Electrochemically assisted deposition of hydroxyapatite on Ti6Al4V substrates covered by CVD diamond films — Coating characterization and first cell biological results

Materials Science and Engineering C 59 (2016) 624–635

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

Electrochemically assisted deposition of hydroxyapatite on Ti6Al4V substrates covered by CVD diamond films — Coating characterization and first cell biological results Paulina Strąkowska a,c,1, René Beutner b,1, Marcin Gnyba c, Andrzej Zielinski a,2, Dieter Scharnweber b,⁎,2 a b c

Gdańsk University of Technology, Mechanical Engineering Faculty, Poland Max Bergmann Center, Technische Universität Dresden, Germany Gdańsk University of Technology, Faculty of Electronics, Telecommunications, and Informatics, Poland

a r t i c l e

i n f o

Article history: Received 23 June 2015 Received in revised form 23 September 2015 Accepted 20 October 2015 Available online 22 October 2015 Keywords: Ti alloys Diamond films Hydroxyapatite coatings CVD deposition ECAD deposition Biological activity

a b s t r a c t Although titanium and its alloys are widely used as implant material for orthopedic and dental applications they show only limited corrosion stability and osseointegration in different cases. The aim of the presented research was to develop and characterize a novel surface modification system from a thin diamond base layer and a hydroxyapatite (HAp) top coating deposited on the alloy Ti6Al4V widely used for implants in contact with bone. This coating system is expected to improve both the long-term corrosion behavior and the biocompatibility and bioactivity of respective surfaces. The diamond base films were obtained by Microwave Plasma Assisted Chemical Vapor Deposition (MW-PACVD); the HAp coatings were formed in aqueous solutions by electrochemically assisted deposition (ECAD) at varying polarization parameters. Scanning electron microscopy (SEM), Raman microscopy, and electrical conductivity measurements were applied to characterize the generated surface states; the calcium phosphate coatings were additionally chemically analyzed for their composition. The biological properties of the coating system were assessed using hMSC cells analyzing for cell adhesion, proliferation, and osteogenic differentiation. Varying MW-PACVD process conditions resulted in composite coatings containing microcrystalline diamond (MCD/Ti-C), nanocrystalline diamond (NCD), and boron-doped nanocrystalline diamond (B-NCD) with the NCD coatings being dense and homogeneous and the B-NCD coatings showing increased electrical conductivity. The ECAD process resulted in calcium phosphate coatings from stoichiometric and non-stoichiometric HAp. The deposition of HAp on the B-NCD films run at lower cathodic potentials and resulted both in the highest coating mass and the most homogenous appearance. Initial cell biological investigations showed an improved cell adhesion in the order B-NCD N HAp/B-NCD N uncoated substrate. Cell proliferation was improved for both investigated coatings whereas ALP expression was highest for the uncoated substrate. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Titanium and its alloys have been extensively used as materials for implants in contact with bone throughout recent decades because of their high biocompatibility and suitable mechanical properties. Additionally, their surface is always covered by a stable, semi-conducting passive layer, which in many cases brings a high corrosion resistance and bioinert in vivo behavior to the material [1,2]. In general this behavior results in a good osseointegration, especially for otherwise healthy patients. Nevertheless, an early implant failure and problems during

⁎ Corresponding author at: Technische Universität Dresden, Max Bergmann Center of Biomaterials, Budapester Str. 27, 01069 Dresden, Germany. E-mail address: [email protected] (D. Scharnweber). 1 Shared first authorship due to equal contribution to this work. 2 These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.msec.2015.10.063 0928-4931/© 2015 Elsevier B.V. All rights reserved.

healing may occur for patient groups with certain risk factors like smoking or systemic diseases such as diabetes, osteoporosis or chronic inflammation [3], or because of aseptic loosening phenomena due to insufficient wear behavior under critical situations [4]. These challenges require either the use of the more biocompatible titanium alloys Ti6Al7Nb [5] or Ti13Zr13Nb [6], or surface treatments to improve biocompatibility. These may include techniques such as oxidation and nanooxidation, deposition of hydroxyapatite (HAp) and nanoHAp coatings, deposition of carbon (diamond-like-carbon DLC and diamond) films, deposition of glass layers and ion implantation [7,8]. Of these, the HAp coatings play an important role in the enhancement of bioactivity, while carbon films are known to increase biocompatibility and wear resistance. Various techniques are used to form HAp deposits, among them magnetron sputtering [9,10], ion-beam physical vapor deposition [11], electrochemically assisted deposition (ECAD) [12,13], electrophoretic deposition [14,15], biomimetic methods [16,17],

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

alternate immersion methods [18,19], and pulsed laser deposition [20]. Due to its robustness and the excellent homogeneity of the coatings, the ECAD process is a low energy method that is increasingly used for the preparation of calcium phosphate phase (CPP) coatings on various implant geometries at an industrial scale [21,22]. Several attempts have been made to improve the HAp adhesion by using an interlayer. Thus, better adhesion of HAp could be achieved on a nanotubular oxide layer compared to air formed oxide layers [23,24]. When HAp coatings were deposited on the titanium substrates covered with porous anodic oxide layers, an enhanced growth of nanophase HAp crystals was observed [25]. An interesting double surface treatment was also proposed [26], which combined the ECAD process with a subsequent plasma vapor deposition (PVD) using radiofrequency magnetron sputtering. Carbon-based films such as microcrystalline diamond (MCD), nanocrystalline diamond (NCD) or diamond-like carbon (DLC) [27–31] combine excellent corrosion resistance [32] with a high biocompatibility [33–35], probably due to an enhanced adsorption of proteins [36]. Therefore they were suggested for use in medicine, mainly in orthopedics [37]. However, large residual stress within the diamond coating may result in poor adhesion of the coatings to a substrate [38]. Nanocrystalline diamond (NCD) layers have been reported to show better bactericidal and bacterial anti-adhesive activities than Ag, though not as good as Cu [39]. No such effects could be observed for microcrystalline diamond (MCD) layers. The introduction of O2 and NH3 during the microwave plasma enhanced chemical-vapor-deposition (MPCVD) process promoted osteoblast adhesion on diamond [40] and the in vitro pattern of cell growth and apoptosis for L929 fibroblast proved that both MCD and NCD films provide non-cytotoxic surfaces suitable for adhesion [31]. The property profiles of diamond films are very complex and properties significantly depend on the CVD parameters. In [41], a pre-etching of the titanium substrate using hydrogen plasma for a short time significantly increased the nuclei density of diamond crystals. Another effective approach was using a DLC layer as the precursor for diamond nucleation, and a graded interlayer combining plasma nitriding followed by plasma carbonitriding. The MPCVD processes using CH4–Ar–H2 gas mixtures were applied to obtain nanocrystalline diamond layer [40]. To effectively protect the titanium against corrosion, the carbon layer must by continuous. In [42] the presence of such barriers was also shown for 3D porous Ti structures. The wear properties of the diamond coated Ti–23Nb–0.7Ta–2Zr–O was far superior to Ti–13Nb– 13Zr alloy due to the presence of beta phase in the alloy [32]. The major results from investigations in this area are that the deposition of diamond layers including NCD layers very effectively reduced corrosion [28,43], but no positive effect was observed in the presence of a DLC layer [44]. Utilizing both the positive effects of diamond layers and DLC films and those of biomimetic HAp coatings is thus expected to combine corrosion stability and improved wear behavior with bioactive and biocompatible implant surfaces. In [45], for a DLC layer prepared on carbon/ carbon composites by CVD technique, the tensile strength between HAp and DLC was only 6.24 MPa. The same authors realized a double-layer system with a metallic base layer and a top coating of Sr substituted HAp on the carbon/carbon composites using a combination of magnetron sputtering and ultrasound-assisted electrochemical deposition [46]. As so far literature results are diverse and indicate the high influence of diamond deposition technologies, we specifically analyzed the effects of the diamond layer property profiles such as homogeneity and electrical conductivity on the resulting HAp coatings. Additionally, the influence of the electrochemical regime of the ECAD process is studied regarding the homogeneity and mass of the resultant HAp coatings. To the best of our knowledge, this is the first such approach to combine these two coating methods for the preparation of sandwich coatings on titanium-based substrates. This may become very useful for better

