Surface functionalization of anodized tantalum with Mn3O4 nanoparticles for effective corrosion protection in simulated inflammatory condition

The study highlights the corrosion behavior of untreated and treated tantalum with addition of trimanganese tetraoxide (Mn 3 O 4 ) nanoparticles in simulated inflammatory media. The anodic layer was produced on pure tantalum by anodization in electrolytes composed of ammonium fluoride, ethylene glycol, and water. Nano-particles were deposited uniformly on the surface of the anodized tantalum with the electrophoretic deposition (EPD) method. The results revealed that the anodic/EPD coating possessed more compact microstructure and higher bond strength than the anodic coating. Simulated inflammatory medium was based on phosphate-buffered saline with additions of H 2 O 2 and HCl. Potentiodynamic polarization and electrochemical impedance spectroscopy studies showed that the anodic and Mn 3 O 4 layers protected the tantalum from corroding in an acidic inflammatory condition. Finally, the corrosion protection mechanism of Mn 3 O 4 NPs in


Introduction
Tantalum and its alloys are widely used in orthopedics and dentistry fields due to their good biocompatibility and corrosion resistance, proper elastic modulus and high damping capacities [1]. The corrosion resistance of tantalum is secured by its ability to form a spontaneous passive oxide film on the surface, protecting it against the corrosive biological medium [2,3]. This passive film is usually only a few nanometers thick, so it may not be sufficient for protection in an acidic medium [2]. During inflammatory conditions after implantation cells release highly reactive oxygen species (ROS), such as hydrogen peroxide (H 2 O 2 ), hydroxyl radical (•OH), superoxide (O 2 − ), and hypochlorous acid (HOCl), as part of their response to infection, foreign bodies, and other inflammation-inducing conditions [4]. In parallel, osteoclast involved in bone remodeling express HCl for bone resorption which significantly drops local pH values together with the ROS release (from 7.4 to 5.5 and even down to 2.0-3.0), create an aggressive environment for the implant materials. Prior studies have demonstrated that simultaneous exposure to H 2 O 2 and HCl can accelerate the degradation and corrosion rate of titanium [5], stainless steel [6], magnesium [7], and cobalt-chrome alloy [8]. However, the features of the corrosion behavior of tantalum in such inflammatory conditions still remain undiscovered. It is expected that by providing a proper surface treatment to tantalum its corrosion properties could be significantly enhanced. This could be done by controlled anodization and introduction of nanoparticles working against the ROS release.
Metal oxide nanoparticles have been used in a variety of medical applications, and in particular it was reported that trimanganese tetraoxide nanoparticles (Mn 3 O 4 NPs) may possess excellent ROS-scavenging activities [9,10], eliminating H 2 O 2 , •OH, and O 2 − up to nearly 75%.
Moreover, the antibacterial activity and biocompatibility of the Mn 3 O 4 NPs were reported by researchers in these years [11,12]. To date, several studies were carried out to incorporate manganese into the coatings matrix on the metallic implants [13], however, without sufficient evidences of the existence of the manganese oxide over the coating. Recently Zhu et al. [14] have demonstrated that electrophoretic deposition (EPD) technology can be applied to deposit Mn 3 O 4 micro-particles on Ni-coated steel for high-temperature applications.
EPD has also already been utilized for producing bioceramics coatings on the surface of metallic biomaterials [15,16]. EPD is based on an electrophoretic movement of charged particles in a suspension (containing an electrolyte, solvent, particles, additives, and dispersant) towards the opposite polarity electrode under an electrical field's influence [17]. This technique presents significant advantages in synthesizing thin films from charged colloidal particles due to its short processing time, high deposition rate, environmentally friendly processing, and easy control of the thickness and uniformity of deposited particles through simple producing apparatus [15][16][17]. Usual drawback of the EPD for ceramic coatings is the insufficient bonding strength between coating and substrate for the requirements of clinical applications. To achieve high adhesion performance and corrosion resistance, pretreatment operation is required before the EPD process success on metallic substrates [18].
The anodizing technique is a cost-effective and efficient approach to increase the performance of metallic materials [19]. The primary aim of the anodizing is to generate a homogeneous, thick porous layer on the surface of a substrate [20]. It has been reported that an anodic oxide layer can act as an intermediate layer to reduce not only the corrosion rate of substrate but also to enhance the adhesion of the EPD coating [21,22]. For example, Ashok raja et al. [21] have reported the development of rGO-incorporated bioglass with 2 μm thickness via EPD on the titanium dioxide (TiO 2 ) anodic layer of 6 μm thickness on titanium substrate.
