Corrosion Behavior of HA-316 L SS Biocomposites in Aqueous Solutions

316L stainless steel and Hydroxyapatite (5, 20 and 50 wt. (%) HA)-316L stainless steel composites were fabricated by mechanical alloying technique, pressing and sintering from 316L and HA powders. The corrosion behavior of both sintered 316L and HA-316L composites was evaluated by electrochemical techniques in simulated body fluid (Ringer’s solution) and in 0.1M HCl solution which simulates occluded cell corrosion conditions. The results indicate that 316L stainless steel and HA-316L composites are passive in Ringer’s solution and active in HCl solution. All materials are highly corrosion resistant in Ringer’s solution with corrosion current density in the order of 10 A cm or less. Both 316L stainless steel and 5% HA-316L composite present good corrosion resistance in HCl with corrosion current density in the order of 10 A cm. The corrosion resistance decreases with increasing HA content in both solutions.


Introduction
316L austenitic stainless steel (316L SS) is widely used as surgical implant material due to its biocompatibility, corrosion resistance and good mechanical strength 1 .Unfortunately, 316L steel only forms weak mechanical bond with the bone tissue due to its biological inertness, which turns difficult the fixing of implant to live bone.In order to improve the osseointegration of these implants, different studies were performed on the deposition of hydroxyapatite (HA) on stainless steel [2][3][4][5] .Indeed, HA (Ca 10 (PO 4 ) 6 (OH) 2 ) has a similar chemical composition and crystallographic structure as the natural apatite of human body bone and consequently can form strong bonds with the bone.Simultaneously, the HA coating helps to prevent the release of metal ions from the steel implant and their accumulation in the internal organs.Another alternative to improve the interaction between the steel implant and the bone is the development of 316L steel based biocomposites containing HA [6][7][8][9][10] .Fan et al. 9 studied the formation of bone-like apatite on HA/316L SS composites in simulated body fluid (SBF) and proved that the composites are excellent bioactive materials.Nevertheless, HA is brittle and the mechanical resistance of the HA-316L composites is lower than that of 316L stainless steel 7 .It is well known that stainless steels are generally passive in neutral media but are susceptible to pitting and crevice corrosion in chloride media.These forms of corrosion are associated to local acidification and generally lead to a severe localized active dissolution 11 .
Tulinski 12,13 showed that HA addition improved the corrosion resistance of HA/ nitrided nickel-free austenitic stainless steel in Ringer's solution.No work was found about the corrosion resistance of HA/316L SS composites.
This work aimed to study the corrosion behavior of HA/316L SS biocomposites in simulated physiological solution (Ringer's solution) and under simulated occluded cell corrosion conditions (0.1M HCl).The investigation was based on electrochemical techniques.

Material
HA powder was prepared by wet precipitation using Ca(NO 3 ) 2 .4H 2 O and H 3 PO 4 reagents, followed by filtration, calcination at 900 °C for 2 h and deagglomeration.The whole procedure of preparation was detailed elsewhere 14 .316L stainless steel powder (size ranging from 5 to 30 µm) and HA powder were mixed in the following proportions, 95/5, 80/20 and 50/50 weight percent.The mixtures were homogenized for 10 h using a dry planetary ball mill (Fritsch, model Pulverisette 5) under the following conditions: stainless steel balls (18 mm-diameter) and vessel (250 mL), ball to powder mass ratio 10:1, speed 120 rpm.Then, the mixtures were uniaxially pressed under 32 MPa pressure and then isostatically pressed under 200 MPa pressure.The densified mixtures were heated under vacuum in a resistive furnace up to 1200 °C at a 10 °C/min rate, maintained at 1200 °C for 2 h and then naturally cooled to room temperature.