625

implant-bone tissue strength, shortening the healing time and increasing the lifetime of titanium implants for orthopedics, dentistry, and maxillofacial surgery. 2. Materials and methods All chemicals used for preparation and analysis were of analytical grade and purchased from Sigma-Aldrich (Germany) if not stated otherwise. 2.1. Sample preparation Coin shaped samples from Ti6Al4V (ASTM 136) were used in the investigations. The surfaces of the samples (∅ 16 mm, height 2 mm) were prepared by grinding and polishing with a diamond suspension of 1–3 μm as the final step. Before use, ultrasonic cleaning was done in 1% Triton X-100 in deionized water, acetone, and ethanol for 20 min each. 2.1.1. CVD coatings Previous to the CVD process the samples were seeded with diamond nuclei in a suspension of detonation nanodiamond (average aggregation size of up to 5 nm, Blueseeds, ITC, USA) in dimethylsulfoxide (DMSO) by either immersion (im) or sonication (so) in an ELMA S40H Elmasonic sonication bath as described in [47]. The suspension concentration was 0.5% and seeding time 60 min. Seeded samples were dried in a nitrogen flow. CVD deposition of diamond films was performed by Microwave Plasma Assisted Chemical Vapor Deposition (Astex AX6500, Japan) from a precursor mixture of methane, hydrogen and diborane (B2H6). A process pressure of 50 Torr and a total flow rate of 300 sccm were applied. The microwave power (PMW) used for plasma excitation was kept at 1.3 kW for all experiments. Three different types of coatings were prepared: (i) microcrystalline diamond containing composite layer (MCD/Ti-C), (ii) nanocrystalline diamond (NCD), and (iii) borondoped nanocrystalline diamond (B-NCD). The deposition time was 180 min for all coating parameters; substrate temperatures (TS) of 700 °C (MCD/Ti-C coating) and 500 °C (NCD coatings) were applied. Deposition parameters are summarized in Table 1. 2.1.2. Calcium phosphate coatings Calcium phosphate coatings were formed by the ECAD method. The electrolyte used for deposition was prepared from solutions of 0.033 M CaCl2 and 0.02 M NH4H2PO4 (Fluka, Neu-Ulm, both analytical grade) by diluting equal volumes in a ratio of 1:20 in distilled water. The final pH of the electrolyte was adjusted to 6.4 with 0.5% ammonium hydroxide solution (Merck, Darmstadt, Germany). ECAD was performed in a galvanostatic mode using a combined potentiostat/galvanostat (Jaissle IMP 88-100) in a two-electrode assembly at 36 ± 1 °C. The potentials were measured as cell potentials via the platinum net counter electrode. The general current density profile consisted of a rectangular signal imposed for n cycles followed by an optional final cathodic polarization as shown in Fig. 1. Sequences with the final polarization are consecutively termed ECAD I and those without final polarization ECAD II. The parameters of the ECAD sequences applied to all types of diamond coatings are summarized in Table 2. Table 1 MW PACVD deposition parameters of diamonds films. Type of the diamond layer

MCD/Ti-C NCD B-NCD a

Ratio of gas precursors H2 [%]

CH4 [%]

(B2H6:H2)a [%]

99 96 86

1 4 4

0 0 10

Precursor composition B2H6/H2 = 1:1000.

626

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

Fig. 1. General ECAD sequence.

2.2. Investigation of the prepared surface states A Laser Reflectance Interferometry (LRI) setup using a 405 nm laser diode with 100 mW output developed in our group and described in detail elsewhere [48] has been used for in situ thickness measurements. Molecular composition of the CVD diamond films was investigated with the Raman Microscope JOBIN-YVON T64000 equipped with 514.5 nm Argon ion laser with a measuring time of 20 s per spectrum. The morphology of the obtained surface states was observed by using scanning electron microscopy (Philips XL13ESEM) at low acceleration voltage (3 kV) after sputter coating with a thin carbon layer (Baltec, Germany). The ohmic resistance of the CVD diamond films was measured via the voltage response at 100 points for a current density range between 0.5 and 50 mA/cm2 using a Keithley 2400 source meter. Each current and voltage response was measured ten times per sample. For the conductivity measurements a few nanometers of aluminum were deposited by physical vapor deposition as contact planes on top and bottom of the samples. The chemical analysis of the calcium phosphate coatings was performed following their dissolution in 0.1 N nitric acid using a Calcium-CPC-kit (Analyticon Biotechnologies AG, Germany) and a Phosphate-FS-kit (DiaSys Greiner GmbH, Germany) according to the manufacturer's instructions.

2.3. Cell biological investigations Human mesenchymal stroma cells (hMSCs) from bone marrow aspirates of a healthy male donor (age 22 years) were isolated as described previously [49], and kindly provided by Katrin Müller, group Professor M. Bornhäuser, Medical Clinic I, Dresden University Hospital Carl Gustav Carus. In brief: a 10 ml aliquot of human bone marrow aspirates was obtained from the iliac crest. Human MSCs were isolated by means of a density gradient (Percoll, Biochrom) and tested negative Table 2 Electrochemical conditions and total charge applied to the samples in the ECAD process. Parameter

ic,1/mA/cm2

ECAD I

−5 −6 −7 −5 −6 −7

ECAD II

ic,2/mA/cm2

n

Δt1/s

Δt2/s

−8

10

60

300

n.a.

10

60

n.a.

