Materials Science and Engineering C 43 (2014) 375–382
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Corrosion behavior and mechanical properties of bioactive sol-gel coatings on titanium implants M. Catauro a,⁎, F. Bollino a, F. Papale a, R. Giovanardi b, P. Veronesi b a b
Department of Industrial and Information Engineering, Second University of Naples, Via Roma 29, 81031 Aversa, Italy Department of Engineering “Enzo Ferrari”, University of Modena and Reggio Emilia, Via Vignolese 905, 41125 Modena, Italy
a r t i c l e
i n f o
Article history: Received 11 April 2014 Received in revised form 30 June 2014 Accepted 13 July 2014 Available online 18 July 2014 Keywords: Sol-gel process Organic–inorganic hybrid Scratch and nano-indentation tests Corrosion process Polarization tests
a b s t r a c t Organic–inorganic hybrid coatings based on zirconia and poly (ε-caprolactone) (PCL) were prepared by means of sol-gel dip-coating technique and used to coat titanium grade 4 implants (Ti-4) in order to improve their wear and corrosion resistance. The coating chemical composition has been analysed by ATR-FTIR. The influence of the PCL amount has been investigated on the microstructure, mechanical properties of the coatings and their ability to inhibit the corrosion of titanium. SEM analysis has shown that all coatings have a nanostructured nature and that the films with high PCL content are crack-free. Mechanical properties of the coatings have been studied using scratch and nano-indentation tests. The results have shown that the Young's modulus of the coatings decreases in presence of large amounts of the organic phase, and that PCL content affects also the adhesion of the coatings to the underlying Ti-4 substrate. However, the presence of cracks on the PCL-free coatings affects severely the mechanical response of the samples at high loads. The electrochemical behavior and corrosion resistance of the coated and uncoated substrate has been investigated by polarization tests. The results have shown that both the coatings with or without PCL don't affect significantly the already excellent passivation properties of titanium. © 2014 Published by Elsevier B.V.
1. Introduction Titanium and its alloys are attractive metallic materials widely used as implants for orthopedic dental and orthodontic wires due to their excellent biocompatibility and corrosion resistance even in very aggressive environments. This aspect is related to their ability to form spontaneously a dense and stable titanium dioxide layer on the surface which provides a barrier between the bio-environment and implant [1]. However, the poor tribological behaviour caused by a high and unstable friction coefficient, the poor wear resistance and the strong tendency to seize [2,3] have restricted applications of titanium as a biomedical implant material. Indeed, when local mechanical abrasion removes the protective oxide film, the corrosion resistance of titanium alloys can be strongly decreased [4]. To avoid the early failure of the implants due to wear and corrosion and to extend the prostheses lifetime, different strategies can be followed, such as to increase the thickness of oxide films using nitric acid passivation protocols [4,5] or to apply a protective coating, which also permits to improve the substrate performances related to its surface, as bioactivity and biocompatibilty.
⁎ Corresponding author. Tel.: +39 081 50 10 360; fax: +39 081 50 10 204. E-mail address: [email protected] (M. Catauro).
http://dx.doi.org/10.1016/j.msec.2014.07.044 0928-4931/© 2014 Published by Elsevier B.V.
The sol-gel dip-coating process is a promising technique for the deposition of functional coatings at low cost on a wide range of substrates. Sol-gel technology is the process of making ceramic and glass materials at a relatively low temperature which allows the entrapping of various inorganic, organic substances and biomolecules in a glassy matrix during its formation [6]. Ten years ago, Guglielmi [7] has already discussed the potential of sol-gel coatings as a corrosion inhibiting system for metal substrates. Since then, various sol-gel based protective coatings were developed. In this work, organic–inorganic hybrid materials based on zirconia (ZrO2) and poly (ɛ-caprolactone) (PCL) were synthesized by sol-gel method and used, in sol phase, to dip-coat commercially pure titanium grade 4 (Ti-4) implants in order to improve their biological properties and to decrease wear and corrosion. ZrO2 has a high expansion coefficient very close to many bulk metals, which can reduce the formation of cracks during the high temperature curing process. ZrO2 also shows good chemical stability and high hardness which makes it a good protective material [8]. However, inorganic sol-gel coatings have many limitations, as brittleness and high temperature treatment. Many works report the introduction of an organic component into the inorganic sol-gel to form an organic–inorganic hybrid sol-gel coating to overcome these limitations [8,9]. For that purpose, the authors have used the poly (ε-caprolactone) a biodegradable,
376
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
synthetic, aliphatic polyester that has attracted a wide interest for its possible applications as a biomaterial: PCL-based three-dimensional scaffolds for orthopaedic surgery [10], and film substrates for tissue engineering [11] have been proposed. Moreover, composite substrates for tissue engineering have been prepared by sol-gel method, consisting of a poly (ε-caprolactone) matrix reinforced with sol-gel synthesized PCL/TiO2 or PCL/ZrO2 hybrid fillers [12,13] as well as several PCL-based hybrids for drug delivery, using different oxides, CaO and/or SiO2 [14–18], TiO2 [19,20], and ZrO2 [21,22]. Moreover, elsewhere the authors proved that such polymer is able to improve the mechanical properties of TiO2-based glassy matrix [20] and acts as plasticizer in the coatings preparations [9]. The Authors [23] already proved that a ZrO2/PCL hybrid coating can be used to modify the surface of Ti-4 implants by improving their biological properties (bioactivity and biocompatibility). The aim of this research is also to investigate the adhesion and the mechanical properties of the synthesized thin films and the corrosion resistance of the Ti-4 implants after coating with the prepared hybrid, when exposed to aggressive human physiological fluids and calcium phosphate surface deposits by means of electrochemical polarization tests, as reported in literature [24] 2. Materials and methods
speed of the substrate was 15 cm/min. The coated substrates were heat-treated at 45 °C for 1 day to promote a partial densification of the film without any polymer degradation. The obtained layers appear to be transparent, uniform and crackfree for Zr (1)-(2)-(3)-(4), while cracks appear on the surface of Zr(0). That observation was confirmed by a microstructural analysis performed using SEM (Quanta 200, FEI Europe Company, Netherlands). 2.3. Materials and coating characterization Scanning electron microscopy (SEM) and Attenuated total reflectance—Fourier transform infrared (ATR-FTIR) spectroscopy were employed to characterize obtained coatings. The microstructure of the films was investigated by a SEM FEI Quanta 200 equipped with EDX (energy dispersive X-ray spectroscopy). ATR/FT-IR allows to analyze the chemical composition of the coating surface. The spectra were recorded on a Prestige-21 FTIR spectrometer equipped with an AIM-8800 infrared microscope (Shimazu, Japan), using the incorporated 3-mm diameter Ge ATR semicircular prism. The spectra were recorded using an incident angle of 30° with the sum of 64 scans at a resolution of 4 cm−1 and in the 650–4000 cm−1. The spectra were elaborated by Prestige software (IRsolution).
2.1. Sol-gel synthesis
2.4. Mechanical characterization techniques
Hybrid organic–inorganic composites, consisting of an inorganic ZrO2 matrix and PCL as organic phase were synthesized by means of the sol-gel process. In order to study the influence of the PCL amount on corrosion behavior and mechanical properties of the coatings several films containing different percentage of the polymer was prepared, as shown in Table 1. Zirconium propoxide Zr(OC3H7)4 and poly (ε-caprolactone) (PCL Mw = 65000) were respectively used as inorganic and organic precursors. The solution of poly (ε-caprolactone) in chloroform was added to the solution of Zr(OC3H7)4 in ethanol–acetylaceton-water mixture. Acetylacetone was added to control the hydrolytic activity of zirconium alkoxide. After the addition of each reactant, the solution was stirred with a magnetic stirrer and the resulting sols were uniform and homogeneous. The time of gelation was 8–18 days depending on the system. Fig. 1 shows the flow-chart of the hybrid synthesis by the sol gel method. After gelation the gels were air-dried at 50 °C for 24 h to remove the residual solvents. That treatment allows to obtain glassy pieces of various sizes (see Fig. 2) without modifying the stability of the polymer, as its melting temperature is about 65 °C.