Electrochemical study
Cylindrical slices (of nearly 9 mm diameter and 3 mm thickness) were cut from the sintered 316L and HA-316L Materials Research composite rods and then mounted in resin.The cross-section of the electrodes (about 0.7 cm 2 area) was mechanically ground up to a 1200-grit finish, rinsed with distilled water and dried.
The electrolytic solutions used for the electrochemical study were the naturally aerated Ringer's solution (whose composition is: 8.6 g NaCl + 0.3 g KCl + 0.33 g CaCl 2 -H 2 O in 1000 mL deionized water) and the naturally aerated 0.1M HCl solution.The counter electrode was a square-shaped platinum sheet of 18 cm 2 area and the reference electrode was the saturated calomel electrode (SCE).The experiments were performed at room temperature.
Open-circuit potential, electrochemical impedance spectroscopy (EIS) and polarization measurements were made using the Electrochemical Interface SOLARTRON mod.1287A and the Frequency Response Analyzer SOLARTRON mod.1260 A, controlled by the Ecorr/Zplot SOLARTRON mod.125587S software.Triplicate assays were performed and reproducibility was verified.
Prior to polarization experiments, the working electrodes were immersed in the solutions for 3 h, taking the moment of immersion as zero time.Then, impedance measurements at open-circuit potential were made using a sinusoidal signal of 10 mV amplitude and frequencies in the 0.01 Hz -100 kHz range.Cathodic and anodic polarization was then carried out potentiodynamically with 1 mV s -1 sweep rate.After each run, the samples were reground with emery paper to a 1200-grit finish in order to remove any product formed on the metal surface which could affect the following tests, rinsed with distilled water and dried.

Microstructure of HA/316L stainless steel biocomposites
Figure 1 shows micrographs of polished surfaces of the investigated materials obtained by scanning electron microscopy (SEM).The microstructure of sintered 316L presents austenitic grains and pores (black spots) (Figure 1a).The microstructure of 5% HA-316L composite is constituted of austenitic grains (grey regions), HA (light regions), an eutectic phase at the grains boundaries (represented by E letter) and pores (Figure 1b).The eutectic phase was also observed by Szewczyk-Nykiel 10 .It can be observed that HA was partially removed during grinding, leaving many holes.The chemical analysis by energy dispersive spectroscopy (EDS) of the austenitic grains in 5% HA-316L composite (P2 and P3-Figure 1b) showed the presence of the constituent elements of 316L stainless steel, Fe, Cr, Ni and Mo and some trace of P (Table 1).P element is not found in sintered 316L steel (P1-Figure 1a).This means that some phosphorus diffused from HA phase into austenitic grains during sintering of the composites 10 .In the light regions (P4 and P5-Figure 1b), large amount of Ca and O are present, as well as Fe and Cr (Table 1).Thus, it seems that Cr and Fe have migrated from austenitic phase.The eutectic phase (P6-Figure 1b) contains significative amounts of Fe, Cr, Mo, Ni and P and some trace of Ca (Table 1), and can be the result of some chemical reaction occured during sintering.

Materials Research
Corrosion Behavior of HA-316L SS Biocomposites in Aqueous Solutions Phosphorus has also diffused from HA phase into austenitic grains during sintering of the 20% and 50% HA-316L composites (P7-Figure 1c and P10-Figure 1d, Table 1).In the 50% HA-316L composite some steel grains are rich in Ca and O too (P12-Figure 1d, Table 1) and in the HA regions, besides the presence of Ca, P and O, Fe and Cr were also detected (P11-Figure 1d, Table 1).This could be related to some reaction between 316L steel and HA during sintering 7 .Similarly, in the HA regions of the 20% HA-316 composite, Fe, Cr and Ni are present (P8-Figure 1c, Table 1).Some regions with high concentrations of Fe, Cr, Ni, Mo, Ca, P and O and that look like an eutectic phase (E) were also observed in the 20% HA-316L composites (P 9-Figure 1c, Table 1).
Besides diffusion, possible transformation during sintering could be considered.The X ray diffraction patterns of 316L and composite samples only revealed the peaks of the austenite phase (JCPDS card number 33-0397) 15 , and also peaks of HA (JCPDS card number 09-0432) 15 for the composites, which attests that the austenite phase did not undergoe transformation during sintering.