Total charge density/C/cm2 5.4 6.0 6.6 3.0 3.6 4.2

for the hematopoietic surface marker CD 45 and positive for CD 9, CD 44, CD 54, CD 73, CD 90, and CD 105 (supplementary Fig. 1). Multipotent differentiation potential of the cells has been confirmed by osteogenic (ALP activity), chondrogenic (Alcian blue staining) and adipogenic differentiation (Oil Red O staining) using standard protocols (Supplementary Fig. 2). Cultivation of the cells was performed in expansion medium (exm = Dulbecco's Modified Eagle's Medium (Biochrom AG, Berlin, Germany) containing 10% fetal calf serum (Lonza, Verviers, Belgium), 100 U penicillin and 100 mg/ml streptomycin/2 mM L-glutamine (Biochrom AG)). Three groups of specimens were used: (A) not coated Ti6Al4V substrate, (B) Ti6Al4V substrate with B-NCD coating, (C) Ti6Al4V substrate with the B-NCD base coating and HAp top coating produced by the ECAD I process with a current density of −6 mA/cm2 during cyclic polarization. Samples were heat sterilized at 120 °C for 30 min. As reference empty 12 well plates from tissue-culture-polystyrene (PS) have been used. Cell culture was performed in 12 well plates with a cell seeding density of 20,000 cells/well. Cultivation was performed both, in expansion medium and osteogenic differentiation medium (osm) (= expansion medium supplemented with 10 nM dexamethasone, 0.2 mM ascorbic acid, and 10 mM ß-glycerophosphate) for up to 26 days. The medium was changed twice a week. The samples for analysis of cell adhesion, proliferation and differentiation were taken after 24 h, 7, 14, 21, and 26 days of culture, washed with PBS and, if not otherwise mentioned, frozen at −80 °C until further analysis.

2.3.1. Biochemical analysis of total LDH and ALP activity For biochemical analysis, samples were defrosted on ice for 30 min. Cells were lysed in PBS containing 1% Triton X-100 for 50 min on ice. Proliferation was determined through measurement of lactate dehydrogenase (LDH) activity in the lysis buffer using a Cytotoxicity Kit (Takara, Saint-Germainen Laye, France). 50 μl of cell lysate was mixed with 50 μl of LDH substrate and incubated for 5 min. Subsequently, enzymatic reactions were stopped with 0.5 M HCl and absorbance was measured at 492 nm. LDH activity was correlated with cell number using a calibration curve of cell lysates with defined cell numbers. Osteogenic differentiation of cells was evaluated by measurement of alkaline phosphatase (ALP) activity. For this, 25 μl of cell lysate was mixed with 125 μl of ALP substrate, containing 1 mg/ml p-nitrophenylphosphate, 0.1 M diethanolamine, 1 mM MgCl2 and 0.1% Triton X-100 (pH 9.8), and incubated at 37 °C for 30 min. Reactions were stopped by adding 73 μl of 1 M NaOH. Afterward, absorbance was measured at 405 nm. For each measurement, absorbance values of blanks, consisting of the respective

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

627

lysis buffer and substrate solution, were subtracted from the absorbance of samples values.

(Life Technologies, Darmstadt, Germany) while nuclei were stained using 0.3 μM DAPI (Sigma).

2.3.2. Cell staining and microscopy For fluorescence microscopy using a cLSM 510 meta (Zeiss, Jena, Germany) equipped with a CCD camera was used. For all cell culture time points, two additional samples per type were washed in PBS and fixed for 20 min with 3.7% formaldehyde in PBS, permeabilized with 0.2% Triton in PBS, washed five times with PBS and blocked with 1% bovine serum albumin (Sigma, Deisenhofen, Germany) in PBS. Cytoskeletal staining was carried out with 0.132 μM Phalloidin-Alexa546

2.4. Statistical analysis Deposited mass densities of calcium were tested for statistically significant influences of surface state and deposition current density by two-way analysis of variance (ANOVA) using the software package Origin 8.6. Sample size was 3 for each combination of surface state and electrochemical parameter set. Experiments were conducted at least in triplicate.

Fig. 2. Raman spectra (A to C) and SEM images (D) of sample surfaces after seeding with diamond nuclei and subsequent deposition of different diamond film types: (A) MCD/Ti-C coatings (a) prepared after is seeding, (b) with defect, prepared after is seeding, (c) prepared after so seeding; (B) NCD coating after so seeding; (C) B-NCD coating after so seeding; (D) left MCD/Ti-C coating, middle NCD coating, right B-NCD coating; insets show details in higher magnification.

628

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

3. Results 3.1. Chemical composition, thickness and morphology of the CVD diamond films Raman spectra of CVD diamond films prepared after seeding either by immersion (is) or by sonication (so) are presented in Fig. 2. The spectra shown for MCD/Ti-C coatings prepared using the two different seeding methods and (is: curve (a), so: curve (c) in Fig. 2(A)) present two bands at about 1336 cm−1 and 1350 cm−1. These bands are sharper for so seeded samples. For that reason, the sonication was selected as seeding method for all subsequent deposition processes. Spectrum (b) in Fig. 2(A) shows a second surface state observed for MCD/Ti-C samples including two additional bands at 391 cm−1 and 612 cm−1 together with two wide maxima around 1364 cm− 1 and 1554 cm− 1. The spectrum of the MCD/Ti-C coating after sonication (curve (c) in Fig. 2(A)) shows another small band at 1586 cm− 1. For the NCD layer in Fig. 2(B) a sharp, high-intensity band is observed at 1350 cm−1; an additional weak shoulder shows at 1338 cm− 1. Moreover, a weak band at 1149 cm−1 and a broad maximum around 1547 cm−1 are observed. In the Raman spectrum of the B-NCD layer in Fig. 2(C) strong bands occur at 481 cm−1 and 1230 cm−1. Additionally a sharp peak at 1318 cm−1 is formed in the shoulder of the higher wavenumber range of the 1230 cm−1 peak. The thickness of the NCB and B-NCD diamond films has been measured by LRI to ca. 3 μm; for MCD/Ti-C no reliable LRI results could be obtained due to the inhomogeneous surface of these coatings.

Fig. 2(D) summarizes the results from SEM investigations of the CVD diamond film morphology. The coatings differ strongly in their appearance between the MCD/Ti-C and the two NCD coating states. The MCD/ Ti-C sample shows only a few crystalline deposits with an average diameter of ca. 1 to 2 μm, covering in maximum 5% of the surface and showing some additional clustering (see inset). Both NCD samples are homogeneously coated presenting an additional sub-μm structure, more pronounced for the NCD sample than for the B-NCD one. Both NCD-type samples show some mostly round defect areas ranging between 5 and 20 μm in diameter as well as a few scratches (especially for the NCD sample).

3.2. Conductivity of the CVD diamond layers The current density voltage diagrams of Ti6Al4V substrates uncoated and coated with the different CVD diamond films are presented in Fig. 3. Table 3 summarizes the linear fits of the curves from Fig. 2. The measured current density voltage curves of all samples show a clear ohmic behavior with the specific resistance increasing in the order uncoated Ti6Al4V b MCD/Ti-C coating b B-NCD coating ≪ NCD coating. Whereas the resistance of the MCD/Ti-C coating is in the range of the values for the uncoated substrate, the presence of a BNCD coating results in an increase of the coatings specific resistance by ca. 50%. For not boron-doped NCD the specific coating resistance is again about threefold increased as compared to the boron doped NCD coating.

Fig. 3. Current density voltage curves for diamond layer coated samples. Top left — NCD coating; top right — B-NCD coating; bottom left — MCD/Ti-C coating; bottom right — Ti6Al4V.

