- Email: [email protected]
Osteoblast interaction with laser cladded HA and SiO2-HA coatings on Ti–6Al–4V
Materials Science and Engineering C 31 (2011) 1643–1652
Contents lists available at ScienceDirect
Materials Science and Engineering C j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m s e c
Osteoblast interaction with laser cladded HA and SiO2-HA coatings on Ti–6Al–4V Yuling Yang a, b, Kaan Serpersu b, Wei He b, c,⁎, Sameer R. Paital b, Narendra B. Dahotre d, 1 a
Department of Physics, Northeastern University, Shenyang 110004, China Department of Materials Science and Engineering, The University of Tennessee, Knoxville, TN 37996, USA c Department of Mechanical, Aerospace and Biomedical Engineering, The University of Tennessee, Knoxville, TN 37996, USA d Department of Materials Science and Engineering, University of North Texas, Denton, TX 76207, USA b
a r t i c l e
i n f o
Article history: Received 30 August 2010 Received in revised form 22 June 2011 Accepted 19 July 2011 Available online 27 July 2011 Keywords: Laser cladding HA/SiO2-HA coating Titanium Osteoblast
a b s t r a c t In order to improve the bioactivity and biocompatibility of titanium endosseous implants, the morphology and composition of the surfaces were modified. Polished Ti–6Al–4V substrates were coated by a laser cladding process with different precursors: 100 wt.% HA and 25 wt.% SiO2-HA. X-ray diffraction of the laser processed samples showed the presence of CaTiO3, Ca3(PO4)2, and Ca2SiO4 phases within the coatings. From in vitro studies, it was observed that compared to the unmodified substrate all laser cladded samples presented improved cellular interactions and bioactivity. The samples processed with 25 wt.% SiO2-HA precursor showed a significantly higher HA precipitation after immersion in simulated body fluid than 100 wt.% HA precursor and titanium substrates. The in vitro biocompatibility of the laser cladded coatings and titanium substrate was investigated by culturing of mouse MC3T3-E1 pre-osteoblast cell line and analyzing the cell viability, cell proliferation, and cell morphology. A significantly higher cell attachment and proliferation rate were observed for both laser cladded 100 wt.% HA and 25 wt.% SiO2-HA samples. Compared to 100 wt.% HA sample, 25 wt.% SiO2-HA samples presented a slightly improved cellular interaction due to the addition of SiO2. The staining of the actin filaments showed that the laser cladded samples induced a normal cytoskeleton and well-developed focal adhesion contacts. Scanning electron microscopic image of the cell cultured samples revealed better cell attachment and spreading for 25 wt.% SiO2-HA and 100 wt.% HA coatings than titanium substrate. These results suggest that the laser cladding process improves the bioactivity and biocompatibility of titanium. The observed biological improvements are mainly due to the coating induced changes in surface chemistry and surface morphology. © 2011 Elsevier B.V. All rights reserved.
1. Introduction To date, a number of materials have been successfully used as implants for hard tissue replacement such as bone. Among them one class of materials has the ability to integrate with host tissue and promote strong bonding at the tissue–material interface. This genre of materials has been dubbed as bioactive materials. In contrast, the other class of materials that remain completely neutral without any toxicity or reaction with the surrounding tissue is called the bioinert materials. Titanium and its alloys, especially Ti–6Al–4V, are the most commonly used bioinert materials for both dental and orthopedic applications [1,2], owing to their excellent mechanical properties, biocompatibility, corrosion resistance, and tissue compatibility. As bio-implants, they adsorb protein from surrounding biological fluid to
⁎ Corresponding author at: 209 Dougherty Engineering Bldg., Department of Materials Science and Engineering, The University of Tennessee, Knoxville, TN 37996, USA. Tel.: + 1 865 974 5275; fax: + 1 865 974 4115. E-mail address: [email protected] (W. He). 1 Formerly at Department of Materials Science and Engineering, The University of Tennessee, USA. 0928-4931/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.msec.2011.07.009
