Electrochemical behavior of additively manufactured patterned titanium alloys under simulated normal, inflammatory, and severe inflammatory conditions

The electrochemical behavior of a biomaterial surface in local conditions is a signiﬁcant factor affecting the success of the implant placement. This is of a particular importance of metallic biomaterials which can undergo oxidation, corrosion and subsequent degradation. This study reports new data on the electrochemical behavior of additively manufactured (AM) patterned titanium alloys, analyzed after 1 and 12 h immersion in three different media mimicking normal, inﬂammatory and severe inﬂammatory conditions. Polarization study showed that corrosion resistance increases with increasing immersion time in all cases. It was found that in inﬂammatory condition a destructive effect on the passive layer ' s resistance was triggered by H 2 O 2 whereas in severe inﬂammatory condition, albumin, lactate, and H 2 O 2 all have a synergistic effect towards decreasing the corrosion resistance of patterned titanium layers. Electrochemical impedance data suggests that in the severe inﬂammatory condition the charged albumins are attracting to the localized pitting areas, changing diffusion transport of corrosive species at the interface of the metal/passive layer. The electrochemical tests also proven that laser-assisted patterned titanium alloys surfaces have an improved corrosion resistance in simulated solutions compared to untreated titanium of the same composition. It is suggested that new surface topography and wettability are also positive factors contributing to this improved corrosion performance in patterned specimens. ©


Introduction
Commercially pure titanium (CPeTi, Grade 2, UNS R50400) and Tie6Ale4V alloy (Grade 5 alloy, UNS R56400) are two among the most commonly used metallic biomaterials in medical implants devices [1,2].They are usually deployed for hard tissue (bone) replacement because of their low specific weight, high strength, good corrosion resistance, and proper biocompatibility [3,4].Since the nature of the implant surfaces influences the adhesion of cells to implanted biomaterials, the surface characteristics and the ability of adjusting them are important to ensure optimal bone-to-implant integration [5,6].For example, it has been demonstrated that macro-pores on Ti implants surface can improve wettability, tribological performance and biological behavior [7e9].It has also been reported that a porous surface with periodic patterns can promote early cell-to-surface interactions, including adhesion, proliferation, and metabolic activity in osteoblastic cells [10e12].The design of specific surface patterns may also positively contribute to successful bone ingrowth into an implant by enhancing the interlocking between the biomaterial and the host bone [13,14].
Additive manufacturing (AM) is an emerging technology in the fabrication of the macro-and micro-architectural framework of metallic biomaterials both externally and internally with high flexibility and sufficient precision [15e17], comprising also various porosities and surface topographies.One of popular AM techniques is powder bed fusion (PBF), particularly laser-driven PBF (L-PBF), which was currently demonstrated to be an effective AM method for geometrically complex metallic implants, including those with periodic patterns [18,19].There is still debate about the optimal pore size for osteointegration in porous structures of metallic biomaterials.Some sources suggest a range of 100 … 400 mm [20,21] or even 650 … 1400 mm [22], although there are other ranges too-cells were found even in the deep pores even as small as dozens of micrometers [21,23,24].Porous titanium implants manufactured using L-PBF with pore sizes of 1000 mm in an in vitro study were shown a significant increase in metabolic activity and cell density compared with implants with pore sizes of 500 mm [25].However, in another study [26] a significant improvement in cells fixation capability and in vivo bone ingrowth after 2 weeks were demonstrated for 600 mm pore size L-PBF-made porous titanium implant compared to structures with the average pore sizes of 300 and 900 mm.
Less attention was paid to electrochemical behavior of such implants in their reactions with body fluids, as this results in changes, sometimes drastic, over time [27].This might be an underexplored area for AM-made titanium implants because new techniques like L-PBF create local heat affected zones and can therefore change both microstructure and local electrochemical stability of the materials.Both surface modification of the material and the selection of the corrosion media [28] are very important to mimic specific in vivo conditions.Commonly used options (simulated body fluids, artificial saliva, etc.) cannot sometimes adequately represent the complex environment in loco the implantation site [29], as the implantation of orthopedic devices usually triggers a local inflammatory response in the body [30].This needs to be properly reflected in the media composition and reactivity analysis when measured in the laboratory.During an inflammatory response, the release of reactive oxygen species (ROS), lactic acid, hydroperoxyl radicals (HO 2 ), hypochlorous acid (HOCl) by leukocytes and HCl by osteoclasts causes the extracellular medium to become oxidative and acidic (pH z 4e5) [31].Subsequently, microorganisms, neutrophils, and macrophages can produce more acidic (pH z 2e3) and oxidative environment surrounding an implant associated with severe inflammation [32,33].Generally, Ti-based biomaterials are relatively corrosion resistant in normal physiological conditions due to the dense oxide layers protecting the surfaces [34].However, it was found that CP-Ti and Tie6Ale4V alloys indeed undergo intensive corrosion and visible surface roughening when exposed to media reflecting these inflammatory and severe inflammatory conditions [35], mimicked by the addition of H 2 O 2 and HCl to a phosphate-buffered saline (PBS) solution (which itself can be ranked as a normal medium).To simulate the severe inflammatory condition, H 2 O 2 , calcium L-lactate hydrate (CLH), bovine serum albumin (BSA), and more HCl were added to the normal biological solution [35].
In the literature, multiple surface patterning techniques on titanium alloys have been used to enhance implant performance and sometimes also the corrosion resistance [36e38].However, till now, there are very scarce information about regularly patterned surfaces of Ti alloys made with L-PBF method.Most of the corrosion studies performed on AM Ti implants have been conducted in normal conditions [39,40], lacking consistent reports on their performance in more complex environments corresponding to post-implantation.The main objective of this study was to perform a deeper study the electrochemical and corrosion behavior of L-PBFmade patterned CP-Ti and Tie6Ale4V substrates under the simulated normal, inflammatory, and severe inflammatory conditions, comparing the results with previous electrochemical study on untreated CP-Ti and Tie6Ale4V substrates [35].