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

629

3.4. Surface morphology after HAp coating deposition investigated by SEM

Table 3 Resistance of the coated samples. Substrate/coating

Specific resistance [Ω cm2]

R2

Ti6Al4V/NCD Ti6Al4V/B-NCD Ti6Al4V/MCD/Ti-C Ti6Al4V/no coating

4.4101 ± 0.0135 1.4798 ± 0.0213 0.9702 ± 0.0143 0.9477 ± 0.0231

0.9586 0.9616 0.9963 0.9940

3.3. Analysis of the CPP coatings Fig. 4 shows the deposited mass of calcium, the calculated mass of HAp based on the calcium mass (A) and the calcium/phosphate (Ca/P) ratio (B) of CPP coatings deposited on substrates pre-coated with the three different CVD diamond layers for all investigated ECAD conditions. The HAp coating masses range between ca. 30 and 60 μg/cm2 with the masses prepared under ECAD I conditions being above those prepared under ECAD II conditions for all types of diamond base coating. This mass difference is most pronounced for MCD/Ti-C and B-NCD samples and only slightly visible for NCD samples. Additionally the coating masses depend on the type of the base coating. Highest coating masses have been determined for MCD/Ti-C samples, followed by B-NCD samples, and NCD samples. The Ca/P ratios of all coatings are in the range between 1.65 and 2.0 with no significant differences between the parameter sets (Fig. 5). CPP coatings have been additionally investigated by Raman spectroscopy. All spectra show a strong band at 961 cm− 1 (stretching vibration ν3PO4) indicating the deposition of HAp irrespective of the chosen specific electrochemical conditions (Supplementary Fig. 3). In Fig. 6 the maximum cathodic potential during the ECAD polarization cycles of samples pre-coated with different diamond layers is plotted for different ECAD conditions against the deposited CPP mass (given as the calcium mass of the coating). For all pre-coatings these potentials are slightly less negative for the applied ECAD I conditions. More pronounced are the differences between the types of diamond base coatings. Highest cathodic potentials are measured for NCD coatings, followed by B-NCD coatings and MCD/Ti-C coatings.

Fig. 7 summarizes the results from SEM investigations of the morphology for sample surfaces after deposition of CPP films via ECAD. The sample surfaces coated with CPP for optimal ECAD conditions (middle row of Fig. 7) show in the macro-scale mostly homogeneous coatings. For the MCD/Ti-C substrate the CPP layer seems slightly thicker than for the NCD substrates, whereas for these substrates the macroscopic appearance shows some larger aggregates which are in their arrangement comparable to the defect areas observed for the diamond coatings. These aggregates are more pronounced for the NCD coated substrate as compared to the B-NCD substrate. In the microscale the CPP coatings form a dense, but structured layer of needle like nanocrystals for all types of CVD diamond coated substrates. The structure of the CPP coatings seems more open for the MCD/Ti-C substrate as compared to the NCD substrates, however the deposits from the CVD process are no longer visible. The macro-scale appearance of the CPP coated samples prepared with electrochemical parameters from Table 2 giving worst coating appearance (bottom line of Fig. 7) differs only moderately from that for optimal ECAD conditions. For both NCD samples however the aggregates seem more pronounced. For the MCD/Ti-C sample the micro-scale appearance of the coating is less homogeneous than for optimal ECAD conditions. For B-NCD the micro-scale inset shows a part of an aggregate. When comparing the micro-scale images of the three diamond pre-coated substrates the coatings on the NCD samples seem substantially thinner. 3.5. Cell biological investigations Initial results for cell adhesion and proliferation rate of human MSCs on uncoated Ti6Al4V, Ti6Al4V coated with B-NCD, and Ti6Al4V coated with B-NCD with a top coating from HAp for up to 26 days are presented in Fig. 8(A). All surface types allowed cells to adhere with an adhesion efficiency referred to the number of seeded cells ranging from ~ 38% for uncoated substrate to ~54% for B-NCD coated samples as determined from the cell number after 24 h. An additional HAp coating on a B-NCD base layer resulted in an intermediate adhesion efficiency of ~46%.

Fig. 4. Deposited mass density of calcium and calculated mass density of HAp on substrates pre-coated with CVD diamond layers.

630

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

Fig. 5. Ca/P ratio determined from Ca and P concentration of dissolved coatings in 0.01 M HNO3.

Regarding proliferation, the cell number on all surface types steadily increased without clear indications for a plateau phase. Typically the cell numbers on samples cultivated in differentiation medium were higher than those cultivated in expansion medium reaching up to 15-fold higher cell numbers compared to those after 24 h. For a culture time of 26 days cell numbers on HAp coated samples and on B-NCD coated samples cultured in differentiation medium are significantly higher than for the three other combinations from surface state and culture medium. The activity of alkaline phosphatase (ALP) (ALP activity related to cell number determined from LDH activity) as a marker for osteogenic differentiation, was evaluated in the same time interval as given above (Fig. 8(B)). Values for culture in differentiation medium are given only, as values in expansion medium did not exceed 5 units independent on surface type and culture time (data not shown). The specific ALP activity of cells cultivated in medium with osteogenic supplements rose significantly in all samples showing a time response with an increase over the whole culture period. The highest ALP levels were detected for cells cultivated on uncoated Ti6Al4V

followed by B-NCD coated samples and samples with additional HA top coating. Fluorescence images of hMSC adhering to the three surface types are presented in Fig. 8 C. All surfaces are almost homogeneously covered with cells after 7 days with cells showing a tendency to form colonies on Ti6Al4V and being more homogeneously separated from each other on B-NCD and HAp/B-NCD surfaces. 4. Discussion For excellent long term behavior even in multi-morbid patients suffering in diabetes, osteoporosis, etc. surfaces of orthopedic implants such as hip stems have to meet numerous requirements. For the development of such surface property profiles, combinations of coatings presenting excellent properties in at least one of the required properties will allow for improved overall behavior of implant surfaces. Against this background the aim of the present work was to combine two established coating methods — Microwave Plasma Assisted Chemical Vapor Deposition (MW-PACVD) for the preparation of diamond films, and the biomimetic electrochemically assisted deposition of

Fig. 6. Polarization (voltage) of CVD diamond coated samples during ECAD deposition of CCP at the end of the 60 s polarization intervals plotted against the Ca amount of the coatings; current densities A = −5 mA/cm2, B = −6 mA/cm2, C = −7 mA/cm2.