create a protein layer that will support cell growth [3]. Titanium and Ti–6Al–4V being completely neutral also make them the ideal choice for patients who may develop toxic reactions to other metal alloys. Although, titanium is bioinert, an oxide layer forms almost immediately on the surface interacting with the surrounding biological fluid. This naturally forming oxide layer occurs as a result of titanium high reactivity with oxygen [4]. Due to its thin layer, and amorphous and porous structures, the oxide layer possesses a risk of dissolving titanium ions into the body plasma and therefore is not a sound barrier for such dissolutions [5]. However, the field of applications can be expanded if the bioactivity and biocompatibility of the titanium implants are improved through a bioceramic coating. The techniques most commonly used to provide a bioceramic coating on the titanium implants are dip coating [6], sol–gel [7], plasma-spraying [8–10], simultaneous vapor deposition [11], and laser deposition [12] procedures. The criteria for judging the success of these coating processes are whether they can achieve a high crystallinity in the coatings, good adherence between the coating and the metal substrate, control over coating thickness and the ability to coat porous and complex-shaped implants. Of these methods, laser cladding (LC) has gained extensive use due to its ability to achieve appropriate surface textures and
1644
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
surface chemistry and thereby improve the biocompatibility of metallic surfaces at the tissue–implant interface [13–16]. Further, the LC technique has also demonstrated distinct advantages which can be listed as follows [17]: (1) localized heating which reduces thermal distortion and the size of the heat-affected zone; (2) controlled levels of dilution; (3) controlled shape of the clad within certain limits; (4) controlled surface roughness; (5) fine microstructures and metallurgical bonding; (6) high deposition rate and flexibility of the process. Therefore, LC is justifiably becoming a popular choice as a method for modifying surface properties [18–20], and lasers have been used to produce ceramic coatings on metallic materials for biomedical applications. In this way, the benefits of the bioceramics are combined with the mechanical performance of titanium implants. The most common examples of this type of combination are titanium coated with hydroxyapatite (HA) [8,9,11], and titanium with pseudowollastonite [21]. HA (chemical formula Ca10(PO4)6(OH)2) due to its chemical and crystallographic resemblance to bone and tooth minerals [22], has been widely studied as an artificial bone and tooth replacement material in orthopedic and dental implants. They are most commonly used as a coating of hard tissue implants, bone fillers, and for drug delivery. However, HA coated implants have the disadvantage that, in comparison with bioactive glasses and glass ceramics, the rate of bonding of bone tissue with bone implants is relatively low [23] and this has implications for prolonging the time required for patient recovery. Approaches toward improving the integration rates of HA with bone have included the incorporation of biological entities such as growth factors, proteins, and cells into the HA implant [24,25]. As an alternative to improve the biocompatibility of the implant, HA may also be chemically doped with small amounts of elements which are commonly found in physiological bone [26]. It was reported that trace elements of silicon and titanium in calcium HA ceramics and coatings can influence both the biological response of implant materials [27], and the crystallographic, mechanical and chemical properties of manufactured ceramics and coatings [28]. Best and co-workers have put their efforts in studying the development of silicate-substituted HA (Si-HA) [29–31], and demonstrated a significant increase in the amount of bone apposition and organization around Si-HA implants, illustrating their potential as bone graft materials [32]. Other researchers have reported the results of firing a stoichiometric HA precipitate to which SiO2 has been added [33–35]. In these reports, SiO2 was incorporated either by diffusion from an adjacent substrate [33] or by incorporation of SiO2 into the precipitate using the thermal decomposition of a metallorganic additive [34,35]. There are also several reports by Thian and co-workers [36–38] where they have synthesized biocompatible silicon substituted hydroxyapatite composite coatings on Ti substrates by a magnetron co-sputtering technique. The authors reported that the presence of a thin Si-HA coating synthesized by this process stimulated the osteoblast attachment, proliferation and differentiation. So far, few published papers have reported laser cladding of SiO2-HA on metallic alloys and corresponding bone cell response to such coatings. In the present study, HA and SiO2-HA coatings were prepared on Ti–6Al–4V substrates using a laser cladding (LC) technique. Here, a highly intense laser beam was used to melt the precursor (HA and SiO2-HA) and the surface of the Ti–6Al–4V substrate to achieve a micro-textured, multi-phase coating as well as metallurgical bonding at the interface. The in vitro bioactivity was assessed by immersing the samples in a simulated body fluid (SBF). The in vitro biocompatibility, including cell attachment, cell viability, and cell proliferation was investigated by culturing of mouse MC3T3-E1 pre-osteoblast cell line.