2.
Materials and method

Preparation of patterned CP-Ti and Tie6Ale4V layers
The supplied CP-Ti (99.6% purity) and Tie6Ale4V alloy (Al 5.5 … 6.76%, V 3.5 … 4.5% wt., Ti -balance) plates by Goodfellow Cambridge Ltd. (Huntingdon, England) were used as substrates.The plates were cut into disks with 100 mm diameter for the L-PBF machine's working platform, and rinsed ultrasonically in acetone for 10 min in order to remove surface contaminants before L-PBF process.The micromorphology of the CP-Ti (99.3% Ti), and Tie6Ale4V (Al 5.4%, V 3.8%, Fe 0.2%, C 0.009% wt., balance Ti) powders (Merck, Germany) used for L-PBF are shown in Supplementary Fig. S1.
The computer-aided design (CAD) models of square patterns with the average lengths of ~600 mm were made with nTopology software (2020, nTopology, NY, USA).The patterned Ti layers were manufactured with optimized parameters (Supplementary Table ST1) at Btech Innovation Ltd., Istanbul, Tu ¨rkiye.A 3D selective L-PBF printer (Mysint100, Sisma S.p.A., Vicenza, Italy) with a fiber laser of wavelength 1060e1100 nm, power <200 W and spot diameter of 55 mm was used to overprint CP-Ti and Tie6Ale4V layers on flat CP-Ti and Tie6Ale4V substrates, respectively (Supplementary video SV1), therefore producing patterned layers of the same composition as the substrate.Eleven slices of the powder were applied to reach a layer thickness of ~300 mm for both samples.The process was conducted under high purity Ar atmosphere, protecting the powder from oxidation.After L-PBF, the built parts (Supplementary Fig. S2a) were cut off into small disks with a 15 mm diameter (Supplementary Fig. S2b) by wire j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 6 : 3 5 6 e3 7 0 electrical discharge machining, washed with water and acetone using ultrasound.

Surface characterization
The surface morphology and microstructure were observed under an optical microscopy (OM, Olympus GX71, Olympus Corp., Japan) and scanning electron microscope (SEM) with an energy dispersive spectroscopy (SEM-EDS, Hitachi TM-4000Plus-RAMI, Japan).3D profilometry in OM was made with a dedicated software (analysisDocu, Olympus Corp., Japan) with an automatic Z-stage movement and assembling of the resulted images into the package.
The phase structure of the alloys was tested using X-ray diffraction (XRD, Philips PW3040/60, The Netherlands) with a monochromatic Cu-Ka radiation source at 40 KV and 30 mA in the range 2q ¼ 20 … 80 , using a step size of 0.02 .The wettability was determined by measuring the contact angles of 5 mL phosphate buffered saline (PBS) solution's droplet via a Theta Flex optical tensiometer (Biolin Scientific, Finland) in the sessileedrop mode.Three contact angles on different regions were measured, and the mean values and standard deviations for patterned and flat Ti samples were calculated.All wettability measurements would be performed under the same conditions of temperature and humidity.