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

631

Fig. 7. SEM images of HAp substrate surfaces: (top) after deposition of calcium phosphate coatings with optimal coating conditions (ECAD I, −6 mA/cm2); (bottom) worst coating conditions (ECAD II, −5 mA/cm2); insets show details in higher magnification.

hydroxyapatite. The resulting surfaces are expected to combine excellent corrosion resistance and improved wear behavior – realized via the diamond films – with good osteoconductivity – realized by the HAp top coating. Thus, this work was designed to answer the question whether the physico-chemical film properties of diamond films prepared under differing MW-PACVD conditions allow for the successful application of the ECAD method and to combine this with a preliminary characterization of the biological properties of the resulting two layer coating systems. 4.1. MW-PACVD prepared diamond films The Raman spectra of the CVD diamond films prepared under varying processing conditions and using different seeding methods show a number of bands in the wavenumber region between 1200 cm−1 and 1600 cm− 1. For MCD/Ti-C coatings, the bands at 1336 cm−1 and 1350 cm−1 can be assigned to diamond and distorted sp3-phase. Sharper bands indicate a better crystallinity of the coating. The additional bands in spectrum (b) of Fig. 2(A) which can be assigned to Ti-C (391 cm− 1 and 612 cm− 1 and sp2 phase (1364 cm− 1 and 1554 cm− 1)) [50] represent processes running in parallel to the formation of diamond and distorted sp3-phase. Also the bands observed at 1558 cm−1 (curve (a) in Fig. 2(A)) and 1586 cm−1 (curve (c) in Fig. 2(A)) can be assigned to amorphous sp2 phase (“G” band). The amorphous sp2 carbon acts as interlayer between other structures like e.g. metallic substrate and diamond layers to fit the mismatch between substrate and the diamond lattice. Moreover, it can be present in areas between the diamond crystallites, where it fits the difference between their orientations. Thus, it improves the adhesion for the film to the surface and reduces the risk of microcracks. There is more sp2 carbon in NCD and B-NCD than in MCD/Ti-C. In our results we can see sp2 carbon as a wide “G” band in the Raman spectra as well as it slightly increased conductivity of the films. For NCD layers similar to MCD/Ti-C coatings the band with the highest intensity at 1350 cm−1 can be assigned again to distorted sp3-phase, whereas the shoulder at 1338 cm−1 is indicative for stressed diamond. Moreover, the band at 1149 cm−1 can be assigned to nanocrystalline diamond. Both bands observed at 481 cm−1 and 1230 cm−1 in the Raman spectrum of the B-NCD layer (Fig. 2(C)) are indicative for highly boron doped diamond films [51]. According to Bernard et al. [51] from the intensity ratio of ca. two for the bands at 481 and 1230 cm−1 in Fig. 2(C) a doping intensity of about 6000 ppm (corresponding

to ~1.5 ∗ 1021 cm−3) can be roughly estimated in agreement with the shift of the ‘500 cm−1’ maximum to lower wavenumbers, i.e. to 481 cm−1. Additionally, this coincides with the shift of the diamond band to lower wavenumber (1318 cm−1) [51], which is the result of diamond lattice stress caused by boron dopants. In summary the Raman results show that we have for the MCD/Ti-C coatings diamond structures and distorted sp3-phase in combination with dominating surface states Ti-C and sp2-phase. For NCD coatings we find mostly stressed diamond and distorted sp3-phase and, for BNCD boron doped diamond represented by the intensity ratio of the bands at 481 and 1230 cm−1 and band shifts to 481 and 1318 cm−1. These coated samples show distinct differences in their ohmic resistance which can be assigned to differing property profiles of the diamond coatings. Whereas the value for the MCD/Ti-C sample (0.97 ± 0.01 Ω cm2) is comparable with the uncoated substrate (0.95 ± 0.02 Ω cm2), the resistance of the sample with B-NCD coating (1.48 ± 0.02 Ω cm2) and the sample with NCD coating (4.4101 ± 0.0135 Ω cm2) are increased by about 50% and 450% respectively. Given the fact that for the MCD/Ti-C coating microcrystalline diamonds cover less than 5% of the substrate surface (see top of Fig. 3), the specific resistance of this sample will be mostly determined by not diamond coated surface representing Ti-C and sp2 phase, both surface states showing an electronic conductivity comparable to that of the native oxide layer on the uncoated substrate. Both NCD coatings appear dense in the SEM investigations. Their processing conditions differ only in the composition of the gas atmosphere (addition of borane). Thus differences in the ohmic resistance of the coatings are determined by the coatings inherent properties and differences can be clearly assigned to the doping with boron which significantly improves the layers conductivity to values only about 50% above that for uncoated substrates in agreement with work published by Levy-Clement et al. [52] who demonstrated a decrease of the resistivity for boron doped diamond by more than 5 magnitudes for an increase of the doping level from ca. 1 ∗ 1018 to ca. 2 ∗ 1022 cm−3. In summary the conductivity experiments show together with the SEM investigations of diamond coated samples that increasing the CH4 concentration in the gas atmosphere during coating preparation from 1 to 4% together with reducing the substrate temperature from 700 to 500 °C results in dense nanocrystalline diamond coatings instead of microcrystals from diamond covering less than 5% of the substrate surface. Furthermore, additional 10% of diborane (B2H6:H2) precursor significantly increases the electrical conductivity of the resulting diamond

632

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

Fig. 8. Cell biological investigation on selected surface conditions: a) Proliferation as determined via LDH measurements of hMSCs cultivated either in expansion medium (exm) or osteogenic differentiation medium (osm); b) ALP activity as early marker of osteogenic differentiation of hMSCs cultivated in osteogenic differentiation medium; (C) fluorescence images of adhering hMSCs at day 7 in osteogenic differentiation medium (red bars: 100 μm). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

coatings in agreement with Agnes et al. [53] who used substrates coated with a microcrystalline boron doped diamond layer for electrografting of a ruthenium complex film. Boron doping did not change the morphology of the NCD films.

4.2. Properties of the CPP coatings Both from the micro-scale morphology of the CPP coatings and their Ca/P ratio it can be concluded that the electrochemically assisted