Most literature has studied tantalum anodization using electrolytes composed of hydrofluoric (HF) and sulfuric (H 2 SO 4 ) acids. Such electrolytes are very harmful to the environment and require specific safety measures for volume production [23]. There are very few studies focusing on tantalum anodization in HF-free electrolytes [24]. Fialho et al. [25] developed a pentavalent tantalum oxide (Ta 2 O 5 ) layer on tantalum by anodizing in two different voltages and processing time using HF-free electrolytes. They studied the effect of the mentioned parameters on morphologies, structure, and chemical composition of the produced oxide layers, but did not address the effect of the anodization process on the corrosion behavior of tantalum.
The main aims of the current study are to design and make an anodic layer on tantalum in HF-free media as an intermediate layer for the EPD of Mn 3 O 4 NPs on the tantalum surface and to discover the effects of these layers on the corrosion behavior of tantalum under normal physiological and simulated inflammatory conditions containing H 2 O 2 and HCl.
The anodization process was performed using a two-electrode configuration with a direct-current (DC) power supply (Zhaoxin, KXN-6020D), comprising the tantalum sheet as the anode and the stainless steel as the cathode. The distance between the electrodes was kept constant at 1.5 cm, and the electrolyte was magnetically stirred. The anodic layer was achieved at 40 V for 20 min in an aqueous electrolyte containing ethylene glycol anhydrous (EG; 99.8%, Sigma-Aldrich), 3 vol % deionized (DI) water, and 1.2 g L −1 ammonium fluoride (NH 4 F; 98%, Sigma-Aldrich) [25]. The temperature of the anodization electrolyte was kept below 40 • C by using a cooling water circulation. The anodized sample was then ultrasonically cleaned with ethanol and subsequently dried by airflow.
For the preparation of the EPD suspension, 10 g L −1 of Mn 3 O 4 NPs (99.95%, average particle size 30 nm, US-Nano) and 1.2 g L −1 iodine (I 2 ; 99.8%, Merck) as a dispersant were added into the 125 mL ethanol and ultrasonically were mixed for 30 min. It was reported that Mn 3 O 4 NPs are positively charged in the prepared suspension [14], so the anodized-tantalum should be connected as cathode and two stainless steel sheets electrodes as anodes. The distance between the parallel anodes and the cathode was set to be about 1 cm. The EPD process was carried out at 1 mA cm −2 using a DC voltage source at 45 • C [14]. The deposition process was conducted for 1 min. The suspension was magnetically stirred during the deposition. After deposition, the samples were dried in air at room temperature for 24 h. A schematic illustration of layers preparation by anodization and EPD techniques is shown in Fig. 1.

Materials characterization
The surface morphology, cross-sectional microstructure, and elements of the coatings were observed by field emission scanning electron microscope (FESEM; TESCAN, MIRA III) coupled with energy dispersive spectroscopy (EDS; Oxford INCA-energy 300). The transmission electron microscopy (TEM; Philips, EM208S) of Mn 3 O 4 NPs was carried out at an acceleration voltage of 100 kV. To examine the surface profile on the layers, three-dimensional (3D) pictures were acquired with microscopy of atomic force (AFM; Ara Research Co). The layers phase analysis was analyzed using X-ray diffraction (XRD; Siemens D500) with copper (Cu) Kα radiation source (λ = 1.5417 Å) and a monochromator at 40 kV, 30 mA in the 2θ range of 20-80 • and 0.03 • step per second. Fourier transform infrared spectroscopy (FT-IR; PerkinElmer, Spectrum 100) in the range of 4000-400 cm −1 with the resolution of 2 cm −1 was used to investigate the chemical compositions of surfaces. The bonding strength of the anodized-and EPD-coated samples was measured according to ASTM C-633 standard using universal testing machine (Instron 5569) with the 1 mm min −1 cross-head displacement rate and a 10 kN load cell [26].
Potentiodynamic polarization and electrochemical impedance spectroscopy (EIS) tests were performed using potentiostat/galvanostat (EG&G, PARSTAT 2263A). All electrochemical measurements were conducted in a typical three-electrode cell of 500 mL at 37 • C with the treated-and untreated tantalum as the working electrode, graphite as the counter electrode, and a saturated calomel electrode (SCE) as the reference electrode [5][6][7][8]. Phosphate-buffered saline (PBS; pH = 7.4, Sigma-Aldrich) was used to simulate normal physiological conditions; and a 150 mM H 2 O 2 (30%, Merck) solution in PBS titrated to pH 5.2 with hydrochloric acid (HCl; 37%, Merck) was considered to simulate inflammatory conditions around the metallic implants [5][6][7]. Before electrochemical measurements, the samples were soaked in the normal and inflammatory solutions for 24 h to stabilize the soaked sample's open-circuit potential (OCP). The samples to be examined in solutions were soaked to a surface of 1 cm 2 . Polarization studies were carried out at a sweep rate of 1 mV s −1 and were analyzed with CorrView (version 2.8) software (Scribner Associates). With a 10 mV sinusoidal potential in amplitude around E ocp , the EIS plots were obtained across a frequency range of 0.01 Hz-100 kHz. The EIS plots were examined and fitted by the ZView (version 3.1) software (Scribner Associates). The thermodynamic calculations were performed using HSC Chemistry software (Chemistry Software Ltd).