Electrochemical measurements
Electrochemical tests like open circuit potential (OCP), electrochemical impedance spectroscopy (EIS), and potentiody-namic polarization (PDP) were used to evaluate the corrosion behavior by the three-electrode configuration with a Ag/AgCl as a reference electrode (R.E.), a graphite rod (3 mm diameter) as the auxiliary electrode, and patterned Ti substrate as a working electrode (W.E.).The electrochemical measurement procedure comprised with extra details shown earlier [35].The simulated media for electrochemical studies with their respective compositions are summarized in Supplementary Table ST2.The post-corrosion morphologies were studied after the PDP test using SEM/EDS on surfaces immersed in simulated biological media for 12 h.

Results and discussion
3.1.

Microstructures and chemical composition
Fig. 1 shows two-dimensional (2D) and three-dimensional (3D) OM images of the patterned layers.Fig. 1a reveals that CP-Ti layer is patterned squarely with an average window size of about 610 mm in diameter.Similarly, the printed Tie6Ale4V pattern (Fig. 1c) revealed square columns with a side length of ~560 mm and a strut size of ~500 mm.It can be seen that the average window size in both patterned samples is between 500 … 600 mm, which might be considered close to the optimal values reported in other studies [23,24,41].There are some differences between the CP-Ti and Tie6Ale4V printed surfaces.In the patterned CP-Ti sample, the size of the strut was higher compared to sample Tie6Ale4V, and the height profile of the strut branch has become a stair-step.
Since the laser beam has a Gaussian distribution profile, the powder (partially melted by laser) occupies more space near well boundaries than what is predicted by ideal model made with CAD.The difference between the sizes of the struts in the two samples can be related to the size of the powders, as explained in Supplementary Fig. S3a.The laser beam provided in the L-PBF machine (55 mm) possesses the highest intensity near the center and decreasing outward (Supplementary Fig. S3b).CP-Ti powders used in this study has dimeter ~20 mm, so the central region that had a better beam intensity.The CP-Ti powder particles would melt within this 20 mm space and be better attached to the edges of the patterns, resulting in more particles being able adhere to the edges leading to a wider strut than the one obtained with Tie6Ale4V powder (~50 mm size).
SEM images of the top surface of the patterned CP-Ti and Tie6Ale4V layers are shown in Fig. 2aed in two magnifications.They show that the regular square patterns with a complete strut are distributed on the top surface of samples.The surface of the patterned samples has a large number of small spherical particles (Fig. 2c and d).These particles are very similar in size and shape to as-received CP-Ti and Tie6Ale4V powders (Supplementary Figs.S1 and S2).Attached metallic particles on the lattice strut surfaces are originating from the complete or partial melting of each layer of bent struts with different inclined angles, followed by adherence to the strut surface with diffusion [42,43].The EDS maps and spectra for corresponded areas of patterned CP-Ti and Tie6Ale4V (Supplementary Fig. S3a and Fig. S3b) show that the patterned Tie6Ale4V layers constitute of Ti, Al, and V and small amounts of oxygen.The printed layer composition was consistent with the original composition of Tie6Ale4V powders (Ti 90%, Al 5.4%, V 3.8%, Fe, O, C < 1% wt.), indicating that laser-powder interaction did not cause changes in alloying elements concentration during L-PBF.The EDS maps of patterned Tie6Ale4V are characterized by solid solution alloying elements without non-homogeneous precipitates.It  indicates that during the L-PBF process, the melted metallic particles have solidified under fast cooling rates without segregation.
Fig. 2e represents the XRD profiles of the surface of patterned CP-Ti and Tie6Ale4V layers on the same substrates.For both printed samples, the a-Ti peaks were identified using the reference cards JCPDS file #44e1294, and no oxide-related peaks were observed.The peak at 2q ¼ 39.5 in the graph of the printed Tie6Ale4V layers shows a combination of a-and bphase.Generally, due to the rapid cooling process used in the L-PBF production of Tie6Ale4V alloys, only a small amount of the b-phase remains and thus it is difficult to detect the bphase in Tie6Ale4V alloys using XRD patterns in the vast majority of L-PBF cases [44,45].