Scratch tests were carried out on a CSM Micro-Combi tester. In those tests, a controlled scratch on the coating surface was made with a diamond tip (Rockwell C diamond scratch indenter with tip radius of 200 μm). The tip was drawn across the coated surface of the sample under an increasing load (from 0.1 to 20 N) at a load rate of 4.98 N min−1 along a scratch length of 1 mm. The instrument was equipped with an integrated optical microscope, an acoustic emission detection system and a device to measure the tangential frictional force in the scratch direction. The sensors allowed to determine the critical load (LC1), evaluated as the normal load at which the first damage occurred to the coating. Three scratches were carried out in different zones for each specimen and the average value of the critical load was evaluated. Nanoindentation was performed using a Berkovitch indenter (Nanoindenter, CSM Instruments, Peseux, CH) operating at a constant load of 50 mN (loading and unloading rate = 360 mN/min), applied for 15 s, in order to maintain the indentation depth below 1 μm. A second set of nanoindentations was performed in penetration depth control, imposing a maximum penetration depth of the indenter equal to 100 nm. Nanoindentation results were analyzed according to the Oliver and Pharr procedure [25] in such a way as to determine the coating hardness and elastic module. The same penetration depth vs. load curves were also used to evaluate the amount of work required to cause the nanoindentation, and the percentage of work performed in the elastic regime (integral of the unloading curve) or plastic regime (difference between the integral of the loading curve and the unloading curve). Ten nanoindentations were performed in different zones of each specimen, and the average values were evaluated. Four specimens belonging to the same ZrO2 + 10 wt% PCL series were used to assess the reproducibility of measurements, and preliminary results indicated that the within sample variations resulted larger than the variations between samples. Hence a mechanical testing was performed on two specimens per series, in different regions of the coating.
2.2. Coating procedure The hybrid ZrO2/PCL materials synthesized by sol-gel process, before gelation, when they were still in a sol phase, were used to coat titanium grade 4 (Sweden & Martina, Padua, Italy) disks of 1 cm diameter. Thin films were obtained by means of the dip-coating technique and a KSV LM dip coater was used. The substrates were ultrasonically cleaned with acetone and were subjected to the passivation process with HNO3 65%, for 60 min. After passivation, the disks were dipcoated in a ZrO2/PCL hybrid synthesized solution. The withdrawal
Table 1 Label and composition of the prepared coatings. Label
Composition
Zr(0) Zr(1) Zr(2) Zr(3) Zr(4)
ZrO2 ZrO2 ZrO2 ZrO2 ZrO2
+ + + +
PCL 5 wt% PCL 10 wt% PCL 20 wt% PCL 30 wt%
2.5. Corrosion tests The accelerated corrosion tests were performed in an electrochemical cell (Flat Cell K0235 PAR), using the samples as working electrode (exposing a flat and circular area of area of 1.6 cm2), a platinum grid as counter electrode and an Ag/AgCl/KCl(sat.) electrode (SSCE) as
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
377
Fig. 1. Flow chart of ZrO2/PCL gel synthesis.
reference electrode (all the potentials reported in this paper refer to this reference electrode). All the electrochemical tests were made in a GIBCO 1X DPBS stock solution (Dulbecco's Phosphate-Buffered Saline solution, pH 7.4) with the addition of 0.1 gL−1 of CaCl2, 0.1 gL−1 of MgCl2 · 6H2O and 0.05 gL−1 of BSA (bovine serum albumine) at 37 °C. Each sample was immersed in the test solution for 1 hour before applying any polarization, in order to stabilize its rest potential (Er); then potentiodynamic polarization scans were carried out with a scan rate of 0.5mVs− 1 in the range from –0.8 V (vs Er) to 3.0 V (vs Er) using a PAR VersaStat 3 potentiostat. The following polarization sequence was applied: (i) identification of the rest potential of the sample (Er) by measuring the open circuit potential of the cell; (ii) cathodic polarization from rest potential (Er) to a value of (Er – 0.8)V; (iii) anodic polarisation from (Er – 0.8)V to (Er + 3.0)V. In addition to the five coated samples, i.e. Zr(0)-(1)-(2)-(3)-(4), an uncoated titanium sample was tested to assess the corrosion behavior of the bare titanium; the uncoated sample was subjected to the same passivation process used as pretreatment for the coated samples (with HNO3 65% for 60 min). In order to better understand the behavior of the applied coating in terms of corrosion resistance, impedance spectra were acquired at different immersion times (100 kHz ÷ 1 Hz frequency range, AC amplitude 10 mV rms, without DC polarization, i.e. 0 V vs Er). The samples obtained at the end of polarization and immersion test (the last was interrupted after 100 h) were analyzed by SEM and EDX spectroscopy to evaluate the degradation of the coatings.
Fig. 2. PCL/ZrO2 bulk after drying at 45 °C for 24 h.
3. Results and discussion 3.1. Sol-gel synthesis Sol-gel is a very attractive technique used to synthesize hybrid materials with various compositions because of the low processing temperature (e.g., room temperature) that allows to entrap thermolabile molecules in an inorganic glassy matrix. Gelation is the result of hydrolysis and polycondensation of metal alkoxides according to the following reactions: ZrðOC3 H7 Þ4 þ nH2 O→ZrðOC3 H7 Þ4n ðOHÞn þ nC3 H7 OH
ð1Þ
≡Zr OH þ C3 H7 O Zr≡→≡Zr O Zr≡ þ C3 H7 OH
ð2Þ
≡Zr OH þ OH Zr≡→≡Zr O Zr≡ þ H2 O
ð3Þ
The reaction mechanism is generally accepted to proceed through a second order nucleophilic substitution [6,26]. The gelation time is between 8 and 18 days depending on the system. 3.2. Coating characterization The coating surfaces were examined using SEM and all layers appear transparent and homogeneous without any phases separation (micrographs shown in Fig. 3). Generally, thin coatings have a low susceptibility to thermal shock cracking and facilitate the ease of gas (including alcohol) removal. In addition, the thermal gradient within the coating is very small and the sintering conditions in all locations of the coating are similar. However, a severe cluster of cracks appears on the ZrO2 coating (Fig. 3A) due to the heat-treatment which removes volatile species even in the presence of a relatively small amount of the deposited material. The cracks are localized mainly towards the edges where they were thinner (observed as interference fringes) due to the edge effect. However, their number and size decrease as the PCL amount increases and disappear when 30 wt% of polymer is added (see figure from 3B to 3E). Therefore, the PCL confers elasticity to the coatings, according to the literature [9]. The chemical composition of the obtained coating was carried out using an infrared microscope which allowed to record ATR-FTIR spectra. Detailed FTIR spectroscopy study, reported in our previous works [21,23], have been shown that the two components (organic and
378
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
Fig. 3. SEM micrograph of A) Zr(0); B) Zr(1); C) Zr(2); D) Zr(3); E) Zr(4).
inorganic) of the hybrid materials interact through the formation of a hydrogen bonds between the carboxylic group of the PCL and the hydroxyl group of the inorganic matrix. Fig. 4 shows the ATR-FTIR spectra of ZrO2 (curve a) and PCL (curve f) compared with those of the hybrid coatings (curves from b to e). In the ZrO2 coating spectrum (curve a) typical peaks of AcAccontaining zirconia sol-gel materials [27,28] are present. The broad intense band in the region 3600–3000 cm−1 is due to –OH vibrations. The position and the shape of that band suggest the presence of hydrogen-bonded solvent molecules (H2O) and hydrogen-bonded OH groups attached to the Zr atom. The low-intensity sharp peaks at frequencies higher (about 3900–3700 cm−1), instead, are due to free \OH. The bands observed at 1552 and 1342 cm −1 are assigned to C = O vibrations of the AcAc bidentate binding. The band at 1529 and the weak shoulder at 1280 cm−1 are attributed to C\C vibrations in AcAc. The peak at 1410 cm−1 is due to methyl C\H symmetric bending [29], whereas the bands at 1026 and 931 cm−1 are assigned to the C\C\H bending mixed with stretching C\C vibrations of AcAc [27,30].