Results and discussion
Fig. 2,a shows the FESEM images of the untreated tantalum surface after polishing and etching, used as the reference sample during the electrochemical measurements. Fig. 2b and 2,c depict the micrographs of anodized-tantalum in different magnifications. These exhibited a compact anodic layer with randomly distributed small diameter pores. No visible delaminations and crack was observed on the surface. NPs EPD layer on anodized-tantalum, shown in two magnifications in Fig. 2e and 2,f, indicate that the layer was uniform and homogenous. The positively charged Mn 3 O 4 NPs continuously migrate toward the anodized tantalum substrate under the application of an electrical field, as shown schematically in Fig. 1 [17]. The porous structure increases the contact area between NPs and the anodized layer during the EPD. As it can also be seen in Fig. 2,f, the Mn 3 O 4 NPs are more or less evenly deposited on the surface of the anodic layer. The presence of some particles with a larger size in the coating is due to higher agglomeration behavior of NPs [27,28].
The AFM surface topography of anodized-and EPD coated-tantalum is shown in Fig. 2g and 2,h, respectively. The root mean square (RMS) values of anodic and anodic/EPD coatings were 1.24 μm and 0.58 μm, respectively. The measured topography parameters indicate that the anodic layer has become rougher compared to the EPD coating. It has been reported that the rougher surface of the interlayer is beneficial for EPD coating adhesion to the metallic substrate [18,21,22].
The XRD patterns of anodized and EPD-coated samples are depicted in Fig. 3,a. The Ta peaks in both patterns are attributed to the contribution from the substrate. The diffraction peaks revealed Ta 2 O 5 (JCPDS card No. 25-0922) as the main components of the anodic layer [29]. The following mechanism can be suggested for the formation of an anodic layer on the tantalum, analogous to data [30,31] [33]. In the EPD-coated sample's spectrum, the two peaks at 512 and 625 cm −1 are attributed to Mn-O stretching modes of tetrahedral and octahedral sites, respectively [34]. The weak absorption band at 1675 cm −1 ascribed to the bending vibration of O-H groups joined with Mn 3 O 4 NPs during EPD in an ethanol-based electrolyte [35]. A band relating to the vibration of adsorbed water was observed in 3400 cm −1 (stretching mode mode) [33,34]. Fig. 4a and 4,b show the FESEM micrographs of the cross-sectional view of the anodic layer. It has a uniform thickness, and it appears continuous, smooth, and well-bonded layer-substrate interfaces. At the lowest magnification in Fig. 4,b it is seen that each roughly cylindrical pore extending from the interface is connected to the tantalum substrate by a thin barrier layer of around ⁓50 nm thickness [36]. The FESEM analysis of the cross-section of the EPD layer (Fig. 4,c) indicated that Mn 3 O 4 NP coating is rather dense and adherent to the underlying anodized layer. The lowest magnification in Fig. 4,d shows the detail of some Mn 3 O 4 NPs included on the surface of coatings near the epoxy, which has been formed during mounting. As seen in Fig. 4a and 4,c, the anodic layer and EPD coating thickness were about 4.6 μm and 17.8 μm, respectively.
The elemental maps for EPD coating's cross-section in high magnification are shown in Fig. 4,e. The presence of Ta, O, and F elements from the porous Ta 2 O 5 layer underneath the Mn 3 O 4 coating is also confirmed from the EDS mapping. The existence of F and C is attributed to the presence of NH 4 F and EG in the anodization electrolyte composition [25]. Moreover, the O and Mn elements mappings suggest a well distribution of the Mn 3 O 4 NPs on the anodized layer.