Surface wettability
The surface wettability of titanium implant materials significantly influences corrosion behavior, protein adsorption, and cell adhesion [46].Fig. 3 shows the results of an experiment evaluating the wettability of flat and patterned Ti samples based on the contact angle goniometry of a PBS solution's droplet.Surfaces that have a contact angle less than 90 are commonly classified as hydrophilic, otherwise, they are hydrophobic.All samples considered in this study were found to be hydrophilic according to this criterion: the contact angle values increased from 48.4 ± 3.1 for the flat CP-Ti surface to 83.8 ± 2.4 for the patterned sample, and from 40.9 ± 2.8 for the flat Tie6Ale4V surface to 75.5 ± 1.6 for its patterned sample.This is expected, as surface roughness influences the wettability behavior of surfaces according to the Wenzel and Cassie-Baxter models.Similar contact angle values for flat Tie6Ale4V to PBS solution have been reported by other studies [47].
In this study, contact angle values could have increased by coupled effects of increasing surface roughness and air pockets trapped beneath the PBS droplet which is well justified using principal CassieeBaxter's theory.Nevertheless, the patterned surfaces were still hydrophilic in nature.It has previously been reported that Ti-based implants manufactured using L-PBF process might have a hydrophilic surface [48,49].The noticeable variations in the contact angle values of patterned specimens could be related to a geometrical dissimilarity between the CP-Ti and Tie6Ale4V patterns, especially in struts morphology.The surface chemical composition also influences wettability.It has been reported that the Tie6Ale4V surface makes lower contact angle than CP-Ti due to the presence of aluminum and vanadium oxides on its surface [50,51].

Open circuit potential
Data on OCP variations of patterned CP-Ti and Tie6Ale4V layers after 1 and 12 h of immersion in simulated media are shown in Fig. 4. OCP values move toward electropositive potentials in the simulated inflammatory and severe inflammatory environment after implantation.A possible explanation for this is the decomposition of H 2 O 2 into H 2 and O 2 , thereby serving as an additional cathodic reaction in a sluggish oxygen reduction reaction [52].Earlier studies indicate that the positive OCP values in inflammatory solutions can be attributed to the adsorption of oxidizing intermediates, such as ,HO 2 and ,HO, onto the passive film on Ti-based biomaterials [53].In severe inflammatory conditions, adding BSA and CLH to the inflammatory solution (H 2 O 2 þPBS) decreases OCP values.Nevertheless, OCP values remain higher and more positive than normal conditions (PBS).
BSA presented in media adsorbs on the surface of Ti by chemisorption through carboxylate/amino functional groups or electrostatic interactions, inhibiting oxygen reduction reactions [54], resulting in decreasing surface potential.CLH adsorption on the surface of Ti implants also significantly blocks the cathodic site in H 2 O 2 -containing solutions, giving a lower OCP value [55].It can also be observed that under all conditions with increasing immersion time, OCP values increased.Due to suppressed anodic reactions anodic current density has decreased and passive oxide layers have chemically reconstructed or thickened [56].The patterned Tie6Ale4V layer shows a much higher potential than patterned CP-Ti layer under all conditions, which is likely to be due to the formation of a passive layer which may have also aluminum and vanadium oxides, being also denser than the pure oxide layers present on CP-Ti [1,2].H 2 O 2 may cause fluctuations in the OCP evolution of patterned Tie6Ale4V layer in an inflammatory environment due to the high resistance of the passive layer [52].Fig. 4 also revealed that the patterned specimens had different OCP values compared with flat counterpart specimens (OCP evolutions for flat CP-Ti and Tie6Ale4V were reported in our previous study [35].).In patterned specimens, OCP shows more of a shift towards electropositive values under normal and inflammatory conditions compared to flat Ti substrate, whereas OCP values do not change much in severe inflammatory conditions.