In ZrO2 + 5wt%PCL spectrum (curve b) all peaks of ZrO2 are visible and other peaks appear assignable to PCL presence [31,32], such as the shoulder at 1455 cm−1 due to methylene C\H bending and the weak peaks at 1290 cm− 1, 1050 cm− 1and 960 m− 1 ascribed to O\C stretching and bending and C\C stretching in the crystalline phase. The intensity of all those peaks increases in the spectra of the hybrid systems containing higher amounts of polymer (curve from c to e). Moreover, bands at 2933 and 2864 cm− 1, due to CH2 asymmetric and symmetric stretching of PCL respectively, appear together to the typical C_O peak at 1732 cm−1 and the weak band at 1182 cm−1 ascribed to asymmetric OC\O stretching. 3.3. Mechanical properties characterization The scratch performed are shown, one per each sample, in Fig. 5. The measured values of the critical load, indicated as LC1, and identified as the load at which the underlying substrate emerges to the surface, are shown in Fig. 6.
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
379
Fig. 6. Results of scratch tests: critical load (LC1).
Fig. 4. ATR-FTIR spectra of (a) ZrO2, and (b, c, d, e) ZrO2/PCL (6, 12, 24 and 50 wt%) coatings. (•) PCL peaks visible in the hybrids spectra.
The PCL-free coatings [Zr(0)] show the highest value of the critical load, despite a severe scattering of the results, ascribable to the rough surface finish of the Ti-4 substrates. The PCL containing coatings present a lower critical load, ascribable to a possible lack of homogeneity of the deposited hybrid coatings in presence of a high surface roughness. Such microstructure, together with a possible lower cohesion of the hybrid coatings, containing both high modulus ZrO2 and low modulus PCL, could be responsible for the measured decrease of the critical load. The different behaviour of the hybrid coatings compared to the PCLfree ones is also evident examining the damaged zone of the scratch: while PCL-free coatings tend to present damage by localized lack of adhesion and brittle scale formation, favoured by pre-existing cracks, as shown in Fig. 7, the PCL-containing ones show evidence of coating deformation along the scratch direction. This should be reflected into the elastic modulus of the coating, hence measurements have been performed to investigate such possible dependence. Fig. 8 shows the measured hardness, expressed in equivalent HV scale and the calculated Young's modulus of the coatings as a function of the PCL content.
Fig. 5. Scratch on ZrO2 + 0–5–10–20–30 wt% PCL samples.
Hardness values results slightly lower than those available in literature concerning grade 4 titanium. Rather unexpectedly, the PCL-free coating presents the lowest measured hardness. However, it must be taken into account that the PCL-free coating is severely cracked, while the PCL containing ones are crack-free. This is expected to affect the hardness measured through nanoindentation. The calculated elastic modulus, derived from the unloading curve, remains almost constant while varying the PCL content. The elastic modulus is close to the titanium one, and to a half of the bulk zirconia one. This seems to suggest that, despite the low loads applied, and hence the small indentation penetration depth, there is a non negligible substrate effect. For this reason, a second set of nanoindentations has been performed, imposing a maximum penetration depth of 100 nm and selecting damage-free regions. Under these conditions, the correct PCL-free coating elastic modulus could be measured and resulted 255 ± 33 GPa, as shown by the dotted line in Fig. 8. The PCL-containing samples, when subjected to lower load testing, presented elastic modulus values in agreement to the ones presented in Fig. 8, but with higher scattering. Hence, the PCL-free coatings, as expected, are stiffer and less deformable, for a given load. The energy required for deformation during nanoindentation has been measured as well. Fig. 9 shows the total amount of energy required for deformation and the percentage in the elastic (Def El) or plastic (Def Pl) field; the latter being interpreted as the amount of non recoverable deformation, as a function of PCL content. Results show that in all the investigated coatings containing less than 20 wt% PCL, as the PCL content increases, the amount of work in the plastic field decreases, to rapidly increase for the PCL-richer samples. Despite this, the amount of work in the elastic field remains
Fig. 7. PCL-free coating damaged at the critical load condition, resulting in the localized detachment of brittle scales.
380
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
Fig. 8. Young modulus and hardness of the coatings, determined by nanoindentation.
practically constant at 16% of the overall deformation energy, indicating that the elastic response of the samples, from an energetic point of view, is not affected by the PCL content. Also in this case, a possible substrate effect can be assumed, and nanoindentations performed with the maximum penetration depth of 100 nm confirmed this trend, except for the PCL-free samples, as shown in Fig. 10. Moreover, the amount of deformation energy in the elastic field remained below 10%, hence much lower than what measured with higher loads. In case of the PCL-free coating, such energy resulted practically 3.5% of the overall deformation energy, indicating the brittle nature of the zirconia coating, which undergoes fracture under loading and which therefore becomes unable to recover the imposed deformation when the load is removed. It can be concluded that the ZrO2 rich coatings generally present a higher scratch resistance and are stiffer, but with a stronger tendency to crack, also because of pre-existing damages in the coating. This can be an important issue when considering the corrosion resistance properties. On the contrary, the coatings containing the higher percentage of PCL are softer but can recover a slightly higher amount of deformation work. However, the application of higher loads makes the nanoindentation responses affected by the nature of the substrate, and the PCLricher samples behavior becomes comparable to the pure zirconiacoated ones. 3.4. Corrosion resistance evaluation The potentiodynamic polaritazion curves of bare and coated titanium are shown in Fig. 11. Uncoated titanium exhibits a corrosion potential of approximately − 0.3 V (SSCE), a passivation potential of 0.7 V (SSCE) and a very low passive current density (ranging between 4 2 and 10 μAcm−2), in agreement with the data reported in literature for corrosion test in this media [24,33]. The PCL free zirconia coating, Zr(0), shows a polarization curve similar, in shape, to that of bare titanium; there is a slight shift of the corrosion potential, but no significant
Fig. 9. Total amount of energy required for deformation and the percentage in elastic or plastic field.
Fig. 10. Total amount of energy required for deformation and the percentage in elastic or plastic field using low load nanoindentations.
change of the passivation potential: those results allow to assume that the zirconia coating does not modify the corrosion mechanism which takes place on the titanium. This hypothesis is in agreement with SEM microscopy results, as the coating of zirconia has numerous cracks from which the electrolyte may easily reach the titanium surface. All the four PCL/ZrO2 mixed coatings shows very similar polarization curves, both in term of shape and current values; in addition these samples show very little differences with the bare titanium behavior. This result can be justified considering that the PCL coating is moderately soluble in the electrolyte solution used [34], so that during the preliminary immersion of the samples in the test solution (1 hour) the coating may have allowed the entry of an amount of electrolyte sufficient to support the corrosion process: in this case the corrosion mechanism is ruled again by the passivation of the titanium substrate, bringing to results similar to those obtained on the bare titanium. The trend of the rest potentials (free corrosion potentials) acquired during the preliminary immersion of the samples in the test solution, shown in Fig. 12 for bare titanium and Zr(1) coating, gives a further confirm of the fact that the PCL-based coatings undergo a partial damage during this pretreatment. The graph of Fig. 12 shows that the potential of bare titanium is very stable in the test solution (stable passivation), while the potential of Zr(1) coated sample is affected by sharp oscillations ranging in the active corrosion region. In order to confirm the hypothesis regarding the partial dissolution of the PCL coatings in test solution, impedance spectra were acquired at increasing immersion time, as shown in Fig. 13 for Zr(0) and Zr(4) coatings (only the coating with the higher PCL content is shown,
Fig. 11. Potentiodynamic polarization curves for bare titanium and coated samples after 1 h of immersion in DPBS stock solution with the addition of 0.1 gL−1 of CaCl2, 0.1 gL−1 of MgCl2 · 6H2O and 0.05 gL−1 of BSA. Scan rate 0.5 mVs−1.