Corrosion behavior of HA/316L biocomposites in Ringer's solution
The open-circuit potentials (OCPs) of 316L and HA/316L composites tended to stabilize before 3 h immersion in Ringer's solution (Figure 2).The values of all OCPs are found in the stability regions of Fe 2 O 3 and Cr 2 O 3 of the Fe-H 2 O and Cr-H 2 O Pourbaix' diagrams 16 , which shows the passive behavior of all materials.The OCPs values measured at 3 h (E corr corrosion potentials) shifted in the negative direction as the HA content increased (Table 2), which means that the addition of HA turned the materials less noble.Tulinski 13 found the same tendency for HA (0, 5, 10 and 15 wt.%) -nickel free stainless steel composites in Ringer's solution.
The polarization curves of 316L stainless steel and HA-316L composites in Ringer's solution at room temperature are presented in Figure 3.The shape of the curves is very similar for all materials, which shows that the anodic and cathodic reactions occurring on the different surfaces are identical.The cathodic domain is related first to the reduction of dissolved oxygen and at lower potentials to the simultaneous reduction of water.The anodic branch presents a pseudo-plateau corresponding to the passive region.It was shown that the passive film on 316 stainless steel is mainly constituted of chromium oxide and lower contents of iron and molybdenum oxides, without presence of nickel oxide 17 .For higher potentials the anodic current density increases strongly due to the breaking of passivity related to pitting corrosion.
Nevertheless, the anodic and cathodic current densities increase as the HA content in the composites increases.
Accordingly to the variation of E corr values, the nullcurrent potential E i=0 shifted in the negative direction as the HA content increased (Figure 3 and Table 2).The E i=0 values are more negative than the E corr ones, which may be due to the partial removal of oxide passive film during the cathodic polarization.
The corrosion current density i corr determined by extrapolation of the cathodic Tafel lines to the null-current potential are reported in Table 2 for the four materials.A significant decrease in corrosion resistance is observed for  the 20 and 50 % HA-316L composites.This may be related to the enrichment of HA phase with Cr.Nevertheless, the corrosion current densities are in the order of 10 -6 A cm -2 or less for all materials, which is characteristic of corrosion resistant materials.Tulinski 13 showed that the corrosion resistance of HA-nickel free stainless steel composites increases with increasing HA content between 0 and 10% but further decreases for 15% HA-SS composite.
The passive current density i pass measured for 20 and 50% HA-316L composites is also higher than that obtained for 316L steel and 5% HA-316L composite which depicts a lower protective character of their passive film (Table 2).
The breakdown potential E b (measured when the anodic current density suddently increases) shifts in the negative direction as the HA content increases, which is characteristic of a higher susceptibility to pitting (Table 2).
The values of E corr , i corr , i pass and E b for 316L steel are close to the values obtained by Talha 18 in Ringer's solution containing NaHCO 3 at 37 °C.
The spontaneous corrosion behavior of metals and alloys is better evaluated from the impedance measurements at corrosion potential.Indeed, the use of this technique only disturbs the potential from +10 mV to -10 mV around the spontaneous corrosion potential.Thus, during the measurements, the electrode surface is only slightly changed.
Figure 4 shows the Bode diagrams obtained for 316L stainless steel and HA-316L composites in Ringer's solution at corrosion potential.For 316L stainless steel, the impedance diagram presents one time constant.The slope of log (Impedance) versus log (Frequency) is close to -1 and the phase angle close to -90° on a large frequency range, which is characteristic of passive material.For the HA-316L composites, Bode diagrams with two time constants were obtained.Only the 5% HA-316L composite maintains a behavior close to that of 316L steel, i.e. high impedance values, slope of log (Impedance) versus log (Frequency) close to -1 and a phase angle around -70° on a large frequency range.
The corrosion resistance of the materials can be estimated for comparison by the values of impedance modulus at low frequencies (Table 2).316L stainless steel and 5% HA-316L composite present values close to 10 5 Ω cm 2 and 20 and 50% HA-316L composites values around 10 4 Ω cm 2 .
Though there is a clear evidence that corrosion resistance decreases with increasing HA content, the values of i corr , i pass and Z (at low frequencies) must be considered carefully.Indeed, the porosity of the composites also increases with increasing HA content (Figure 1), which can lead to a significant error on the measured sample area exposed to the corrosive medium, and consequently to an error on i corr , i pass and Z values.

Table 2. Corrosion potential (E corr
), null-current potential (E i=0 ), breakdown potential (E b ), corrosion current density (i corr ), passive current density (i pass measured at the middle of the passive region) and impedance modulus (|Z| measured at 0.01 Hz) of sintered 316L stainless steel and HA-316L composites in Ringer's solution at room temperature.