Potentiodynamic polarization
The values of E corr , b a , b c , I corr , and R p extracted from the plots are summarized and compared with previous study in Supplementary Table ST3.As seen in Fig. 5a and b, all patterned specimens display passivation behavior in anodic branches after 1 h and 12 h immersion under normal conditions.This behavior is consistent with the literature [58].However, the presence of H 2 O 2 in the inflammatory and severe inflammatory solutions, has prevented specimens from exhibiting passive behavior under these conditions [59].Hence anodic branches do not exhibit Tafel behavior [60], making it impossible to obtain accurate measurements of anodic Tafel slopes (b a ) meaning the SterneGeary analysis became invalid.
The I corr values were increased after adding the H 2 O 2 to the PBS for all specimens, indicating higher corrosion rates during inflammation (Supplementary Table ST3).For the inflammatory condition compared to the normal condition, the E corr values were electropositive, similar to the OCP values in  From the point of view of immersion times' effect on electrochemical behavior, it can be said that the Icorr and Rp values have decreased and increased respectively in all conditions after 12 h of immersion.It can be explained by the formation of a stable protective oxide film in the normal condition that protects patterned materials' surfaces from further damage.H 2 O 2 undergoes catalytic decomposition into H 2 O and O 2 a prolonged immersion times during severe inflammatory and inflammatory conditions, thus reducing the oxidizing power of the solutions in these cases.In both inflammatory conditions, I corr values for patterned CP-Ti specimens exhibits a gradual increase, indicating that the implant surface oxidizes when exposed to H 2 O 2 -containing solutions [61].Nevertheless, the formed oxide dissolves more quickly than is necessary for optimal protection.The patterned Tie6Ale4V layer demonstrated superior corrosion resistance than patented CP-Ti under all simulated conditions, which correlates with other data [62].When inflammation and severe inflammation conditions were present, I corr of patterned Tie6Ale4V specimens increased as compared to normal conditions.As a result of vanadium oxides (VO 2 , V 2 O 4 , V 2 O 5 ) dissolving in the native oxide layer on the Tie6Ale4V alloy under inflammatory conditions described in the literature, cation vacancies are generated in the passive film, resulting in an increase in I corr [62,63].It is further enhanced by the presence of H 2 O 2 molecules, lactate and chlorine ions.It is not straightforward to describe how BSA and CLH influence the corrosion process of patterned Ti6Al4V layer, as they can have an effect on both anodic and cathodic reactions [54].
It has been reported that the BSA and CLH can accelerate corrosion through their effect on the dissolution of the oxide on the Tie6Ale4V alloy through anodic reactions.As a result of the generation of soluble complexes like metal-biomolecule matrices, the available metal ion species in the oxide passive layer will be consumed and the passive film dissolution will be Fig. 5 e PDP curves of the patterned CP-Ti and Tie6Ale4V layers exposed in simulated normal, inflammatory and severe inflammatory conditions for 1 and 12 h.accelerated, which may be a major cause of the expedited anodic reaction kinetics associated with the co-existence of CLH and BSA biomolecules [54,55].When comparing the I corr of patterned layers with the results obtained from the previous study on flat specimens, it is evident that in both cases of Tie6Ale4V and CP-Ti, the patterned specimens have a lower I corr , indicating better corrosion performance of them in all simulated conditions.The lower I corr of patterned Ti layers can be attributed to a number of factors arising from the surface topography difference induced by the L-PBF process.
The beneficial influence of periodic patterns on the corrosion response of Ti implants was observed in earlier studies [36,37].As seen in Fig. 3, creating square patterns on the surface of flat substrates increases the contact angle.Indeed, patterned samples had a relatively higher contact angle (Fig. 3), which resulted in fewer tendencies to absorb water molecules on the surface, leading to a high corrosion resistance.Consequently, aggressive ions reach the substrate at a slower rate and the substrate is less exposed to corrosive media and agents.It has also been established that Ti6Al4V implants fabricated through the L-PBF process exhibit a lower I corr and a nobler E corr in biological solutions when compared to wrought counterparts [60,64,65].