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
381
surface of the samples before and after the immersion and polarization tests. In both cases, the only difference between the samples before and after the test concerns the carbon content, which undergoes a dramatic drop, e.g. for the Zr(2) sample it decreases from 6% to 2% (in weight). 4. Conclusions
Fig. 12. Free corrosion potential (rest potential) acquired for bare titanium and Zr(1) coating during the immersion in the test solution.
to facilitate the reading of the graph). The spectra of Fig. 13 underline that just after 20 minutes of immersion the impedance of the coatings is the same of bare titanium. These spectra do not undergo further changes even after 100 hours of immersion, confirming the detection of the passivation phenomenon of the titanium substrate. It's interesting to note that the behavior, in terms of impedance after 20 minutes of immersion, of Zr(0) and Zr(4) samples is similar, although for the first this is due to the presence of numerous fractures on the coating, while for the second this is related to the solubility of the PCL coating in the test solution. The only difference between the two samples concerns the spectra acquired for immersion times below 20 minutes. While these spectra for the Zr(0) sample are identical to those shown in the graph of Fig. 13, for the Zr(4) sample (and all the other PCL coatings) was impossible to acquire these spectra because the sample appeared electrically isolated (very high and unstable impedance values). This result is a further confirmation about the very fast dissolution process of the PCL coating, that allow to acquire stable impedance spectra after 20 minutes of immersion, when the underlying titanium is almost completely exposed. Another experimental evidence that confirms the phenomenon of dissolution of the PCL coating is the EDX analysis performed on the
Dip coating has proved to be a promising technique for the deposition of functional coatings at a low cost to modify the surface properties of a substrate. The film properties change as function of the PCL amount. SEM micrographs have shown that for high PCL content the presence of cracks on coatings decreases. Moreover, scratch test and nanoindentation results show that, depending on the PCL amounts, both the hardness and the stiffness of such coatings can also be varied, obtaining higher values of elastic modulus in the case of a lower percentage of the polymer. Adhesion proved satisfactory, with critical loads in excess of 4 N for all samples, with minimum scratch resistance in case of the PCL-richer samples. The results obtained in corrosion tests, especially the polarization curves, allow to asses that the PCL-modified coatings do not significantly affect the passivation process that titanium develops in contact with the test solution used, because these coatings undergo a moderately fast degradation in such electrolyte. In conclusion, ZrO2/PCL crack-free coatings can be obtained by means of sol-gel dip coating methods, adding a high PCL content to the zirconia matrix. Those films can be used to coat metal implants to improve their bioactivity without significantly altering the already excellent passivation properties of titanium. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
[14] [15] [16] [17] [18] [19] [20] [21] [22]
Fig. 13. Impedance spectra acquired after different immersion times of the samples Zr(0) and Zr(4) in the test solution, compared with the spectrum of bare titanium.
[23] [24] [25] [26]
F.E.-T. Heakal, K.A. Awad, Int. J. Electrochem. Sci. 6 (2011) 6483–6502. Y.M. Wang, B.L. Jiang, L.X. Guo, T.Q. Lei, Appl. Surf. Sci. 252 (2006) 2989–2998. P.D. Miller, J.W. Holladay, Wear 2 (1958) 133–140. M. Masmoudi, M. Assoul, M. Wery, R. Abdelhedi, F. El Halouani, G. Monteil, Appl. Surf. Sci. 253 (2006) 2237–2243. M. Masmoudi, D. Capek, R. Abdelhedi, F. El Halouani, M. Wery, Surf. Coat. Technol. 200 (2006) 6651–6658. C. Brinker, G. Scherer, Sol-Gel Science: the Physics and Chemistry of Sol-Gel processing, Academic Press, San Diego, 1989. M. Guglielmi, J. Sol-Gel Sci. Technol. 8 (1997) 443–449. D. Wang, G.P. Bierwagen, Prog. Org. Coat. 64 (2009) 327–338. M. Catauro, F. Bollino, P. Veronesi, G. Lamanna, Mater. Sci. Eng. C 39 (2014) 344–351. D.W. Hutmacher, T. Schantz, I. Zein, K. Woei Ng, S. Hin Teoh, K. Cheng Tan, J. Biomed. Mater. Res. 55 (2001) 203–216. H.L. Khor, K.W. Ng, J.T. Schantz, T.-T. Phan, T.C. Lim, S.H. Teoh, D.W. Hutmacher, Mater. Sci. Eng. C 20 (2002) 71–75. R. De Santis, A. Gloria, T. Russo, U. D'Amora, V. D'Antò, F. Bollino, M. Catauro, F. Mollica, S. Rengo, L. Ambrosio, Polym. Compos. 34 (2013) 1413–1417. T. Russo, A. Gloria, V. D'Antò, U. D'Amora, G. Ametrano, F. Bollino, R. De Santis, G. Ausanio, M. Catauro, S. Rengo, L. Ambrosio, J. Appl. Biomater. Biomech. 8 (2010) 146–152. M. Catauro, F. Bollino, J. Appl. Biomater. Fundam. Mater. 11 (2013) 172–179. M. Catauro, F. Bollino, J. Appl. Biomater. Funct. Mater. (2014), http://dx.doi.org/10. 5301/JABFM.5000160. M. Catauro, F. Bollino, M. Cristina Mozzati, C. Ferrara, P. Mustarelli, J. Solid State Chem. 203 (2013) 92–99. M. Catauro, F. Bollino, F. Papale, J. Biomed. Mater. Res. A (2013), http://dx.doi.org/10. 1002/jbm.a.34836. M. Catauro, F. Papale, G. Roviello, C. Ferone, F. Bollino, M. Trifuoggi, C. Aurilio, J. Biomed. Mater. Res. A (2013), http://dx.doi.org/10.1002/jbm.a.34978. M. Catauro, M.G. Raucci, D. De Marco, L. Ambrosio, J. Biomed. Mater. Res. A 77 (2006) 340–350. R. De Santis, M. Catauro, L.D. Silvio, L. Manto, M.G. Raucci, L. Ambrosio, L. Nicolais, Biomaterials 28 (2007) 2801–2809. M. Catauro, D. Verardi, D. Melisi, F. Belotti, P. Mustarelli, J. Appl. Biomater. Biomech. 8 (2010) 42–51. M. Catauro, F. Bollino, F. Papale, S. Pacifico, S. Galasso, C. Ferrara, P. Mustarelli, Drug Deliv. (2013), http://dx.doi.org/10.3109/10717544.2013.865816. M. Catauro, F. Bollino, F. Papale, J. Biomed. Mater. Res. A 102 (2014) 392–399. X. Cheng, S.G. Roscoe, Biomaterials 26 (2005) 7350–7356. W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564–1583. C. Sanchez, J. Livage, M. Henry, F. Babonneau, J. Non-Cryst. Solids 100 (1988) 65–76.
382
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
[27] I. Georgieva, N. Danchova, S. Gutzov, N. Trendafilova, J. Mol. Model. 18 (2012) 2409–2422. [28] A. Gallo, A.M. Piccirillo, Boundary Value Problems, 20092009. [29] S.F. Tayyari, F. Milani-nejad, Spectrochim. Acta A Mol. Biomol. Spectrosc. 56 (2000) 2679–2691. [30] A. Gallo, A.M. Piccirillo, Nonlinear Stud. 19 (2012) 115–126.