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

deposition of CPP resulted in the formation of HAp on all substrates precoated with CVD diamond layers under all applied ECAD conditions. A similar crystal morphology of HAp has been observed by Rößler et al. [13] for the ECAD process using the same electrolyte and temperature as in this work, but constant galvanostatic polarization, and for the formation of mineralized networks of collagen type I. This has been investigated in more detail by Sewing et al. [54] who examined the deposition of different CPP as a function of the electrochemical parameters and the electrolyte composition. The results are also in agreement with recent investigations in our group using cyclic polarization with different pulse ratios for the deposition of HAp either alone or in codeposition with copper [55]. Both, the ECAD regime and the type of diamond pre-layer strongly influence the deposited mass of HAp. For all diamond precoatings the identical ECAD I regime results in the deposition of higher masses of HAp. This effect is most pronounced for MCD/Ti-C samples and becomes weaker in the order B-NCD and NCD. This observation corresponds both with the specific ohmic resistance of the diamond coatings and the maximum cathodic potentials during ECAD polarization. The lower the ohmic resistance of a diamond coating, the lower is the cathodic potential during the ECAD process and the higher the deposited mass of HAp. In parallel this means that the mass yield per charge unit decreases with increasing resistance of the diamond coating. The two ECAD regimes which were applied for the formation of the CPP coatings, which are presented in their properties in Fig. 1, differ only in the final step of 300 s polarization with a constant current density of −8 mA/cm2, i.e. a charge of 2.4 C/cm2. Whereas this additional charge causes an increase of the coating mass of ca. 17 μg/cm2 for MCD/Ti-C samples, this increase goes down to ca. 14 and 3 μg/cm2 for B-NCD and NCD samples respectively. This observation is again in agreement with the abovementioned effect that the mass yield per charge follows the order MCD/Ti-C N B-NCD N NCD, corresponding to the substrate surface conductivities. 4.3. Cell biological response to coatings The cell biological investigations have been focused on the coating system HAp/B-NCD solely because of two reasons. Firstly, corrosion investigations of Mohan et al. [56] for coating systems dominated by Ti-C and sp2-phase similar to the MCD/Ti-C coating in our work gave no improvement of the corrosion behavior of titanium based alloys. Secondly, among the nanocrystalline diamond coatings the B-NCD state has been selected due to its improved behavior in the ECAD coating investigations. The B-NCD coatings show a slight positive effect on the adhesion of human MSCs when compared to not coated substrates and samples with a HAp top coating. Taking the surface topography of these samples into consideration (see Figs. 2(D) and 7) an interpretation could be given that the regular sub-μm topography of the B-NCD samples improves cell adhesion in relation to the smother uncoated sample (not shown; topography comparable to not diamond coated areas of the MCD/Ti-C sample, top line left in Fig. 7). The same is true for a comparison with the HAp coated sample presenting a much rougher surface. An influence of sub-μm to μm surface topography on cell adhesion is also reported by Zinger et al. [57] for titanium based model surfaces. Cell proliferation up to day 14 shows a tendency to HAp/BNCD N B-NCD N Ti6Al4V which is more pronounced for exm culture conditions in agreement with our working hypothesis. Similar effects have been observed by Kim et al. [58] for sol–gel derived HAp coatings on c.p. titanium using the osteoblast like HOS cell line and by Chen et al. [59] for rabbit MSC. For longer culture duration a plateau is indicated. Similar to improved cell adhesion on the two nanorough surfaces also cell proliferation might be improved by the surface topographies of the HAp/B-NCD and B-NCD samples. This goes in line with results published by Dolatshahi-Pirouz et al. [60] studying the effect of nanotopography on the proliferation of dental pulp derived MSC on

633

nanorough tantalum surfaces. The authors correlate this effect to the adsorption of fibronectin and the availability of fibronectin cellbinding domains on the nanorough surfaces. Further, the nanocrystalline HAp coating itself might positively affect cell proliferation similar to observations of Bae et al. [61] when comparing the proliferation of mouse osteoblasts on smooth, chemically treated and additionally nanoHAp coated titanium surfaces. Contrary to cell proliferation, a tendency to a reduced ALP expression on HAp coated surfaces can be observed in agreement with gene expression data observed by Pereira et al. [62] using primary rat osteoblasts for similar HAp coatings directly deposited on titanium. Kim et al. [58] however found increased ALP expression for HAp coatings using an osteoblast-like cell line suggesting a different response of primary cells and cell lines to HAp coated substrates. Interestingly, Mine et al. [63] in their investigations using RAW264.7 osteoclast precursor cells detected an inhibition of the RANKL-dependent osteoclast differentiation for DLC coatings on c.p. titanium as compared to uncoated substrates which points to a potential additional beneficial effect of the bilayer coating system developed in this work. 5. Conclusions Aiming at the improvement of the biological potential of implant surface designed for contact with bone the results clearly show that the deposition of dense and homogeneous diamond coatings realized via a MW-PACVD as base layer can be combined with the biomimetic deposition of HAp as top coating via the ECAD approach. This coating system has been realized on Ti6Al4V as substrate material. For MWPACVD conditions that produce nano-diamond coatings the formation of homogeneous and dense layers with a regular sub-μm topography has been observed. This topography seems to improve the adhesion of human MSCs. The formation of ECAD produced HAp top coatings on diamond base layers was improved for boron doped nanocrystalline diamond films because of their higher electrical conductivity which results additionally in the deposition of higher masses of HAp as compared to not doped nanocrystalline diamond coatings. The in vitro cell testing suggests an improved biological behavior of the developed two layer coating system which is expected to combine improved corrosion resistance and wear behavior with favorable osteoconductivity. 6. Statements The experiments in this study include in vitro studies using human MSCs. These cells were isolated from surplus material derived from bone marrow collections performed in healthy donors for allogenic transplantation and made available by the Medical Clinic I, Dresden University Hospital Carl Gustav Carus in Germany. The donor declaration of consent included the use of stem cells for research purposes (EK263122004). The authors declare explicitly that no conflicts of interests exist and that there is no prior or duplicate publication or submission elsewhere of any part of the work. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.10.063. Acknowledgments The authors would like to thank Mr. H. Szymikowski, M. Sobaszek and L Goluński for their technical assistance. The financial support by the Polish National Centre for Research and Development (NCBiR) under research project No. LIDER/32/205/L-3/11 is gratefully acknowledged. References [1] S.S. Zieliński, Corrosion of titanium biomaterials, mechanisms, effects and modelisation, Corros. Rev. 26 (2008) 1–22.