Electrochemical impedance spectroscopy
Figs. 6 and 7 present the Nyquist and Bode modulus/phase plots of EIS tests.In Nyquist plots, the diameter of capacitive semicircles decreases when H 2 O 2 and HCl are added to PBS, which indicates that corrosion resistance, is deteriorating under the inflammatory condition.When BSA and LCH are introduced into the medium of severe inflammatory conditions, the diameter of semicircles decreases sharply, indicating a drastic reduction in corrosion resistance [54,55].The patterned Tie6Ale4V exhibits nobler electrochemical behavior than patterned CP-Ti in all conditions, as evidenced by the largest area of capacitive semicircles and highest modulus values at low frequency (Figs. 6 and Fig. 7aec).Phase angles of about 40 and 60 in Fig. 7f for patterned specimens implies almost complete destruction of passive film under severe inflammatory conditions.The EIS data is analyzed using equivalent circuits shown in Supplementary Fig. S5.In both circuits, R s represents the resistance of the corrosion test solution between the R.E. and W.E. The circuit in Supplementary Fig. S5a was used for fitting the data obtained in normal and inflammatory conditions and it contained the constant phase element of the passive layer parallel with the resistance of the passive layer (CPE pl /R pl ).A constant phase element (CPE) formalism is applied to express the capacitive characteristics of coating surfaces due to the "scattering effect" resulting from the nonhomogeneous nature of the passive layer's surface [66]: where j 2 ¼ À1, u is an angular frequency (rad.sÀ1 ), and n ¼ 0 … 1 is the power coefficient (with n ¼ 1, CPE behaves as a pure capacitance, and for n ¼ 0 as a pure resistance).The value of Q is usually used to approximate interfacial capacitance, but this approach does not always consistently work [66].The effective capacity (C eff ) for each passive layer was calculated using the Brug model [67]: where R T is the total resistance sum (R ct þ R pl ).The passive layer thickness (d eff ) can be estimated using the equation [68] as follows: where ε is the relative permittivity of the material (around 45 for TiO 2 passive film [68]), and ε 0 is the vacuum permittivity (8.85 Â 10 À14 F cm À1 ).As seen in Supplementary Fig. S5b, the circuit for severe inflammatory solution contains two-time constants and a Warburg component due to the difference in impedance characteristics.Here the capacitance of the electrical double layer at the interface between the substrate and solution is denoted by CPE dl and the resistance to charge transfer is shown by R ct [69].As a phase element, Warburg open terminus (W o ) demonstrates semi-infinite length diffusion of corrosion ions across the passive layer [70].A straight line at an angle of 45 appears at the low-frequency region of the Nyquist plots in the severe inflammatory condition (Fig. 7c).This can be attributed to the Warburg impedance formed at this frequency [70].It was found that the best fit to the data was achieved with the W o element connected in series with the R ct .W o element can be expressed mathematically as follows [71]: where W o-R and W o-t correspond to the Warburg pseudoresistance and diffusion time, respectively.The W o-P represents the power coefficient varying as 0 … 1.In the diffusional process assumes that characteristic time W o-t ¼ d 2 /D diff , where d is the diffusion layer thickness and D is the effective diffusion coefficient of the corrosive ions.The fitted results of EIS and computed values by equations (1)e( 4) are itemized in Supplementary Table ST4 and the values of each individual parameter of Warburg elements in Supplementary Table ST5.All R pl values for patterned specimens decreases under inflammatory and severe inflammatory conditions, indicating that corrosion resistance is weakening.The reaction between Ti and H 2 O 2 results in the formation of the less stable Ti(IV)eH 2 O 2 complex compounds, which do not offer a high level of resistance against corrosive agents [56].This lower compactness of the passive film is reflected in the higher CPE pl [66].Much higher values of CPE pl in the inflammatory and severe inflammatory solutions indicate further drop in corrosion resistance.When a specimen exposed to H 2 O 2 -containing strong acidic solutions, the passive film is damaged, causing the surface to dissolve more and increasing the cathode reaction rate [52].
Values of R pl in the severe inflammatory have been noticeably decreased compared to inflammatory conditions, showing a synergetic behavior of both CLH and BSA on patterned Ti layers degradation [54,55].It was recently reported that biomolecules in the active layer are aligned to produce an electric field throughout the passive layer due to ROS intermediates' high oxidation potential [72].It is also possible CLH produces a lactate-chelating compound with Ti that would explain a decrease in R pl during severe inflammation [73].All specimens displayed an increase in R pl after 12 h in all conditions, indicating that a passive protective film was forming and growing, facilitating the self-healing of passive layers in inflammatory media including the effect of the decomposition of H 2 O 2 to H 2 O and O 2 [52].The rate of metal dissolution with the broken passive film is generally related to the R ct and here it is seen that R ct for patterned Tie6Ale4V is slightly larger than that for patterned CP-Ti for both immersion times.Therefore, patterned Tie6Ale4V layers display better corrosion protection performance than patterned CP-Ti layers on their own substrates in the severe inflammatory solution [62].
Supplementary Table ST4 shows that the surface patterning has effectively improved the corrosion resistance of the flat specimens in all simulated media.A lower CPE pl of patterned Ti layers than flat Ti specimens indicates that the passive film that forms over patterned layers has a lower defect density.W o-R values rise in patterned Tie6Ale4V layers under severe inflammation (Supplementary Table ST5) as a result of the formation of strong oxide layers, as reported [74,75] for CP-Ti and Tie6Ale4V corrosion in HCl-based fluids.Both W o-R and W o-t values are likely to increase with immersion time due to the increased penetration pathway of electrolyte through a thicker passive layer.
The equivalent circuit proposed for patterned specimens under severe inflammatory conditions (Supplementary Fig. S5b) differs from the circuit previously reported in our study for fitting EIS plots of flat Tie6Ale4V and CP-Ti specimens under the same condition [35].The Warburg impedance and R ct indicates that diffusion of corrosive ions across the interface between the passive layer and the patterned substrate is responsible for electrochemical reactions in severe inflammatory condition [70].Indeed, patterned layers on the surface Ti have caused a change in the corrosion mechanism compared to flat Ti in the severe inflammatory environment.The proposed corrosion mechanism for patterned Ti layers under the severe inflammatory condition is schematically illustrated in Fig. 8 where the cross-sectional view of a square printed pattern on the Ti substrate is shown.The gradient in oxygen concentration between the exterior and interior parts of the square patterned wells results from the hindered mass transport process within these struts.This gradient leads to an inferior surface potential of internal parts than at the outside parts, so more metallic ions are released and accumulated at the interior due to the anodically polarized Ti surface in localized corrosion pits.On the other hand, electron transfer to the outer part of the pit causes the cathodically polarized outer parts of the struts.
Taking into account that BSA is positively charged in the severe inflammatory condition at pH z 3 [76], migration of positively charged BSA into the exterior part of the localized pit is essential in order to maintain charge balance.Larger amount of BSA hence accumulates in the outer parts, possessing higher electric charge on sharp and rough structures, with their number gradually increasing with the immersion time.BSA adsorption can also be further enhanced by the surface potential difference [77].Additional negative charges are also generated by the potential difference between the exterior and interior of the corrosion pit, enabling them to interact with positive BSA particles, promoting their adsorption onto the alloy surface within the interior parts the localized corrosion regions [78].In the vicinity of corrosion pits, BSA absorption and accumulation affect the penetration and interaction of corrosive agents across the passive layer and the impedance response exhibits the unique Warburg impedance characteristic.Additionally, Cl À ions in the solution can trap under a mass of BSA particles near the broken passive film to form a charge transfer and an electric double layer [79].