[31] [32] [33] [34]
J. Krzaczkowska, Z. Fojud, M. Kozak, S. Jurga, Acta Phys. Pol. A 108 (2005) 187–196. K.P. Chellamani, R.S. Vignesh Balaji, J. Sudharsan, J. Acad. Ind. Res. 2 (2013) 341–344. S. Luiz de Assis, S. Wolynec, I. Costa, Electrochim. Acta 51 (2006) 1815–1819. J. Degner, F. Singer, L. Cordero, A.R. Boccaccini, S. Virtanen, Appl. Surf. Sci. 282 (2013) 264–270.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Corrosion behavior and mechanical properties of bioactive sol-gel coatings on titanium implants M. Catauro a,⁎, F. Bollino a, F. Papale a, R. Giovanardi b, P. Veronesi b a b
Department of Industrial and Information Engineering, Second University of Naples, Via Roma 29, 81031 Aversa, Italy Department of Engineering “Enzo Ferrari”, University of Modena and Reggio Emilia, Via Vignolese 905, 41125 Modena, Italy
a r t i c l e
i n f o
Article history: Received 11 April 2014 Received in revised form 30 June 2014 Accepted 13 July 2014 Available online 18 July 2014 Keywords: Sol-gel process Organic–inorganic hybrid Scratch and nano-indentation tests Corrosion process Polarization tests
a b s t r a c t Organic–inorganic hybrid coatings based on zirconia and poly (ε-caprolactone) (PCL) were prepared by means of sol-gel dip-coating technique and used to coat titanium grade 4 implants (Ti-4) in order to improve their wear and corrosion resistance. The coating chemical composition has been analysed by ATR-FTIR. The influence of the PCL amount has been investigated on the microstructure, mechanical properties of the coatings and their ability to inhibit the corrosion of titanium. SEM analysis has shown that all coatings have a nanostructured nature and that the films with high PCL content are crack-free. Mechanical properties of the coatings have been studied using scratch and nano-indentation tests. The results have shown that the Young's modulus of the coatings decreases in presence of large amounts of the organic phase, and that PCL content affects also the adhesion of the coatings to the underlying Ti-4 substrate. However, the presence of cracks on the PCL-free coatings affects severely the mechanical response of the samples at high loads. The electrochemical behavior and corrosion resistance of the coated and uncoated substrate has been investigated by polarization tests. The results have shown that both the coatings with or without PCL don't affect significantly the already excellent passivation properties of titanium. © 2014 Published by Elsevier B.V.
1. Introduction Titanium and its alloys are attractive metallic materials widely used as implants for orthopedic dental and orthodontic wires due to their excellent biocompatibility and corrosion resistance even in very aggressive environments. This aspect is related to their ability to form spontaneously a dense and stable titanium dioxide layer on the surface which provides a barrier between the bio-environment and implant [1]. However, the poor tribological behaviour caused by a high and unstable friction coefficient, the poor wear resistance and the strong tendency to seize [2,3] have restricted applications of titanium as a biomedical implant material. Indeed, when local mechanical abrasion removes the protective oxide film, the corrosion resistance of titanium alloys can be strongly decreased [4]. To avoid the early failure of the implants due to wear and corrosion and to extend the prostheses lifetime, different strategies can be followed, such as to increase the thickness of oxide films using nitric acid passivation protocols [4,5] or to apply a protective coating, which also permits to improve the substrate performances related to its surface, as bioactivity and biocompatibilty.
⁎ Corresponding author. Tel.: +39 081 50 10 360; fax: +39 081 50 10 204. E-mail address: [email protected] (M. Catauro).
http://dx.doi.org/10.1016/j.msec.2014.07.044 0928-4931/© 2014 Published by Elsevier B.V.
The sol-gel dip-coating process is a promising technique for the deposition of functional coatings at low cost on a wide range of substrates. Sol-gel technology is the process of making ceramic and glass materials at a relatively low temperature which allows the entrapping of various inorganic, organic substances and biomolecules in a glassy matrix during its formation [6]. Ten years ago, Guglielmi [7] has already discussed the potential of sol-gel coatings as a corrosion inhibiting system for metal substrates. Since then, various sol-gel based protective coatings were developed. In this work, organic–inorganic hybrid materials based on zirconia (ZrO2) and poly (ɛ-caprolactone) (PCL) were synthesized by sol-gel method and used, in sol phase, to dip-coat commercially pure titanium grade 4 (Ti-4) implants in order to improve their biological properties and to decrease wear and corrosion. ZrO2 has a high expansion coefficient very close to many bulk metals, which can reduce the formation of cracks during the high temperature curing process. ZrO2 also shows good chemical stability and high hardness which makes it a good protective material [8]. However, inorganic sol-gel coatings have many limitations, as brittleness and high temperature treatment. Many works report the introduction of an organic component into the inorganic sol-gel to form an organic–inorganic hybrid sol-gel coating to overcome these limitations [8,9]. For that purpose, the authors have used the poly (ε-caprolactone) a biodegradable,
376
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
synthetic, aliphatic polyester that has attracted a wide interest for its possible applications as a biomaterial: PCL-based three-dimensional scaffolds for orthopaedic surgery [10], and film substrates for tissue engineering [11] have been proposed. Moreover, composite substrates for tissue engineering have been prepared by sol-gel method, consisting of a poly (ε-caprolactone) matrix reinforced with sol-gel synthesized PCL/TiO2 or PCL/ZrO2 hybrid fillers [12,13] as well as several PCL-based hybrids for drug delivery, using different oxides, CaO and/or SiO2 [14–18], TiO2 [19,20], and ZrO2 [21,22]. Moreover, elsewhere the authors proved that such polymer is able to improve the mechanical properties of TiO2-based glassy matrix [20] and acts as plasticizer in the coatings preparations [9]. The Authors [23] already proved that a ZrO2/PCL hybrid coating can be used to modify the surface of Ti-4 implants by improving their biological properties (bioactivity and biocompatibility). The aim of this research is also to investigate the adhesion and the mechanical properties of the synthesized thin films and the corrosion resistance of the Ti-4 implants after coating with the prepared hybrid, when exposed to aggressive human physiological fluids and calcium phosphate surface deposits by means of electrochemical polarization tests, as reported in literature [24] 2. Materials and methods
speed of the substrate was 15 cm/min. The coated substrates were heat-treated at 45 °C for 1 day to promote a partial densification of the film without any polymer degradation. The obtained layers appear to be transparent, uniform and crackfree for Zr (1)-(2)-(3)-(4), while cracks appear on the surface of Zr(0). That observation was confirmed by a microstructural analysis performed using SEM (Quanta 200, FEI Europe Company, Netherlands). 2.3. Materials and coating characterization Scanning electron microscopy (SEM) and Attenuated total reflectance—Fourier transform infrared (ATR-FTIR) spectroscopy were employed to characterize obtained coatings. The microstructure of the films was investigated by a SEM FEI Quanta 200 equipped with EDX (energy dispersive X-ray spectroscopy). ATR/FT-IR allows to analyze the chemical composition of the coating surface. The spectra were recorded on a Prestige-21 FTIR spectrometer equipped with an AIM-8800 infrared microscope (Shimazu, Japan), using the incorporated 3-mm diameter Ge ATR semicircular prism. The spectra were recorded using an incident angle of 30° with the sum of 64 scans at a resolution of 4 cm−1 and in the 650–4000 cm−1. The spectra were elaborated by Prestige software (IRsolution).
2.1. Sol-gel synthesis
2.4. Mechanical characterization techniques
Hybrid organic–inorganic composites, consisting of an inorganic ZrO2 matrix and PCL as organic phase were synthesized by means of the sol-gel process. In order to study the influence of the PCL amount on corrosion behavior and mechanical properties of the coatings several films containing different percentage of the polymer was prepared, as shown in Table 1. Zirconium propoxide Zr(OC3H7)4 and poly (ε-caprolactone) (PCL Mw = 65000) were respectively used as inorganic and organic precursors. The solution of poly (ε-caprolactone) in chloroform was added to the solution of Zr(OC3H7)4 in ethanol–acetylaceton-water mixture. Acetylacetone was added to control the hydrolytic activity of zirconium alkoxide. After the addition of each reactant, the solution was stirred with a magnetic stirrer and the resulting sols were uniform and homogeneous. The time of gelation was 8–18 days depending on the system. Fig. 1 shows the flow-chart of the hybrid synthesis by the sol gel method. After gelation the gels were air-dried at 50 °C for 24 h to remove the residual solvents. That treatment allows to obtain glassy pieces of various sizes (see Fig. 2) without modifying the stability of the polymer, as its melting temperature is about 65 °C.