Materials Research
Corrosion Behavior of HA-316L SS Biocomposites in Aqueous Solutions

Corrosion behavior of HA/316L biocomposites in 0.1M HCl solution
The open-circuit potentials (OCPs) of 316L and HA-316L composites also stabilized before 3 h immersion in 0.1M HCl solution (Figure 5).The values of all OCPs are found in the stability regions of Fe 2+ and Cr 3+ ions of the Fe-H 2 O and Cr-H 2 O Pourbaix' diagrams 16 , which depicts the active behavior of all materials.The OCPs values measured at 3 h (corrosion potentials) shifted in the negative direction as the HA content increased (Table 3), which means that the addition of HA also turned the materials more active in HCl solution.It can be noted that the corrosion potentials of both 316L and 5% HA-316L composite are very close.The same occurred for both 20% and 50% HA-316L composites.The E corr values are generally more negative than those measured in Ringer's solution, which shows that the materials are less noble in HCl solution.
The polarization curves of 316L stainless steel and HA-316L composites in 0.1M HCl solution at room temperature are shown in Figure 6.The cathodic branches are related to hydrogen evolution.It seems that the kinetics of H + reduction on 316L steel is controlled by activation but is diffusion-controlled on the HA-316L composites.The anodic curves present different shapes.316L steel showed a narrow passive region between nearly -0.260 and -0.160 V/ SCE followed by the film breakdown above -0.160V/SCE.For the 5% HA-316L composite, the cathodic polarization seems to have led to the dissolution of the passive film since during the anodic polarization, the curve presented an anodic peak followed by a wide passive region between -0.150 and 0.150 V/SCE.The corrosion and passive current densities are similar for both 316L and 5% HA-316L composite (Table 3).Nevertheless, the passive film on the 5% HA-316L surface only undergoes pitting corrosion for potentials above 0.150 V/SCE.The anodic branches for both 20 and 50% HA-316L composites are very similar.Both materials are active and no passivation is observed.Thus E b and i pass values

Conclusions
Sintered 316L stainless steel and 5, 20 and 50 wt.(%) HA-316L composites are passive in Ringer's solution and active in 0.1M HCl solution which simulates occluded cell corrosion conditions.
The corrosion resistance decreases with increasing HA content in both solutions.
Only 316L stainless steel and 5 wt.(%) HA-316L composite can be considered corrosion resistant in 0.1M HCl solution.
are not available for these materials.Their corrosion current densities are one order of magnitude higher than the values measured for 316L and 5% HA-316L composite (Table 3).
Figure 7 shows the Bode diagrams obtained for 316L stainless steel and HA-316L composites in 0.1M HCl solution at corrosion potential.Both 316L stainless steel and 5% HA-316L composite show a slope of log (Impedance) versus log (Frequency) close to -1 and a phase angle near -70° in the intermediary frequency range frequency range.
The corrosion resistance of the materials can be estimated by the values of impedance modulus at low frequencies (Table 3).316L stainless steel and 5% HA-316L composite present values in the order of 10 3 Ω cm 2 and 20 and 50% HA-316L composites values around 10 2 Ω cm 2 .

Figure 2 .
Figure 2. Variation of open-circuit potential with time for sintered 316L stainless steel and HA-316L composites in Ringer's solution at room temperature.

5 Figure 3 .
Figure 3. Polarization curves of sintered 316L stainless steel and HA-316L composites in Ringer's solution at room temperature.

Figure 4 .
Figure 4. Bode diagrams of sintered 316L stainless steel and HA-316L composites in Ringer's solution at room temperature and at corrosion potential.

Table 3 .
Corrosion potential (E corr ), null-current potential (E i=0 ), breakdown potential (E b ), corrosion current density (i corr ), passive current density (i pass measured at the middle of the passive region) and impedance modulus (|Z| measured at 0.01 Hz) of sintered 316L stainless steel and HA-316L composites in 0.1M HCl solution at room temperature.

Figure 5 .
Figure 5. Variation of open-circuit potential with time for sintered 316L stainless steel and HA-316L composites in 0.1M HCl solution at room temperature.

Figure 6 .
Figure 6.Polarization curves of sintered 316L stainless steel and HA-316L composites in 0.1M HCl solution at room temperature.

Figure 7 .
Figure 7. Bode diagrams of sintered 316L stainless steel and HA-316L composites in 0.1M HCl solution at room temperature and at corrosion potential.