634

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635

[2] B.G. Pound, Corrosion behavior of metallic materials in biomedical applications. I. Ti and its alloys, Corros. Rev. 32 (2014) 1–21. [3] B.R. Chrcanovic, T. Albrektsson, A. Wennerberg, Diabetes and oral implant failure: a systematic review, J. Dent. Res. 93 (2014) 859–867. [4] D.R. Sumner, R. Ross, E. Purdue, Are there biological markers for wear or corrosion? A systematic review, Clin. Orthop. Relat. Res. 26 (2014) 1–12. [5] D. Scharnweber, R. Beutner, S. Rößler, H. Worch, Electrochemical behavior of titanium-based materials — are there relations to biocompatibility? J. Mater. Sci. Mater. Med. 13 (2002) 1215–1220. [6] A. Zielinski, S. Sobieszczyk, W. Serbinski, T. Seramak, A. Ossowska, Materials design for the titanium scaffold based implant, Solid State Phenom. 83 (2012) 225–232. [7] S.V. Dorozhkin, Calcium orthophosphates, J. Mater. Sci. 42 (2007) 1061–1095. [8] S.V. Dorozhkin, Calcium orthophosphate deposits: preparation, properties and biomedical applications, Mater. Sci. Eng. C (2015) (online). [9] J.A. Jansen, J.G.C. Wolke, S. Swann, J.P.C.M. Van Der Waerden, K. De Groof, Application of magnetron sputtering for producing ceramic coatings on implant material, Clin. Oral Implants Res. 4 (1) (1993) 28–34. [10] V. Nelea, C. Morosanu, M. Iliescu, I.N. Mihailescu, Microstructure and mechanical properties of hydroxyapatite thin films grown by RF magnetron sputtering, Surf. Coat. Technol. 173 (2003) 315–322. [11] M.-S. Kim, U.-W. Jung, S. Kim, J.-S. Lee, I.-S. Lee, S.-H. Choi, Bone apposition on implants coated with calcium phosphate by ion beam assisted deposition in oversized drilled sockets: a histologic and histometric analysis in dogs, J. Periodontal Implant Sci. 43 (2013) 18–23. [12] A. Zielinski, P. Antoniuk, K. Krzysztofowicz, Nanotubular oxide layers and hydroxyapatite coatings on ‘Ti–13Zr–13Nb’ alloy, Surf. Eng. 30 (2014) 643–649. [13] S. Rößler, A. Sewing, M. Stolzel, R. Born, D. Scharnweber, M. Dard, H. Worch, Electrochemically assisted deposition of thin calcium phosphate coatings at nearphysiological pH and temperature, J. Biomed. Mater. Res. A 64 (2003) 655–663. [14] M. Mona Goudarzi, F. Batmanghelich, A. Afshar, A. Dolati, G. Mortazavi, Development of electrophoretically deposited hydroxyapatite coatings on anodized nanotubular TiO2 structures: corrosion and sintering temperature, Appl. Surf. Sci. 301 (2014) 250–257. [15] O. Albayrak, O. El-Atwani, S. Altintas, Hydroxyapatite coating on titanium substrate by electrophoretic deposition method: effects of titanium dioxide inner layer on adhesion strength and hydroxyapatite decomposition, Surf. Coat. Technol. 202 (2008) 2482–2487. [16] R. Narayanan, H.-J. Lee, T.-Y. Kwon, K.-H. Kim, Anodic TiO2 nanotubes from stirred baths: hydroxyapatite growth & osteoblast responses, Mater. Chem. Phys. 125 (2011) 510–517. [17] A. Roguska, M. Pisarek, M. Andrzejczuk, M. Dolata, M. Lewandowska, M. JanikCzachor, Characterization of a calcium phosphate–TiO2 nanotube composite layer for biomedical applications, Mater. Sci. Eng. C 39 (2011) 906–914. [18] A. Kodama, S. Bauer, A. Komatsu, H. Asoh, S. Ono, P. Schmuki, Bioactivation of titanium surfaces using coatings of TiO2 nanotubes rapidly pre-loaded with synthetic hydroxyapatite, Acta Biomater. 5 (2009) 2322–2330. [19] L.-N. Wang, J.-L. Luo, Fabrication and formation of bioactive anodic zirconium oxide nanotubes containing presynthesized hydroxyapatite via alternative immersion method, Mater. Sci. Eng. C 31 (2011) 748–754. [20] C.F. Koch, S. Johnson, D. Kumar, M. Jelinek, D.B. Chrisey, A. Doraiswamy, C. Jin, R.J. Narayan, I.N. Mihailescu, Pulsed laser deposition of hydroxyapatite thin films, Mater. Sci. Eng. C 27 (2007) 484–494. [21] B.G. Bøe, S.M. Röhrl, T. Heier, F. Snorrason, L. Nordsletten, A prospective randomized study comparing electrochemically deposited hydroxyapatite and plasma-sprayed hydroxyapatite on titanium stems, Acta Orthop. 82 (2011) 13–19. [22] M. Stoffel, M. Behr, A. Reinke, C. Stüer, F. Ringel, B. Meyer, Pedicle screw-based dynamic stabilization of the thoracolumbar spine with the Cosmic®-system: a prospective observation, Acta Neurochir. 152 (2010) 835–843. [23] Y.-H. Jeong, B.-H. Moon, H.-C. Choe, W.A. Brantley, Surface characteristics of hydroxyapatite-coated layer prepared on nanotubular Ti–35Ta–xHf alloys by EBPVD, Thin Solid Films 549 (2013) 147–153. [24] H.-J. Kim, Y.-H. Jeong, H.-C. Choe, W.A. Brantley, Surface characteristics of hydroxyapatite coatings on nanotubular Ti–25Ta–xZr alloys prepared by electrochemical deposition, Surf. Coat. Technol. 259 (2014) 274–280. [25] K. Lee, Y.-H. Jeong, Y.-M. Ko, H.-C. Choe, W.A. Brantley, Hydroxyapatite coating on micropore-formed titanium alloy utilizing electrochemical deposition, Thin Solid Films 549 (2013) 154–158. [26] H.-J. Kim, Y.-H. Jeong, H.-C. Choe, W.A. Brantley, Hydroxyapatite formation on biomedical Ti–Ta–Zr alloys by magnetron sputtering and electrochemical deposition, Thin Solid Films 572 (2014) 119–125. [27] M.D. Drory, J.W. Hutchinson, Diamond coating of titanium alloys, Science 263 (1994) 1753–1755. [28] J. Wang, J. Zhou, H.Y. Long, Y.N. Xie, X.W. Zhang, H. Luo, Z.J. Deng, Q. Wei, Z.M. Yu, J. Zhang, Z.G. Tang, Tribological, anti-corrosive properties and biocompatibility of the micro- and nano-crystalline diamond coated Ti6Al4V, Surf. Coat. Technol. 258 (2014) 1032–1038. [29] P.A. Nistor, P.W. May, F. Tamagnini, A.D. Randall, M.A. Caldwell, Long-term culture of pluripotent stem-cell-derived human neurons on diamond — a substrate for neurodegeneration research and therapy, Biomaterials 61 (2015) 139–149. [30] H. Zanin, A.C. Perlevitz, H.J. Ceragioli, A.A. Rodrigues, W.D. Belangero, Magnetic and cytotoxic properties of hot-filament chemical vapour deposited diamond, Mater. Sci. Eng. C 32 (2012) 2340–2343. [31] H. Zanin, P.W. May, A.O. Lobo, E. Saito, J.P.B. Machado, G. Martins, V.J. Trava-Airoldi, E.J. Corat, Effect of multi-walled carbon nanotubes incorporation on the structure, optical and electrochemical properties of diamond-like carbon thin films, J. Electrochem. Soc. 161 (2014) H290–H295.