Scratch tests were carried out on a CSM Micro-Combi tester. In those tests, a controlled scratch on the coating surface was made with a diamond tip (Rockwell C diamond scratch indenter with tip radius of 200 μm). The tip was drawn across the coated surface of the sample under an increasing load (from 0.1 to 20 N) at a load rate of 4.98 N min−1 along a scratch length of 1 mm. The instrument was equipped with an integrated optical microscope, an acoustic emission detection system and a device to measure the tangential frictional force in the scratch direction. The sensors allowed to determine the critical load (LC1), evaluated as the normal load at which the first damage occurred to the coating. Three scratches were carried out in different zones for each specimen and the average value of the critical load was evaluated. Nanoindentation was performed using a Berkovitch indenter (Nanoindenter, CSM Instruments, Peseux, CH) operating at a constant load of 50 mN (loading and unloading rate = 360 mN/min), applied for 15 s, in order to maintain the indentation depth below 1 μm. A second set of nanoindentations was performed in penetration depth control, imposing a maximum penetration depth of the indenter equal to 100 nm. Nanoindentation results were analyzed according to the Oliver and Pharr procedure [25] in such a way as to determine the coating hardness and elastic module. The same penetration depth vs. load curves were also used to evaluate the amount of work required to cause the nanoindentation, and the percentage of work performed in the elastic regime (integral of the unloading curve) or plastic regime (difference between the integral of the loading curve and the unloading curve). Ten nanoindentations were performed in different zones of each specimen, and the average values were evaluated. Four specimens belonging to the same ZrO2 + 10 wt% PCL series were used to assess the reproducibility of measurements, and preliminary results indicated that the within sample variations resulted larger than the variations between samples. Hence a mechanical testing was performed on two specimens per series, in different regions of the coating.
2.2. Coating procedure The hybrid ZrO2/PCL materials synthesized by sol-gel process, before gelation, when they were still in a sol phase, were used to coat titanium grade 4 (Sweden & Martina, Padua, Italy) disks of 1 cm diameter. Thin films were obtained by means of the dip-coating technique and a KSV LM dip coater was used. The substrates were ultrasonically cleaned with acetone and were subjected to the passivation process with HNO3 65%, for 60 min. After passivation, the disks were dipcoated in a ZrO2/PCL hybrid synthesized solution. The withdrawal
Table 1 Label and composition of the prepared coatings. Label
Composition
Zr(0) Zr(1) Zr(2) Zr(3) Zr(4)
ZrO2 ZrO2 ZrO2 ZrO2 ZrO2
+ + + +
PCL 5 wt% PCL 10 wt% PCL 20 wt% PCL 30 wt%
2.5. Corrosion tests The accelerated corrosion tests were performed in an electrochemical cell (Flat Cell K0235 PAR), using the samples as working electrode (exposing a flat and circular area of area of 1.6 cm2), a platinum grid as counter electrode and an Ag/AgCl/KCl(sat.) electrode (SSCE) as
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
377
Fig. 1. Flow chart of ZrO2/PCL gel synthesis.
reference electrode (all the potentials reported in this paper refer to this reference electrode). All the electrochemical tests were made in a GIBCO 1X DPBS stock solution (Dulbecco's Phosphate-Buffered Saline solution, pH 7.4) with the addition of 0.1 gL−1 of CaCl2, 0.1 gL−1 of MgCl2 · 6H2O and 0.05 gL−1 of BSA (bovine serum albumine) at 37 °C. Each sample was immersed in the test solution for 1 hour before applying any polarization, in order to stabilize its rest potential (Er); then potentiodynamic polarization scans were carried out with a scan rate of 0.5mVs− 1 in the range from –0.8 V (vs Er) to 3.0 V (vs Er) using a PAR VersaStat 3 potentiostat. The following polarization sequence was applied: (i) identification of the rest potential of the sample (Er) by measuring the open circuit potential of the cell; (ii) cathodic polarization from rest potential (Er) to a value of (Er – 0.8)V; (iii) anodic polarisation from (Er – 0.8)V to (Er + 3.0)V. In addition to the five coated samples, i.e. Zr(0)-(1)-(2)-(3)-(4), an uncoated titanium sample was tested to assess the corrosion behavior of the bare titanium; the uncoated sample was subjected to the same passivation process used as pretreatment for the coated samples (with HNO3 65% for 60 min). In order to better understand the behavior of the applied coating in terms of corrosion resistance, impedance spectra were acquired at different immersion times (100 kHz ÷ 1 Hz frequency range, AC amplitude 10 mV rms, without DC polarization, i.e. 0 V vs Er). The samples obtained at the end of polarization and immersion test (the last was interrupted after 100 h) were analyzed by SEM and EDX spectroscopy to evaluate the degradation of the coatings.
Fig. 2. PCL/ZrO2 bulk after drying at 45 °C for 24 h.
3. Results and discussion 3.1. Sol-gel synthesis Sol-gel is a very attractive technique used to synthesize hybrid materials with various compositions because of the low processing temperature (e.g., room temperature) that allows to entrap thermolabile molecules in an inorganic glassy matrix. Gelation is the result of hydrolysis and polycondensation of metal alkoxides according to the following reactions: ZrðOC3 H7 Þ4 þ nH2 O→ZrðOC3 H7 Þ4n ðOHÞn þ nC3 H7 OH
ð1Þ
≡Zr OH þ C3 H7 O Zr≡→≡Zr O Zr≡ þ C3 H7 OH
ð2Þ
≡Zr OH þ OH Zr≡→≡Zr O Zr≡ þ H2 O
ð3Þ
The reaction mechanism is generally accepted to proceed through a second order nucleophilic substitution [6,26]. The gelation time is between 8 and 18 days depending on the system. 3.2. Coating characterization The coating surfaces were examined using SEM and all layers appear transparent and homogeneous without any phases separation (micrographs shown in Fig. 3). Generally, thin coatings have a low susceptibility to thermal shock cracking and facilitate the ease of gas (including alcohol) removal. In addition, the thermal gradient within the coating is very small and the sintering conditions in all locations of the coating are similar. However, a severe cluster of cracks appears on the ZrO2 coating (Fig. 3A) due to the heat-treatment which removes volatile species even in the presence of a relatively small amount of the deposited material. The cracks are localized mainly towards the edges where they were thinner (observed as interference fringes) due to the edge effect. However, their number and size decrease as the PCL amount increases and disappear when 30 wt% of polymer is added (see figure from 3B to 3E). Therefore, the PCL confers elasticity to the coatings, according to the literature [9]. The chemical composition of the obtained coating was carried out using an infrared microscope which allowed to record ATR-FTIR spectra. Detailed FTIR spectroscopy study, reported in our previous works [21,23], have been shown that the two components (organic and
378
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
Fig. 3. SEM micrograph of A) Zr(0); B) Zr(1); C) Zr(2); D) Zr(3); E) Zr(4).
inorganic) of the hybrid materials interact through the formation of a hydrogen bonds between the carboxylic group of the PCL and the hydroxyl group of the inorganic matrix. Fig. 4 shows the ATR-FTIR spectra of ZrO2 (curve a) and PCL (curve f) compared with those of the hybrid coatings (curves from b to e). In the ZrO2 coating spectrum (curve a) typical peaks of AcAccontaining zirconia sol-gel materials [27,28] are present. The broad intense band in the region 3600–3000 cm−1 is due to –OH vibrations. The position and the shape of that band suggest the presence of hydrogen-bonded solvent molecules (H2O) and hydrogen-bonded OH groups attached to the Zr atom. The low-intensity sharp peaks at frequencies higher (about 3900–3700 cm−1), instead, are due to free \OH. The bands observed at 1552 and 1342 cm −1 are assigned to C = O vibrations of the AcAc bidentate binding. The band at 1529 and the weak shoulder at 1280 cm−1 are attributed to C\C vibrations in AcAc. The peak at 1410 cm−1 is due to methyl C\H symmetric bending [29], whereas the bands at 1026 and 931 cm−1 are assigned to the C\C\H bending mixed with stretching C\C vibrations of AcAc [27,30].
