( Figure 6) reveal the presence of the following constituents: MoTa; Mo, (Ti7Zr3)0.2; Ti2Fe and (FeMo)Zr (for this last phase, taking into account the elemental maps from Figures 4c -4e, the iron is associated in alloy matrix with zirconium and not with molybdenum). The peaks are distinct, clear, their intensity allowing to appreciate the degree of crystallization of the constituents of studied alloy after casting. The most intense peak, at 2θ ~ 40°, belongs to the compound MoTa. Its presence confirms the results obtained by elemental mapping, regarding the composition of the dendrites in the case of the FeMoTaTiZr multicomponent alloy. The presence of the other two constituents, (Ti7Zr3)0.2 at angle of 2θ = 34° and Ti2Fe at angle of 2θ = 41.5°, brings further clarifications concerning the matrix, in which the dendrites are arranged, namely the fact that it contains two constituents. The presence of Mo is explained by the fact that the melt was not homogenized during elaboration, which allowed that at solidification, a part of Mo, left undissolved, to solidify again (see the peak from the angle 2θ = 40.5°).
After the indexing, the diffraction data was subjected to Rietveld refinement, Figure 7 showing the Rietveld refinement patterns for the FeMoTaTiZr alloy. The refined lattice parameters and Rietveld refinement results are listed in Table 2. The Rietveld refinement results indicated the presence, along with Mo, of two main crystalline cubic phases of Ti2Fe and MoTa. In addition to these main constituents, a hexagonal phase was identified in the alloy -(Fe Mo)Zr -and, in a very small proportion, the compound Ti7Zr3. This is consistent with the SEM/EDS analysis and elemental mapping results.

Morphology and elemental composition of the coatings
The EDS mapping of consisted elements in the coatings can be observed in Figure   8. The EDS analysis confirmed the presence of Ca and P in both coatings. The elements of substrates are also found, due to the small thickness of the coatings. The EDS Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 9 April 2021 doi:10.20944/preprints202104.0259.v1 mapping analysis was used to investigate elemental distribution (Ca, P, and Zn) and the results are showed in Figure 8.  Figure   8). No cracks or other defects were found on the surface. It is well known that the addition of Zn into HAP matrix induces distortions in its lattice, leading to the Ca-deficient/rich hydroxyapatite, which probably will increase/decrease its solubility and bioactivity [46]. The morphology of the coatings is displayed in Figure 9. As it is shown in Figure 9, no microcracks can be found on the coatings surfaces. The doped HAP coatings show a variety of agglomerated spherical particles with a wide size distribution (size range of 10 -30 nm), which are larger than of the HAP crystal found in native bone [47]. It was reported that the smaller particles lead to a higher bioactivity, more proper for biomedical applications [47].

HAP+Zn/MoTaTiZrNb
Ca/P = 1.95±0.07; Ca/(P+Zn)=1.66±0.01 The X-ray diffraction patterns of HAP coatings and Zn doped HAP coatings are presented in Figure 10.  The ATR spectra of HAP and HAP+Zn coatings are presented in Figure 11. As it can be noticed, all ATR spectra confirmed the presence of two characteristic functional groups such as phosphate and hydroxyl. In the case of HAP coating, two sharped and well-defined phosphate ν3 bands can be clearly observed at 1088 and 1025 cm -1 , while the phosphate ν4 bands can be distinguished starting from 618 cm -1 up to 574 cm -1 . The band assignments for asymmetric stretching mode of PO4 3-(ν3) and O-P-O bending (v4) are in good agreement with other reported ones [48,49]. By addition of Zn into the HAP structure, the bands corresponding to PO4 3-(ν3) became wider, but no peak wavenumber shift was revealed. This finding validates the assumption that the Zn 2+ ions were successfully introduced into the HAP structure despite the fact that several bands assigned to the functional group of PO4 3-(ν4) exhibited a slightly shift (<15 cm -1 ) towards higher wavenumbers. The broad bands ranged between 3600 cm -1 and 3500 cm -1 can be attributed to the O-H stretching. Figure 11. ATR-FTIR spectra of both HAP and HAP+Zn coatings deposited on FeMoTaTiZr alloy.

Hardness, elastic modulus, and roughness
Hardness and elastic modulus of the investigated samples are presented in Table   3, the values were determined according to the Oliver-Pharr method by forcedisplacement curves presented in Figure 12.  [1,5,8,9,18,32]. In Table 3, the H 3 /E 2 ratio is also calculated, which is related to the resistance to plastic deformation, a parameter sensitive to corrosion properties of the materials [50]. The roughness of the investigated surface before corrosion tests is presented in Table 3. The roughness is a parameter very important in clinical practice. For example, the roughness of endosseous dental implants should not exceed the 100 μm [51]. On the other hand, it was reported that the bone grows better on the rough implant sur- faces, as a result of the high interfaces between bone and implant [52][53][54][55]. According to

