Enhanced Osteoblast Proliferation and Corrosion Resistance of Commercially Pure Titanium Through Surface Nanostructuring by Ultrasonic Shot Peening and Stress Relieving

Titanium and its alloys are frequently used for dental implants because of their excellent combination of mechanical properties, chemical stability, and biocompatibility.^1,2  The predictability and success of dental implants is well established; however, the success rate has been reported to vary with sites and patients.³  Therefore, modification in design as well as the surface of implants is essential for improving the process of osseointegration and reducing healing time, stress shielding, and force concentration, especially at bone sites that are less dense.^3–7  The major concern in the near future is to improve bioactivity of titanium implants, especially with respect to bone cells, to enhance the process of osseointegration and its effectiveness.

Several factors contribute to the process of osseointegration of dental implants and consequently the duration of healing. Surface characteristics of implants, such as topography, chemistry, charge, and wettability play a major role in biologic success of implants. Therefore, various surface modification processes have been developed and applied to titanium implants to achieve more biocompatible surface.^7,8 

Nanosize roughness on titanium implant surface through nanostructured coatings or compacting nanosize constituent particles and fibers has been observed to increase osteoblast functions, especially the adhesion, proliferation, and deposition of calcium-containing minerals on nanosize roughness, with respect to the conventional surfaces with micron size grains.^9–12  However, nanosize metal particles are known to be expensive and technique sensitive; therefore, alternative methods of creating nanosize surface roughness on titanium implants are highly warranted.

Fatigue life of commercially pure titanium (c p-Ti) has been reported to be improved by surface nanostructuring^13,14 ; therefore, surface nanostructure would be an additional attribute to implants experiencing fatigue damage. Furthermore, in view of human tissue being nanostructured, a practical approach for the design of the next generation of implants, with nanostructured surface, is considered to be highly promising.¹⁰  This study was undertaken to characterize surface microstructures of c p-Ti resulting from ultrasonic shot peening by steel balls as well as on subsequent annealing and to examine their effects on osteoblast proliferation and corrosion resistance in simulated body fluid (SBF).

Ultrasonic shot peening is a process of cold working, in which small spherical peening media with sufficient hardness are stroked onto the substrate surface. Each shot striking the metal surface acts as a tiny peening hammer, causing severe plastic deformation and formation of nanostructure on the metal surface. The impact directions of the balls onto the sample surface are random. Indigenous c p-Ti (grade 2) was procured from M/S Mishra Dhatu Nigam (MIDHANI, Hyderabad, India) in the form of a plate of 10-mm thickness. A disc of 65-mm diameter and 4-mm thickness was machined from the plate. One side of the disc was mechanically polished on emery papers from 1/0 to 4/0 grades and finally on Selvyt cloth with a slurry of alumina powder in water as abrasive media. The polished surface of the disc was subjected to ultrasonic shot peening, with stainless steel balls of 8 mm diameter, to achieve optimum combination of nanoscale surface roughness, surface area, and associated energy. Shot peening was carried out in a vibrator generator (Model SNC 1 SNC, Advanced Technology Co Ltd, Chengdu, China) at a frequency of 50 Hz, for 45 minutes, to create nanostructure at the surface. Using this technique, a very thin region of nanostructure was created at the surface of the c p-Ti disc. The thickness of the shot peened disc was reduced from 4 mm to 2 mm by machining from the other side, opposite to the shot peened one, and square samples of 10 × 10 × 2 mm size were prepared. Fifty percent of these samples, prepared from the shot peened disc, were annealed at 400°C for 1 hour, in a vacuum-sealed (approximately 10⁻⁴ Torr) silica tube. Identical square-shaped samples were prepared also from the as received (unshot peened) material to compare their osteoblast culture and corrosion resistance in SBF with those of the shot peened as well as with the subsequently annealed ones.

The microstructures of c p-Ti in the as received, shot peened, and shot peen-annealed conditions, were examined using an optical microscope (Leitz Metallux-3, Wetzlar, Germany), an atomic force microscope (Nanoscope IV Controller, Veeco, The Helios, Singapore), and a scanning electron microscope (Quanta 200F, Eindhoven, Netherland).