In ZrO2 + 5wt%PCL spectrum (curve b) all peaks of ZrO2 are visible and other peaks appear assignable to PCL presence [31,32], such as the shoulder at 1455 cm−1 due to methylene C\H bending and the weak peaks at 1290 cm− 1, 1050 cm− 1and 960 m− 1 ascribed to O\C stretching and bending and C\C stretching in the crystalline phase. The intensity of all those peaks increases in the spectra of the hybrid systems containing higher amounts of polymer (curve from c to e). Moreover, bands at 2933 and 2864 cm− 1, due to CH2 asymmetric and symmetric stretching of PCL respectively, appear together to the typical C_O peak at 1732 cm−1 and the weak band at 1182 cm−1 ascribed to asymmetric OC\O stretching. 3.3. Mechanical properties characterization The scratch performed are shown, one per each sample, in Fig. 5. The measured values of the critical load, indicated as LC1, and identified as the load at which the underlying substrate emerges to the surface, are shown in Fig. 6.
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
379
Fig. 6. Results of scratch tests: critical load (LC1).
Fig. 4. ATR-FTIR spectra of (a) ZrO2, and (b, c, d, e) ZrO2/PCL (6, 12, 24 and 50 wt%) coatings. (•) PCL peaks visible in the hybrids spectra.
The PCL-free coatings [Zr(0)] show the highest value of the critical load, despite a severe scattering of the results, ascribable to the rough surface finish of the Ti-4 substrates. The PCL containing coatings present a lower critical load, ascribable to a possible lack of homogeneity of the deposited hybrid coatings in presence of a high surface roughness. Such microstructure, together with a possible lower cohesion of the hybrid coatings, containing both high modulus ZrO2 and low modulus PCL, could be responsible for the measured decrease of the critical load. The different behaviour of the hybrid coatings compared to the PCLfree ones is also evident examining the damaged zone of the scratch: while PCL-free coatings tend to present damage by localized lack of adhesion and brittle scale formation, favoured by pre-existing cracks, as shown in Fig. 7, the PCL-containing ones show evidence of coating deformation along the scratch direction. This should be reflected into the elastic modulus of the coating, hence measurements have been performed to investigate such possible dependence. Fig. 8 shows the measured hardness, expressed in equivalent HV scale and the calculated Young's modulus of the coatings as a function of the PCL content.
Fig. 5. Scratch on ZrO2 + 0–5–10–20–30 wt% PCL samples.
Hardness values results slightly lower than those available in literature concerning grade 4 titanium. Rather unexpectedly, the PCL-free coating presents the lowest measured hardness. However, it must be taken into account that the PCL-free coating is severely cracked, while the PCL containing ones are crack-free. This is expected to affect the hardness measured through nanoindentation. The calculated elastic modulus, derived from the unloading curve, remains almost constant while varying the PCL content. The elastic modulus is close to the titanium one, and to a half of the bulk zirconia one. This seems to suggest that, despite the low loads applied, and hence the small indentation penetration depth, there is a non negligible substrate effect. For this reason, a second set of nanoindentations has been performed, imposing a maximum penetration depth of 100 nm and selecting damage-free regions. Under these conditions, the correct PCL-free coating elastic modulus could be measured and resulted 255 ± 33 GPa, as shown by the dotted line in Fig. 8. The PCL-containing samples, when subjected to lower load testing, presented elastic modulus values in agreement to the ones presented in Fig. 8, but with higher scattering. Hence, the PCL-free coatings, as expected, are stiffer and less deformable, for a given load. The energy required for deformation during nanoindentation has been measured as well. Fig. 9 shows the total amount of energy required for deformation and the percentage in the elastic (Def El) or plastic (Def Pl) field; the latter being interpreted as the amount of non recoverable deformation, as a function of PCL content. Results show that in all the investigated coatings containing less than 20 wt% PCL, as the PCL content increases, the amount of work in the plastic field decreases, to rapidly increase for the PCL-richer samples. Despite this, the amount of work in the elastic field remains
Fig. 7. PCL-free coating damaged at the critical load condition, resulting in the localized detachment of brittle scales.
380
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
Fig. 8. Young modulus and hardness of the coatings, determined by nanoindentation.
practically constant at 16% of the overall deformation energy, indicating that the elastic response of the samples, from an energetic point of view, is not affected by the PCL content. Also in this case, a possible substrate effect can be assumed, and nanoindentations performed with the maximum penetration depth of 100 nm confirmed this trend, except for the PCL-free samples, as shown in Fig. 10. Moreover, the amount of deformation energy in the elastic field remained below 10%, hence much lower than what measured with higher loads. In case of the PCL-free coating, such energy resulted practically 3.5% of the overall deformation energy, indicating the brittle nature of the zirconia coating, which undergoes fracture under loading and which therefore becomes unable to recover the imposed deformation when the load is removed. It can be concluded that the ZrO2 rich coatings generally present a higher scratch resistance and are stiffer, but with a stronger tendency to crack, also because of pre-existing damages in the coating. This can be an important issue when considering the corrosion resistance properties. On the contrary, the coatings containing the higher percentage of PCL are softer but can recover a slightly higher amount of deformation work. However, the application of higher loads makes the nanoindentation responses affected by the nature of the substrate, and the PCLricher samples behavior becomes comparable to the pure zirconiacoated ones. 3.4. Corrosion resistance evaluation The potentiodynamic polaritazion curves of bare and coated titanium are shown in Fig. 11. Uncoated titanium exhibits a corrosion potential of approximately − 0.3 V (SSCE), a passivation potential of 0.7 V (SSCE) and a very low passive current density (ranging between 4 2 and 10 μAcm−2), in agreement with the data reported in literature for corrosion test in this media [24,33]. The PCL free zirconia coating, Zr(0), shows a polarization curve similar, in shape, to that of bare titanium; there is a slight shift of the corrosion potential, but no significant
Fig. 9. Total amount of energy required for deformation and the percentage in elastic or plastic field.
Fig. 10. Total amount of energy required for deformation and the percentage in elastic or plastic field using low load nanoindentations.
change of the passivation potential: those results allow to assume that the zirconia coating does not modify the corrosion mechanism which takes place on the titanium. This hypothesis is in agreement with SEM microscopy results, as the coating of zirconia has numerous cracks from which the electrolyte may easily reach the titanium surface. All the four PCL/ZrO2 mixed coatings shows very similar polarization curves, both in term of shape and current values; in addition these samples show very little differences with the bare titanium behavior. This result can be justified considering that the PCL coating is moderately soluble in the electrolyte solution used [34], so that during the preliminary immersion of the samples in the test solution (1 hour) the coating may have allowed the entry of an amount of electrolyte sufficient to support the corrosion process: in this case the corrosion mechanism is ruled again by the passivation of the titanium substrate, bringing to results similar to those obtained on the bare titanium. The trend of the rest potentials (free corrosion potentials) acquired during the preliminary immersion of the samples in the test solution, shown in Fig. 12 for bare titanium and Zr(1) coating, gives a further confirm of the fact that the PCL-based coatings undergo a partial damage during this pretreatment. The graph of Fig. 12 shows that the potential of bare titanium is very stable in the test solution (stable passivation), while the potential of Zr(1) coated sample is affected by sharp oscillations ranging in the active corrosion region. In order to confirm the hypothesis regarding the partial dissolution of the PCL coatings in test solution, impedance spectra were acquired at increasing immersion time, as shown in Fig. 13 for Zr(0) and Zr(4) coatings (only the coating with the higher PCL content is shown,
Fig. 11. Potentiodynamic polarization curves for bare titanium and coated samples after 1 h of immersion in DPBS stock solution with the addition of 0.1 gL−1 of CaCl2, 0.1 gL−1 of MgCl2 · 6H2O and 0.05 gL−1 of BSA. Scan rate 0.5 mVs−1.