Surface morphology after PDP tests
SEM images and corresponding EDS spectra for the morphologies of patterned specimens after PDP in different conditions tests are shown in Fig. 9 and Supplementary Fig. S6.The PDP test of patterned Ti materials after 12 h immersion generally resulted in selective dissolution, pitting occurrence, and passivation decomposition.It can be seen in Fig. 9a that patterned CP-Ti layer is susceptible to inferior pitting corrosion under normal condition.However, fewer pits and localized corrosion products are detectable on the patterned Tie6Ale4V surface in same conditions (Fig. 9d).Compared to the normal conditions, severe inflammatory and inflammatory conditions show a decrease in Ti and an increase in O (Supplementary Fig. S6), indicating a higher growing the oxide film over immersion time for the both patterned layers.
Lots of wide and deep pits can be seen on the struts surface in inflammatory condition (Fig. 9b), indicating the degradation of patterned CP-Ti layer in the presence of H 2 O 2 and Cl À ions [80].The Cl À accumulation at the surface of localized corrosion regions leads to rupture of the oxide film (Supplementary Fig. S6b): when Cl À ions are adsorbing on oxygen vacancies on the surface of a metal oxide in the presence of H 2 O 2 , they are produced extra cation vacancies.The condensation of excessive cation vacancies facilitates the formation of pit nuclei [81].According to Fig. 9b and e, pits seem to propagate along non-melted particle interfaces at different rates and in various directions, starting from a local defect on patterned surfaces (void).It has been noted [82] that pits tend to form in areas with a high residual tensile stress in porous Ti materials produced by L-PBF.
The patterned Ti specimens under the severe inflammatory condition were covered almost completely with corrosion products.The original morphology of these specimens, however, can still be observed.It is seen from Fig. 9c that numerous corrosion pits are randomly distributed on the corroded surface of patterned CP-Ti, and the number of which is significantly higher than patterned Tie6Ale4V (Fig. 9f).The presence of more carbon in regions "C" (Fig. 9c) and "F" (Fig. 9f) correlates with the envisaged CLH and BSA adsorption to patterned Ti layers under the severe inflammation [55,73,83].