The bonding strength of coatings is a crucial factor that directly affects the coated implants' service life [37]. The bonding strength of the anodic layer was 13.1 MPa. However, the bonding strength of the anodized/EPD coated sample was 23.4 MPa which meets the orthopedic implant industry's necessary standards (15 MPa) according to the standard ISO 13779-2 [38]. By introducing the porous Ta 2 O 5 layer on the tantalum surface, the coating/substrate interface bonding improves. Fig. 5a and 5,b show the polarization curves of the as-polished and surface-treated tantalum in normal and inflammatory media solutions, respectively. Anodic branches reveal that the as-polished tantalum exhibited strong passivation behavior in both conditions [39,40]. Regardless of the imposed conditions, the anodization and EPD treatments of the tantalum improved its corrosion resistance, as can be observed from the shift of potentials towards the anodic side, compared to the as-polished tantalum [41]. The corrosion current density (i corr ) can be obtained by Tafel curve extrapolation, as shown in Fig. 5,c [42]. When the as-polished tantalum is exposed to an inflammatory medium, the i corr shows an increase, indicating an increase in corrosion rate. These results show that the passive film is increasingly unstable in acidic conditions when H 2 O 2 is present in the environment. For coated samples immersed in inflammatory conditions, the i corr values were insignificantly increased by simultaneous presence of H 2 O 2 and HCl. On the contrary, compared with that of the as-polished sample, the i corr of the anodic and anodic/EPD coatings decreased, which demonstrated that the coated samples were thermodynamically stable against corrosion in both conditions [43,44]. It can also be seen from Fig. 5,c that the i corr of the anodic/EPD sample is the lowest among the three samples, which is one and two orders of magnitude lower than that of anodized and as-polished tantalum in inflammatory medium, respectively. It can be said that through the deposition of the Mn 3 O 4 NPs on the anodic layer, the increasing coating thickness and decreasing surface roughness eventually lead to the most protective composite coatings on tantalum [45].
The Nyquist plots related to the norrmal and inflammatory conditions are displaced in Fig. 5d and 5,e, respectively. The impedance modulus (|Z|) values at a lower frequency (0.01 Hz) are shown in Fig. 5, f. Generally, larger semicircles and higher |Z| 0.01Hz reflect a higher corrosion resistance [46,47]. It could be found that the distinct difference among the Nyquist plots was the size of the capacitive semicircles. Moreover, the Nyquist plots of anodic and anodic/EPD coatings exhibit two semicircles. The low-frequency semicircle traced by the anodic and anodic/EPD coated sample suggests the higher dielectric constant and considerable resistance to the corrosive ion diffusion than other samples. As presented in Fig. 5,f, the |Z| 0.01Hz value of all samples quickly dropped, indicating the anti-corrosion performance of the samples is greatly reduced during the inflammatory condition. In this condition, the anodic/EPD coating shows the largest |Z| 0.01Hz about 19.2 × 10 3 Ω cm 2 and the substrate is about 5.1 × 10 3 Ω cm 2 . This result suggests that anodic/EPD can improve efficiently the substrate corrosion resistance, which is consistent with the polarization results. Fig. 5,g depicts the equivalent circuit for as-polished tantalum in both conditions. It made of a charge transfer resistance (R ct ) in parallel with a double layer capacitance (Q dl ). To simulate the structure of the anodic and anodic/EPD coatings, a two-constant type equivalent circuit was considered (Fig. 5,h). Compared with the circuit in Fig. 5,g, it contains anodic and Mn 3 O 4 coatings' resistance in parallel with a constant phase element (R c /Q c elements) for the low-frequency loop that reveals a large capacitive behavior observed. In both circuits, R s denotes the electrolyte resistance between the working and reference electrodes. The fitted values of the circuit elements are summarized in Table 1. At angular frequency ω (rad⋅s −1 ), the impedance (Z ω ) of a constant phase element is defined by Refs. [44,48]: where Q 0 is (|Z|) −1 at frequency 1 rad s −1 and i = √-1. The value of n varies between 0 and 1, representing the amount to which the element deviates from its ideal capacitive nature. When H 2 O 2 and HCl are added to the solution, R ct decreases from 5.9 × 10 3 to 2.8 × 10 3 Ω cm 2 for aspolished tantalum, evidencing a higher electrochemical activity in inflammatory medium. A lower R ct implied more efficient charge transfer across the substrate/solution interface, which could decrease the possibility of charge recombination and thus diminish the corrosion resistance [35,41]. In Table 1, in both conditions, the R c values of anodic/EPD coating are higher than anodic coating, indicating that duplex coating efficiently protects the tantalum substrate. Therefore, it can be said that the deposition of Mn 3 O 4 NPs has increased the barrier performance of the anodic coating [49,50]. Furthermore, its R ct values increase noticeably, indicating the improvement of corrosion resistance of composite coatings. The anodic/EPD coating has lower Q c and Q dl values due to fewer pores and defects in the coating' structure, which trap reaction products inside the pores [44]. Moreover, the n c values for anodic/EPD coating are higher than n c for the anodic layer. By deposition of Mn 3 O 4 NPs, n values are typically increased, resulting in a smoother surface with decreased roughness, which matches the AFM study in Fig. 2 [7]. The Mn 3 O 4 coated sample had the highest corrosion resistance in both media. This can be due to the formation of a low-defect and diffusion-less layer.