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
381
surface of the samples before and after the immersion and polarization tests. In both cases, the only difference between the samples before and after the test concerns the carbon content, which undergoes a dramatic drop, e.g. for the Zr(2) sample it decreases from 6% to 2% (in weight). 4. Conclusions
Fig. 12. Free corrosion potential (rest potential) acquired for bare titanium and Zr(1) coating during the immersion in the test solution.
to facilitate the reading of the graph). The spectra of Fig. 13 underline that just after 20 minutes of immersion the impedance of the coatings is the same of bare titanium. These spectra do not undergo further changes even after 100 hours of immersion, confirming the detection of the passivation phenomenon of the titanium substrate. It's interesting to note that the behavior, in terms of impedance after 20 minutes of immersion, of Zr(0) and Zr(4) samples is similar, although for the first this is due to the presence of numerous fractures on the coating, while for the second this is related to the solubility of the PCL coating in the test solution. The only difference between the two samples concerns the spectra acquired for immersion times below 20 minutes. While these spectra for the Zr(0) sample are identical to those shown in the graph of Fig. 13, for the Zr(4) sample (and all the other PCL coatings) was impossible to acquire these spectra because the sample appeared electrically isolated (very high and unstable impedance values). This result is a further confirmation about the very fast dissolution process of the PCL coating, that allow to acquire stable impedance spectra after 20 minutes of immersion, when the underlying titanium is almost completely exposed. Another experimental evidence that confirms the phenomenon of dissolution of the PCL coating is the EDX analysis performed on the
Dip coating has proved to be a promising technique for the deposition of functional coatings at a low cost to modify the surface properties of a substrate. The film properties change as function of the PCL amount. SEM micrographs have shown that for high PCL content the presence of cracks on coatings decreases. Moreover, scratch test and nanoindentation results show that, depending on the PCL amounts, both the hardness and the stiffness of such coatings can also be varied, obtaining higher values of elastic modulus in the case of a lower percentage of the polymer. Adhesion proved satisfactory, with critical loads in excess of 4 N for all samples, with minimum scratch resistance in case of the PCL-richer samples. The results obtained in corrosion tests, especially the polarization curves, allow to asses that the PCL-modified coatings do not significantly affect the passivation process that titanium develops in contact with the test solution used, because these coatings undergo a moderately fast degradation in such electrolyte. In conclusion, ZrO2/PCL crack-free coatings can be obtained by means of sol-gel dip coating methods, adding a high PCL content to the zirconia matrix. Those films can be used to coat metal implants to improve their bioactivity without significantly altering the already excellent passivation properties of titanium. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
[14] [15] [16] [17] [18] [19] [20] [21] [22]
Fig. 13. Impedance spectra acquired after different immersion times of the samples Zr(0) and Zr(4) in the test solution, compared with the spectrum of bare titanium.
[23] [24] [25] [26]
F.E.-T. Heakal, K.A. Awad, Int. J. Electrochem. Sci. 6 (2011) 6483–6502. Y.M. Wang, B.L. Jiang, L.X. Guo, T.Q. Lei, Appl. Surf. Sci. 252 (2006) 2989–2998. P.D. Miller, J.W. Holladay, Wear 2 (1958) 133–140. M. Masmoudi, M. Assoul, M. Wery, R. Abdelhedi, F. El Halouani, G. Monteil, Appl. Surf. Sci. 253 (2006) 2237–2243. M. Masmoudi, D. Capek, R. Abdelhedi, F. El Halouani, M. Wery, Surf. Coat. Technol. 200 (2006) 6651–6658. C. Brinker, G. Scherer, Sol-Gel Science: the Physics and Chemistry of Sol-Gel processing, Academic Press, San Diego, 1989. M. Guglielmi, J. Sol-Gel Sci. Technol. 8 (1997) 443–449. D. Wang, G.P. Bierwagen, Prog. Org. Coat. 64 (2009) 327–338. M. Catauro, F. Bollino, P. Veronesi, G. Lamanna, Mater. Sci. Eng. C 39 (2014) 344–351. D.W. Hutmacher, T. Schantz, I. Zein, K. Woei Ng, S. Hin Teoh, K. Cheng Tan, J. Biomed. Mater. Res. 55 (2001) 203–216. H.L. Khor, K.W. Ng, J.T. Schantz, T.-T. Phan, T.C. Lim, S.H. Teoh, D.W. Hutmacher, Mater. Sci. Eng. C 20 (2002) 71–75. R. De Santis, A. Gloria, T. Russo, U. D'Amora, V. D'Antò, F. Bollino, M. Catauro, F. Mollica, S. Rengo, L. Ambrosio, Polym. Compos. 34 (2013) 1413–1417. T. Russo, A. Gloria, V. D'Antò, U. D'Amora, G. Ametrano, F. Bollino, R. De Santis, G. Ausanio, M. Catauro, S. Rengo, L. Ambrosio, J. Appl. Biomater. Biomech. 8 (2010) 146–152. M. Catauro, F. Bollino, J. Appl. Biomater. Fundam. Mater. 11 (2013) 172–179. M. Catauro, F. Bollino, J. Appl. Biomater. Funct. Mater. (2014), http://dx.doi.org/10. 5301/JABFM.5000160. M. Catauro, F. Bollino, M. Cristina Mozzati, C. Ferrara, P. Mustarelli, J. Solid State Chem. 203 (2013) 92–99. M. Catauro, F. Bollino, F. Papale, J. Biomed. Mater. Res. A (2013), http://dx.doi.org/10. 1002/jbm.a.34836. M. Catauro, F. Papale, G. Roviello, C. Ferone, F. Bollino, M. Trifuoggi, C. Aurilio, J. Biomed. Mater. Res. A (2013), http://dx.doi.org/10.1002/jbm.a.34978. M. Catauro, M.G. Raucci, D. De Marco, L. Ambrosio, J. Biomed. Mater. Res. A 77 (2006) 340–350. R. De Santis, M. Catauro, L.D. Silvio, L. Manto, M.G. Raucci, L. Ambrosio, L. Nicolais, Biomaterials 28 (2007) 2801–2809. M. Catauro, D. Verardi, D. Melisi, F. Belotti, P. Mustarelli, J. Appl. Biomater. Biomech. 8 (2010) 42–51. M. Catauro, F. Bollino, F. Papale, S. Pacifico, S. Galasso, C. Ferrara, P. Mustarelli, Drug Deliv. (2013), http://dx.doi.org/10.3109/10717544.2013.865816. M. Catauro, F. Bollino, F. Papale, J. Biomed. Mater. Res. A 102 (2014) 392–399. X. Cheng, S.G. Roscoe, Biomaterials 26 (2005) 7350–7356. W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564–1583. C. Sanchez, J. Livage, M. Henry, F. Babonneau, J. Non-Cryst. Solids 100 (1988) 65–76.
382
M. Catauro et al. / Materials Science and Engineering C 43 (2014) 375–382
[27] I. Georgieva, N. Danchova, S. Gutzov, N. Trendafilova, J. Mol. Model. 18 (2012) 2409–2422. [28] A. Gallo, A.M. Piccirillo, Boundary Value Problems, 20092009. [29] S.F. Tayyari, F. Milani-nejad, Spectrochim. Acta A Mol. Biomol. Spectrosc. 56 (2000) 2679–2691. [30] A. Gallo, A.M. Piccirillo, Nonlinear Stud. 19 (2012) 115–126.
[31] [32] [33] [34]
J. Krzaczkowska, Z. Fojud, M. Kozak, S. Jurga, Acta Phys. Pol. A 108 (2005) 187–196. K.P. Chellamani, R.S. Vignesh Balaji, J. Sudharsan, J. Acad. Ind. Res. 2 (2013) 341–344. S. Luiz de Assis, S. Wolynec, I. Costa, Electrochim. Acta 51 (2006) 1815–1819. J. Degner, F. Singer, L. Cordero, A.R. Boccaccini, S. Virtanen, Appl. Surf. Sci. 282 (2013) 264–270.