Conclusions
In this work, the authors have studied the electrochemical corrosion behavior of AM-patterned CP-Ti, and Tie6Ale4V layers under simulated normal, inflammatory, and severe inflammatory conditions.The electrochemical stability of patterned layers was evaluated by OCP, PDP, and EIS tests.This study has drawn the following main conclusions: (1) The patterned CP-Ti layer showed different morphology with stair-step struts compared with patterned Tie6Ale4V alloy.It was attributed to the difference in the size of powders with the laser beam diameter.(2) The wettability results suggest that the PBS droplet's contact angles were increased on AM-patterned Ti layers compared with their flat counterparts, but hydrophilic behavior has remained.(3) The OCP monitoring shown values moving towards the nobler direction with the increase of immersion time and it is explained by an increase in the passive layer's thickness under normal conditions with the decomposition of H 2 O 2 during inflammatory and severe inflammatory conditions.(4) The PDP results showed that the I corr values of the patterned layers is increased in the order: normal < inflammatory < severe inflammatory, indicating the role of H 2 O 2 , BSA, and CLH in destruction of the passive layer.(5) The equivalent circuit scheme obtained under the severe inflammatory condition differs from normal and inflammatory conditions, which is connected with altered corrosion mechanism when charged albumins are attracted to the localized pitting areas.There they cause the diffusion of corrosive species controlling the electrochemical process at the interface between the metal and the passive layer.(6) The PDP and EIS tests indicated that patterned Tie6Ale4V layers had better electrochemical performance under all conditions than the patterned CP-Ti.SEM micrographs also confirmed that the pitting tendency of patterned CP-Ti layers was higher than the patterned Tie6Ale4V alloy.(7) Comparing the results of this study with the previous one showed that the patterned specimens have better electrochemical stability than the flat counterparts because of low wettability.Further studies on the biological compatibility of patterned specimens will be required for the application of fabricated biomaterials in implantology.

Fig. 1 e
Fig. 1 e The 2D and reconstructed 3D OM images of patterned (a, b) CP-Ti and (c, d) Tie6Ale4V (grid size ¼ 250 mm).For b) and d) note vertical z-scale number shows the absolute specimen position during scanning and not the depth of the patterned wells.

Fig. 2 e
Fig. 2 e The top-surface SEM micrographs of patterned (a, and b) CP-Ti and (c, and d) Tie6Ale4V layers in different magnifications; (e) XRD patterns of patterned CP-Ti and Tie6Ale4V layers.
j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 6 : 3 5 6 e3 7 0

Fig. 3 e
Fig. 3 e The PBS solution droplet contact angle for the flat (a) CP-Ti and (b) Tie6Ale4V surfaces; and for the patterned (c) CP-Ti and (d) Tie6Ale4V surfaces.In patterned surfaces, the droplet covered one-two wells.

Fig. 5
Fig.5shows the potentiodynamic polarization (PDP) plots for patterned CP-Ti and Tie6Ale4V layers after different immersion times in simulated normal, inflammatory and severe

Fig. 4 e
Fig. 4 e OCP monitoring of the patterned CP-Ti, and Tie6Ale4V layers exposed in simulated (a) normal, (b) inflammatory, and (c) severe inflammatory conditions for 1 h and 12 h.

Fig. 4b .
Fig. 4b.This corrosion behavior is consistent with previous studies of H 2 O 2 -containing biological solutions.A combination of BSA, CLH, and H 2 O 2 in PBS (severe inflammatory media) result in an increase of I corr and a decrease in R p , suggesting a synergetic action of these compounds on the corrosion behavior of patterned Ti specimens.From the point of view of immersion times' effect on electrochemical behavior, it can be said that the Icorr and Rp values have decreased and increased respectively in all conditions after 12 h of immersion.It can be explained by the formation of a stable protective oxide film in the normal condition that protects patterned materials' surfaces from further damage.H 2 O 2 undergoes catalytic decomposition into H 2 O and O 2 a prolonged immersion times during severe inflammatory and inflammatory conditions, thus reducing the oxidizing power of the solutions in these cases.In both inflammatory conditions, I corr values for patterned CP-Ti specimens exhibits a gradual increase, indicating that the implant surface oxidizes when exposed to H 2 O 2 -containing solutions[61].Nevertheless, the formed oxide dissolves more quickly than is necessary for optimal protection.

Fig. 6 e
Fig. 6 e EIS plots of the patterned CP-Ti and Tie6Ale4V layers exposed in simulated normal, inflammatory and severe inflammatory conditions for 1 and 12 h: (aec) Nyquist plots.Nyquist plots for the severe inflammatory condition have been enlarged for better display.

Fig. 7 e
Fig. 7 e EIS plots of the patterned CP-Ti and Tie6Ale4V layers exposed in simulated normal, inflammatory and severe inflammatory conditions for 1 and 12 h: (aec) Bode modulus, and (def) Bode phase plots.

Fig. 8 e
Fig. 8 e Schematic representation of the corrosion mechanism for patterned Ti layers in the severe inflammatory media.
j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 6 : 3 5 6 e3 7 0

Fig. 9 e
Fig. 9 e Top-surface SEM micrographs of patterned specimens exposed in simulated normal, inflammatory and severe inflammatory conditions for 12 h: (aec) CP-Ti, and (def) Tie6Ale4V.SEM micrographs for the severe inflammatory condition have been shown in different magnifications.