The corrosion protection mechanism of the anodic/EPD coated tantalum in the inflammatory condition is shown schematically in Fig. 6. The anti-corrosion mechanism can be mainly influenced by the following factors: the physical barrier effect and corrosion inhibition of Mn 3 O 4 NPs based on catalytic activities. Due to the high surface area of NPs, the pores could be sealed by depositing Mn 3 O 4 NPs on top of the anodic coating, inhibiting the penetration of corrosive molecules and ions via pores, resulting in a significant level of protective property for the tantalum, as seen in the left part of Fig. 6 [7,50]. The corrosion behavior of the Mn 3 O 4 coating in acidic inflammatory medium can be also affected by its excellent catalytic ROS removal efficacy.
A variety of catalyst nanomaterials with ROS scavenging activities have been reported for the surface functionalization of biomedical implants. Recently Li et al. [51] demonstrated that catalase-mimetic  activities of cerium oxide (CeO 2 ) NPs, which deposited by magnetron sputtering on acid-treated titanium, can scavenge the ROS and create an anti-inflammatory microenvironment. It is also known that a range of ROS produced by bone tissue cells during inflammatory response can participate in Fenton reactions, resulting in large quantities of radical species [52]. In Fenton-like reactions, the metal ion provides an electron to H 2 O 2 and creates the •OH radical through following equation:  [53]. According to the Pourbaix diagram of manganese in water (Fig. 7), Mn 3 O 4 is unstable in acidic inflammatory condition, and in pH 5.2 dominant compounds are Mn 2+ and MnO 2 . The Fenton-like catalytic reaction mechanism involves the generation of •OH radicals caused by the circulation processes of Mn 3+ /Mn 2+ , resulting in decomposed H 2 O 2 , as described by Ref. [54]: Besides, the pH of the biological medium influences the catalytic activity of Mn 3 O 4 NPs, mostly dependent on whether the catalyst surface is H + /OH − adsorbed. H 2 O 2 prefers to interact with H + in an acidic environment and decomposes into H 2 O and •OH, according to theoretical calculations [55]: As stated before, besides the Mn 2+ compounds, the other dominant phase of the Mn 3 O 4 in acidic inflammatory solution is MnO 2 . MnO 2 is relatively eased to undergo reversible redox reaction between Mn 3+ and Mn 4+ state and produces the Mn 2 O 3 [56]. As seen in Fig. 7 [57]. It is also reported that the scavenging role of the manganese oxide phases is only efficient when the H 2 O 2 decomposition rate be equal to or higher than the production rate [10]. Therefore, this study developed a novel implant surface coating using Mn 3 O 4 NPs on tantalum surfaces that possessed excellent corrosion resistance and strong anti-inflammatory efficacy. The probable schematic diagram of the corrosion protection mechanism of Mn 3 O 4 NPs based on catalytic activity is drafted in the right part of Fig. 6.

Conclusion
In the present study, we successfully proposed the EPD technique to prepare a corrosion-resistant Mn 3 O 4 coating on tantalum when it is combined with an anodization process. Our results provide evidence that the corrosion rate of untreated and treated tantalum can be accelerated by inflammatory conditions. As the primary coating film on the substrate, the anodic layer can increase the corrosion resistance of tantalum in normal and inflammatory conditions. Combining anodic and EPD coatings makes it possible to develop compact, thicker, and well-adhered coatings on the tantalum surface. Electrochemical studies also revealed that the impedance modulus of the anodic/EPD coating was considerably higher than that of tantalum and anodic layer in normal and inflammatory conditions. We can propose that due to the sealing effect of Mn 3 O 4 NPs, they act as a physical barrier against corrosive agents accessing the anodic layer in acidic inflammatory conditions. Importantly, the Mn 3 O 4 coating protected tantalum from H 2 O 2 oxidant-mediated damage by maintaining catalytic activity and suppressing ROS generation in acidic pH. Taken together, the Mn 3 O 4 NPs could provide an opportunity to be utilized as a potential candidate for bone regeneration under ROS stress.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.  the anodic and EPD coatings was performed at Materials and Energy Research Center, Tehran, Iran. The authors are willing to express their gratitude for this support.