A human osteoblast cell line, MG 63, procured from the National Centre for Cell Sciences(Pune, India), was cultured in Dulbecco's Modified Eagle's Medium (DMEM), supplemented with heat-inactivated fetal calf serum (FCS) (10%), penicillin (100 U/mL), streptomycin (100 U/mL), and gentamycin (20 μg/mL). FCS was from Biological Industries (Haemek, Israel); all other chemicals, unless mentioned otherwise, were procured from Sigma- Aldrich Chemicals (St Louis, Mo). The cells were seeded into 25 cm² culture flasks (Nunc, Roskide, Denmark) and were allowed to grow in a controlled humidified incubator (Model 3121, Thermo Forma, Marrietta, Ohio) with 5% CO₂ and 98% humidity at 37°C. Before seeding of the cells, all of the samples of c p-Ti, in different conditions, were soaked in Extran MAO₃ phosphate-free detergent solution (Merck GmbH, Darmstadt, Germany) and subsequently autoclaved at a pressure of 15 lb/inch²  for 30 minutes. Osteoblasts were seeded at a density of 0.5 × 10⁶ osteoblasts/well onto the surface of each test sample, which was kept in a 12-well plate (Nunc, Roskide, Denmark).

Cytoproliferative activity of osteoblasts was measured by 4,5-dimethylthiazol-2-yl-2,5-diphenyl tetrazolium bromide (MTT) assay. Osteoblasts (0.5 × 10⁶ cells/well) were seeded onto 10 × 10 × 2 mm surfaces in 12-well culture plates in 1 mL complete DMEM and incubated for different durations of 24, 48, and 72 hours. After completion of the incubation period, 50 μL of 5 mg/mL MTT solution was added to the monolayers and incubated for 4 hours at 37°C. The MTT reaction was terminated by addition of 0.04 N HCl in isopropanol. The MTT formazan formed was measured spectrophotometrically (540 nm) using an EMax spectrophotometer (Molecular Devices, Sunnyvale, Calif), and the relative intensities were plotted. Each experiment was performed 4 times in triplicate, and standard deviation and variance were calculated. Proliferation of adherent cells was distinguished among the different groups by using parametric 2-way analysis of variance (Bonferroni test) as a function of time.

The concentration of nitrite, the stable end product of nitric oxide, in culture supernatants was determined by Griess' reaction. Nitrite content was quantified by extrapolation from sodium nitrite standard curve in each experiment (Figure 1). Osteoblasts were plated onto different surfaces and incubated under standard culture conditions for 12 hours; the supernatants were collected and nitric oxide was estimated by titrating for nitrite using Griess' reagent. Bacterial lipopolysaccharide (100 ng/mL/10⁶ cells) was taken as the positive control.

Electrochemical potentiodynamic polarization study was carried out in Ringer solution (NaCl 9 g/L, KCl 0.42 g/L, CaCl₂ 0.48 g/L, NaHCO₃ 0.2 g/L) at a pH of 7.2, using a Gamry Potentiostat, (Gamry PC4, Warminster, Penn). Salt concentration in the Ringer solution corresponded to that in the body fluid. A conventional 3-electrode system with high-density graphite as counter electrode, saturated calomel electrode as reference electrode, and the test specimen as working electrode, was used. The anodic polarization curves were recorded, at a scan rate of 1 mV/s. The polarization scan was conducted in the voltage range from −500 to +1500 mV with reference to the open circuit potential. Tafel extrapolation was used to determine corrosion parameters.

The microstructures of c p-Ti in the 3 conditions are shown by the optical micrographs in Figure 2. Grains of approximately 20 μm mean intercept length may be seen in the as received condition (Figure 2a and b). Refinement of grains resulting from shot peening is quite obvious in the shot peened condition (Figure 2c and d). Similar refined microstructure can also be seen in the shot peen-annealed condition (Figure 2e and f). Microstructures in the above 3 conditions, recorded by atomic force microscope and scanning electron microscope, are displayed in Figures 3 and 4, respectively. As expected, the microstructure in the as received condition was featureless, whereas nanosize grains were seen with inhomogeneous distribution in the shot peened condition. Relatively more distinct nanosize grains, with increased number density and more uniform distribution were seen in the shot peen-annealed condition. The average size of the nanosize grains was in the range of 14–20 nm. The roughness of the surface was highest in the shot peened condition and was followed in decreasing order by those of shot peen-annealed and the as received ones.