[32] D. Durgalakshmi, M. Chandran, G. Manivasagam, M.S. Ramachandra Rao, R. Asokamani, Studies on corrosion and wear behavior of submicrometric diamond coated Ti alloys, Tribol. Int. 63 (2013) 132–140. [33] L. Tang, C. Tsai, W.W. Gerberich, L. Kruckeberg, D.R. Kania, Biocompatibility of chemical-vapour-deposited diamond, Biomaterials 16 (1995) 483–488. [34] C.A. Love, R.B. Cook, T.J. Harvey, P.A. Dearnley, R.J.K. Wood, Diamond like carbon coatings for potential application in biological implants—a review, Tribol. Int. 63 (2013) 141–150. [35] R. Hauert, A review of modified DLC coatings for biological applications, Diam. Relat. Mater. 583-589 (2003). [36] C. Stavis, T.L. Clare, J.E. Butler, A.D. Radadia, R. Carr, H. Zeng, W.P. King, J.A. Carlisle, A. Aksimentiev, R. Bashir, R.J. Hamers, Surface functionalization of thin-film diamond for highly stable and selective biological interfaces, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 983–988. [37] S.A. Catledge, V. Thomas, Y.K. Vohra, Nanostructured diamond coatings for orthopaedic applications, diamond-based materials for biomedical applications, Woodhead Publ. Ser. Biomaterials 2013, pp. 105–150. [38] Y. Fu, H. Du, C.Q. Sun, Interfacial structure, residual stress and adhesion of diamond coatings deposited on titanium, Thin Solid Films 424 (2003) 107–114. [39] O. Medina, J. Nocua, F. Mendoza, R. Gómez-Moreno, J. Ávalos, C. Rodríguez, G. Morell, Bactericide and bacterial anti-adhesive properties of the nanocrystalline diamond surface, Diam. Relat. Mater. 22 (2012) 77–81. [40] R. Pareta, L. Yang, A. Kothari, S. Sirinrath, X. Xiao, B.W. Sheldon, T.J. Webster, Tailoring nanocrystalline diamond coated on titanium for osteoblast adhesion, J. Biomed. Mater. Res. A 95 (2010) 129–136. [41] Y. Fu, B. Yan, N.L. Loh, Effects of pre-treatments and interlayers on the nucleation and growth of diamond coatings on titanium substrates, Surf. Coat. Technol. 130 (2000) 173–185. [42] N.A. Braga, C.A.A. Cairo, M.R. Baldan, V.J. Trava-Airoldi, N.G. Ferreira, Optimal parameters to produce high quality diamond films on 3D porous titanium substrates, Diam. Relat. Mater. 20 (2011) 31–35. [43] S. Rupprecht, A. Bloch, S. Rosiwal, F.W. Neukam, J. Wiltfang, Examination of the bone–metal interface of titanium implants coated by the microwave plasma chemical vapor deposition method, Int. J. Oral Maxillofac. Implants 17 (2002) 778–785. [44] L. Mohan, C. Anandan, V.K. William Grips, Corrosion behavior of titanium alloy Beta21S coated with diamond like carbon in Hank's solution, Appl. Surf. Sci. 258 (2012) 6331–6340. [45] L. Zhang, H. Li, Q. Song, K. Li, J. Lu, Y. Zhang, J. Hampshire, W. Li, Applying diamond like carbon layer on carbon/carbon composites and its effect on the deposition of hydroxyapatite coating, Surf. Interface Anal. 46 (2014) 24–29. [46] L. Zhang, H. Li, K. Li, S. Zhang, K. Fu, Y. Zhang, S. Liu, Double-layer TC4/Sr substituted hydroxyapatite bioactive coating for carbon/carbon composites, Ceram. Int. 41 (2015) 427–435. [47] R. Bogdanowicz, M. Smietana, M. Gnyba, M. Ficek, V. Straňák, L. Golunski, M. Sobaszek, J. Ryl, Nucleation and growth of CVD diamond on fused silica optical fibres with titanium dioxide interlayer, Phys. Status Solidi 210 (2013) 1991–1997. [48] M. Kreszewski, R. Bogdanowicz, Laser reflectance interferometry system with a 405 Nm laser diode for in situ measurements of CVD diamond thickness, Metrol. Meas. Syst. 20 (2013) 543–554. [49] J. Oswald, S. Boxberger, B. Jørgensen, S. Feldmann, G. Ehninger, M. Bornhäuser, C. Werner, Mesenchymal stem cells can be differentiated into endothelial cells in vitro, Stem Cells 22 (2004) 377–384. [50] B.H. Lohse, A. Calka, D. Wexler, Raman spectroscopy as a tool to study TiC formation during controlled ball milling, J. Appl. Phys. 97 (2005) 114912. [51] M. Bernard, C. Baron, A. Deneuville, About the origin of the low wave number structures of the Raman spectra of heavily boron doped diamond film, Diam. Relat. Mater. 13 (2004) 896–899. [52] C. Lévy-Clément, F. Zenia, N. Awa Ndao, A. Deneuville, New Diamond Front. Carbon Technol. 9 (1999) 189. [53] C. Agnès, J.-C. Arnault, F. Omnès, B. Jousselme, M. Billon, G. Bidan, P. Mailley, XPS study of ruthenium tris-bipyridine electrografted from diazonium salt derivative on microcrystalline boron doped diamond, Phys. Chem. Chem. Phys. 11 (2009) 11647–11654. [54] A. Sewing, M. Lakatos, D. Scharnweber, S. Roessler, R. Born, M. Dard, H. Worch, Influence of Ca/P ratio on electrochemical assisted deposition of hydroxyapatite on titanium, Key Eng. Mater. 254 (2004) 419–422. [55] C. Wolf-Brandstetter, S. Oswald, S. Bierbaum, H.-P. Wiesmann, D. Scharnweber, Influence of pulse ratio on codeposition of copper species with calcium phosphate coatings on titanium by means of electrochemically assisted deposition, J. Biomed. Mater. Res. B Appl. Biomater. 102 (2014) 160–172. [56] L. Mohan, C. Anandan, V.K. William Grips, Corrosion behavior of titanium alloy Beta21S coated with diamond like carbon in Hank's solution, Appl. Surf. Sci. 258 (2012) 6331–6340. [57] O. Zinger, K. Anselme, A. Denzer, P. Habersetzer, M. Wieland, J. Jeanfils, P. Hardouin, D. Landolt, Time-dependent morphology and adhesion of osteoblastic cells on titanium model surfaces featuring scale-resolved topography, Biomaterials 25 (2004) 2695–2711. [58] H.-W. Kim, Y.-H. Koh, L.-H. Li, S. Lee, H.-E. Kim, Hydroxyapatite coating on titanium substrate with titania buffer layer processed by sol–gel method, Biomaterials 25 (2004) 2533–2538. [59] F. Chen, W.M. Lam, C.J. Lin, G.X. Qiu, Z.H. Wu, K.D.K. Luk, W.W. Lu, Biocompatibility of electrophoretical deposition of nanostructured hydroxyapatite coating on roughen titanium surface: in vitro evaluation using mesenchymal stem cells, J. Biomed. Mater. Res. B Appl. Biomater. 82 (2007) 183–191.

P. Strąkowska et al. / Materials Science and Engineering C 59 (2016) 624–635 [60] A. Dolatshahi-Pirouz, T. Jensen, D.C. Kraft, M. Foss, P. Kingshott, J.L. Hansen, A.N. Larsen, J. Chevallier, F. Besenbacher, Fibronectin adsorption, cell adhesion, and proliferation on nanostructured tantalum surfaces, ACS Nano 4 (2010) 2874–2882. [61] S.E. Bae, J. Choi, Y.K. Joung, K. Park, D.K. Han, Controlled release of bone morphogenetic protein (BMP)-2 from nanocomplex incorporated on hydroxyapatite-formed titanium surface, J. Control. Release 160 (2012) 676–684. [62] K.K.Y. Pereira, O.C. Alves, A.B. Novaes, F.S. de Oliveira, J.-H. Yi, O. Zaniquelli, C. WolfBrandstetter, D. Scharnweber, F. Variol, A. Nanci, A.L. Rosa, P.T. de Oliveira,

635

Progression of osteogenic cell cultures grown on microtopographic titanium coated with calcium phosphate and functionalized with a type I collagen-derived peptide, J. Periodontol. 84 (2013) 1199–1210. [63] Y. Mine, T. Nakatani, K. Okamoto, S. Makihira, S. Iwata, H. Nikawa, Biomimetic diamond-like carbon coated titanium inhibits RANKL-dependent osteoclast differentiation in vitro, J. Photopolym. Sci. Technol. 25 (2012) 523–528.