Electrochemical tests
In Figure 13, the variation of the open circuit potential (OCP) with time was presented. According to Ref. [57], OCP usually depends on chemical composition of the investigated material, but also on thermomechanical history, the type of electrolyte, temperature etc. Note that the most electropositive potential was recorded by the HAP coating, followed closely by the system, where Zn was present and then by the uncoated alloy. After only 10 min of immersion, the HAP+Zn coating formed a protective film, which showed similar values of potential as in the case of HAP. This result can be assigned due to high passivation rate of Ti, Zr and Ta elements [58].
Relatively stable curves were recorded for all investigated systems, showing that the passivating film formed at the metal-electrolyte interface was uniform and protective.
The variation of the polarization resistance with time is presented in Figure 14,  The variation of the corrosion rate with time is presented in Figure 15. This parameter is indicative of the degree, in which a material degrades in a certain medium. Therefore, similar to the previous results, the most resistant material appeared to be HAP. Even though in this case there was also a certain instability, at the end of the test, the next lowest value of the corrosion rate was observed in the case of HAP. Again, the FBS electrolyte also showed in this case a negative influence that led to an accelerated rate of corrosion for HAP+Zn materials. Considering the variation of the corrosion rate results, the dissolution of the investigated system at the end of the test followed the order: HAP coating < HAP+Zn coating. In Figure 16, the potentiodynamic polarization curves are presented and in Table   4 the kinetic corrosion parameters are shown. Unexpectedly, the corrosion potential presented the least negative value for HAP/FeMoTaTiZr (Ecorr= -464.5 mV), whereas when Zn was added, the potential dropped to Ecorr= -935.7 mV. Considering the anodic and the cathodic slopes, one can note the higher ba values, therefore a tendency of the investigated materials to passivate [61]. Based on the slopes of each system, the corrosion current density (icorr) is estimated. Lower values were presented by HAP coatings (icorr= 0.0640 µA/cm²). An increase up to about 3 µA/cm² was estimated for icorr in the case of Zn-containing coatings. Higher polarization resistance (Rp) and lowest corrosion rate values were ascribed to HAP coating, followed by Zn-containing coatings.   (Figure 18.). In Figure 18, a similar impedance modulus was observed for HAP coating in the low frequency region. Lower values of impedance modulus were recorded for Zn addition HAP depositions.
The use of EIS has the main advantage of analyzing the formed interfaces when a material is immersed in an electrolyte, using their frequency response to an applied potential and their corresponding resistive and capacitive behavior. In order to assess the corrosion kinetics present at the interface, the impedance data was used for fitting procedure with an equivalent circuit. In this case, Randles circuit was employed, taking into consideration a charge transfer resistance (Rp) and an ideal capacitor (Cdl), which describes the flow of the current at the material/ electrolyte interface [25, [62][63][64][65].
The previous mentioned parameters are used in parallel and a solution resistance was also added (Rs). The interface is usually characterized by a double-layer formed between the material and the electrolyte and it depends of each type of investigated system and the FBS electroactive species.
Taking into consideration the extracted electrochemical parameters ( to Rs, the most unstable material in FBS medium appeared to be HAP+Zn coating (55.530 Ω·cm²), whereas the most electropositive value of Eeq was shown by the same system without Zn addition (-84 mV). However, in a recent study [66], Zn proved to be efficient for in vitro degradation as well as for cytotoxicity assay in this medium (FBS). Moreover, Zn tuned BMs showed a great potential for biomedical applications.
In another study [67], Zn based biodegradable composites with 3 and 5 wt. % hydroxyapatite were investigated in simulated body fluid and were found to be biocompatible.
In Figure 19,the investigated surfaces are presented after electrochemical tests.

Conclusions
According to the gained results, the following conclusions can be drawn:

FeMoTaTiZr Alloy
• The high entropy alloys from the FeMoTaTiZr alloying system can be obtained using the VAR method, after performing multiple remelting, because the chemical elements have high melting temperatures. Homogeneity of the alloy has been proven by the concordance between the designed compositions and EDS chemical analysis performed on the mini-ingots. During the metallurgical process the metal losses are very small, due to the high vacuum level and protective argon atmosphere.
• The SEM analysis and the elemental mapping showed that the microstructure of the cast FeMoTaTiZr alloy consists in dendrites based on Mo and Ta, dispersed in matrix formed by Ti, Fe, and Zr. • The Rietveld refinement results indicated the presence, along with Mo, of two main crystalline cubic phases of Ti2Fe and MoTa.

Undoped and Doped Coatings
• The EDS analysis confirmed the presence of Ca and P in both coatings with Ca/(P+Zn) ratio of about 1.70 for undoped HAP and of about 1.66 for Zn doped HAP coatings.
• No cracks or other defects were found on the surface.

•
The XRD and ATR experiments revealed that the HAP phase was formed in all prepared coatings, either doped or undoped.
• The roughness of uncoated substrates increased after coating with HAP, and it was considerably increased by the Zn addition.
• After the Zn addition, an increase in hardness was observed for HAP coatings.
• High values of polarization resistance recorded for HAP coatings, indicating that HAP coatings acted as a more insulating barrier.
• The dissolution of the investigated system at the end of the electrochemical tests followed the order: HAP/FeMoTaTiZr alloy < (HAP+Zn)/FeMoTaTiZr alloy.

•
The lowest corrosion rate was observed in the case of HAP, indicating that the FBS electrolyte accelerated rate of corrosion for HAP+Zn materials.
The present study succeeded in the preparation of HAP, while the Zn addition leads to a high dissolution rate in FBS solution. Thus, it can be concluded that the Zn doped HAP coatings can be used as biodegradable materials in medical applications.