The growth behavior of human osteoblast cell line MG 63 on the as received samples was analyzed using MTT assay at 24, 48, and 72 hours of incubation. It may be seen from the bar diagrams in Figure 5 that, in general, there was increase in cell proliferation in all 3 cases, with increase in duration of proliferation. Further, there was only marginal increase in cell proliferation from shot peening; however, proliferation was enhanced significantly by the subsequent annealing treatment. This relative trend of cell proliferation, on the surface of the 3 different categories of samples, was maintained for all 3 durations of proliferation (24, 48, and 72 hours). Statistical analysis of the data of osteoblast proliferation for the 3 conditions is presented in Tables 1a through c. It may be seen that there was statistically significant difference (P < .001), in particular, in the shot peen-annealed condition. The increase in osteoblast proliferation in the shot peened and shot peen-annealed condition, with respect to that of the untreated condition, is presented in Table 2. It may be seen that while the increase in cell proliferation was almost 10% for the shot peened, it was enhanced to almost 50% for the shot peen-annealed condition. The improvement in cell proliferation on these surfaces was almost comparable for the 3 time intervals of cell proliferation.

The important parameters characterizing corrosion behavior of c p-Ti in the 3 states in SBF, at 37°C, are recorded in Table 3 with chi-square values for the respective conditions. It may be seen that corrosion rate was increased by almost 327% in the shot peened condition with respect to the as received material. On the other hand, the corrosion rate was reduced by 41% following annealing of the peened specimen as compared with that of the as received material. Thus, it is obvious that highest corrosion resistance was exhibited in the shot peen-annealed condition. Tafel plots for the 3 conditions of c p-Ti are shown in Figure 6. It may be seen that corrosion potential had become nobler, and corrosion current was reduced for the shot peen-annealed condition. These features are indicative of the presence of a strong oxide layer with effective barrier action on the surface of the shot peen-annealed specimen, which was formed from annealing at 400°C. It is obvious from the shape of the plot, that the high rate of dissolution in the shot peened condition was due to cathodic depolarization, which is indicative of weak passive film, with less effective barrier action on the shot peened surface. It may be inferred from the Tafel plot for the as received condition that passive film on the as received specimen was relatively stronger than that on the shot peened surface.

It is seen that a nanostructure (14–20 nm) was created at the surface of c p-Ti by ultrasonic shot peening and it was not only stable but became more prominent following subsequent annealing at 400°C. The observed effect of shot peening as well as that of subsequent annealing on osteoblast proliferation may be understood in terms of the surface topography, form, and stability of oxide film, and the residual stresses associated with the surface, in respective conditions. In the shot peened condition, the surface was nanostructured; oxide film could be presumed to be amorphous like that of the untreated surface, as shot peening was carried out at an ambient temperature. The film would have been relatively weaker and less uniform in thickness as compared with that on the untreated surface because of adverse effect of dislocations with high density and residual stresses in the substrate on bonding between the passive oxide film and the substrate.

The passive titanium oxide film on c p-Ti, formed at room temperature, is known to be amorphous^15,16  and to undergo transformation to crystalline form (anatase) at 276.10°C.¹⁶  In light of these facts, little role of the passive oxide film could be visualized in enhanced cell proliferation of approximately 10% in the shot peened condition. A high density of dislocations and considerably high magnitude of residual stresses would have been there in the shot peened region resulting from severe plastic deformation caused by ultrasonic shot peening.¹⁷  A peak subsurface compressive residual stress of 480 MPa was observed by Jiang et al¹⁷  in a 0.8-mm thick sheet of c p-Ti blasted by sand particles of 200–300 μm diameter. Furthermore, the sheet was bent into a curved shape on only one side due to compressive stresses associated with the surface when the sheet was sand blasted. . In the present investigation, the thickness of the shot peened disc was kept at 4 mm to avoid bending following ultrasonic shot peening with stainless steel balls of 8-mm diameter. However, the magnitude of the residual stresses associated with the shot peened surface may be visualized easily. Residual stress at the surface has been found to be detrimental for cell proliferation.¹⁸  In sharp contrast, as mentioned above, surface morphology of nanoscale has been found to be highly effective in promoting cell proliferation. Materials with nanoscale particles/grains (<100 nm) are considered quite unique in the sense that they simulate dimensions of bone constituents.¹⁰ 

It is now well established that surfaces with micro and nano topography are more conducive to proliferation of osteoblast and formation of good quality bone.^19–21  Surface roughness has been found to affect contact angle/wettability.²²  Alloy Ti6Al4V, coated with hydroxyapatite, had the lowest contact angle and best wettability. Overall bone-to-implant contact is the basis for biomechanical anchorage of dental implant. A statistically significant difference (P < .001) was found in the implant-to-bone contact percentage in machined (conventional) and resorbable blast material implants.²³  Resorbable blast material implant is characterized by ultraclean surface with macro roughness, created by blasting the implant surface with coarsely ground (180–425 μm) particles of calcium phosphate, a resorbable material, followed by specific passivation to remove the embedded particles.²⁴  In view of the above facts it is obvious that the observed increase of approximately 10% in cell proliferation on the shot peened surface was essentially due to the surface nanostructure countering the deleterious effect of surface residual stresses on cell proliferation. The marked enhancement of almost 50% in cell proliferation in the shot peen-annealed condition was essentially from the stress recovery and partly from the crystallized form of the oxide layer. The crystallized form of surface oxide on c p-Ti has been shown to be highly effective in the spreading and proliferation of cells.^25,26  While there was drastic decrease in corrosion resistance due to shot peening, there was marked increase in corrosion resistance following subsequent annealing treatment. The drastic decrease in corrosion resistance of c p-Ti in the shot peened condition, despite nanostructure at the surface, would have been from less protective passive film, as mentioned above. There would have been much variation in thickness of the passive oxide film due to inhomogeneity in dislocation density and residual stresses in the substrate. The marked increase in corrosion resistance in the shot peen-annealed condition was due to formation of a strong oxide layer on the surface, with effective barrier action, against the transfer of metal ions to electrolyte. Relatively higher corrosion resistance in the shot peen-annealed condition than in the as received state would have been from adherent and strong oxide film on the shot peen-annealed surface from the significantly large increase in grain boundary area.²⁷  In the present investigation, corrosion behavior of c p-Ti characterized by potentiodynamic polarization was similar to that observed in the earlier studies on c p-Ti¹⁷  and 304 stainless steel²⁸  in 3.5% NaCl solution. In these studies, corrosion resistance was found to be low in the sand blasted condition and significantly improved in the subsequently annealed condition. The improved corrosion resistance of these materials in the sand blasted and annealed condition was attributed to surface nanostructure formed by subsequent annealing and consequent effective passivation.

Thus, it is obvious from this investigation that there is high potential of this novel process of surface modification, through nanostructuring, in enhancing osteoblast proliferation as well as corrosion resistance of c p-Ti without any modification in surface chemistry to avoid risk of deterioration in corrosion resistance. One of the possible limitations of this process with respect to other processes may be the relatively high initial cost of the processing unit.

Abbreviations

c p-Ti

commercially pure titanium

DMEM

Dulbecco's Modified Eagle's Medium

MTT

4,5-dimethylthiazol-2-yl-2,5-diphenyl tetrazolium bromide

SBF

simulated body fluid

Authors are thankful to Dr DDN Singh (National Metallurgical Laboratory Jamshedpur, India) and G. Sudhakar Rao, (Department of Metallurgical Engineering, Indian Institute of Technology-Banaras Hindu University, India) for useful discussions related to corrosion study.