2. Materials and methods 2.1. Precursor materials and laser cladding Ti alloy (Ti–6Al–4V) plates, cut from the rolled sheets with a thickness of 3 mm and dimensions of 50 × 50 mm 2 were used as substrates. They were polished using 30 μm grit silicon carbide (SiC) emery paper and sequentially rinsed with acetone to obtain a clean surface free from rust and grease. The HA (Ca10(PO4)6(OH)2) and silica (SiO2) powders, obtained from Fisher Scientific, were used as the precursor materials. The HA and SiO2 precursor powders had a spherical morphology with a unimodal distribution in the 10–30 μm range. The precursors (25 wt.% SiO2-HA or 100 wt.% HA) were mixed in a water-based organic solvent (LISI W 15853) obtained from Warren Paint and Color Company (Nashville, TN, USA) and mechanically stirred for 25 min to get a viscous slurry. The slurry was then sprayed onto the polished and clean substrate coupons using an air pressurized spray gun. The sprayed coupons were dried in air to remove moisture. The thickness of the precursor deposit was maintained at 80 μm for all samples. The laser source employed in this work was a pulsed Nd:YAG laser equipped with a fiber optic beam delivery system that operates in the infrared region with a wavelength of 1064 nm. A high power laser radiation obtained from the above source is directed to the precursor sprayed surface of the substrate and as the laser beam heats up the precursor and the surface of the substrate a molten pool is created on the metallic substrate. Rapid quenching of the molten pool takes place as the laser beam is scanned away from the irradiated area. The scanning is carried in a way such that overlapping (0.1 mm or ~ 11%) laser cladding tracks were laid to obtain a continuous coating. Thus, a coating with metallurgical bonding with the substrate is achieved. The laser processing parameters employed in the present work are listed in Table 1.
2.2. Bioactivity In vitro bioactivity studies were performed by soaking the samples in a simulated body fluid (SBF) solution with ionic composition similar to that of the human plasma. The SBF solution is prepared by dissolving the reagent grade chemicals in the following order: NaCl (8.026 g), NaHCO3 (0.352 g), KCl (0.225 g), K2HPO4·3H2O (0.230 g), MgCl2·6H2O (0.311 g), CaCl2 (0.293 g) and Na2SO4 (0.072 g) in distilled water (700 mL). The fluid was then buffered to pH 7.4 with tri-hydroxymethylaminomethane [(CH2OH)3CNH2] (6.063 g) and hydrochloric acid (1 M, 40 mL) at 37 °C [39]. To study the bioactivity, the in vitro assays were carried out for a set of three samples from each processing condition. The samples were soaked in 10 mL of SBF solution in plastic containers. Soaking periods were varied for 1, 3, 5 and 7 days. The solution was refreshed every 24 h to maintain a pH value of 7.4, and the temperature was maintained at 37 °C during the soaking course. The samples following SBF immersion were dried in a vacuum desiccator for more than 12 h to remove any entrapped water molecules. The weight of samples before and after immersion in the SBF solution was measured to evaluate the HA deposition rate in SBF. The weight change measurements were carried out using a microbalance.
Table 1 Laser parameters used in this study. Pulse width (ms) Pulse repetition (Hz) Pulse energy (J) Spot diameter (μm) Laser scan speed (cm/min) Line spacing (mm)
1.0 30 4.0 900 75 0.1
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
2.3. Surface characterization After laser cladding and SBF immersion experiments, the surface chemical compositions of the samples before and after SBF immersion were characterized using a Philips Norelco X-ray Diffractometer (XRD) with Cu-Kα radiation of wavelength 0.15418 nm. The XRD system was operated at 20 kV and 10 mA in a 2θ range of 20°–100° using a step size of 0.02° and a count time of 1 s. The morphological evolutions of the surface of the coatings were characterized using low magnification Leica optical microscope. The cross-sectional microstructure of the coatings and the morphological evolutions of the surface of coatings following immersion in SBF are characterized using a LEO 1525 scanning electron microscope (SEM). In the current work, the surface roughness measurements of the laser cladded coatings and untreated Ti–6Al–4V substrates were measured using a Mahr Federal profilometer. The profilometer is equipped with a stylus based tip which traverses the surface following surface irregularities. The motion of the stylus tip is recorded by a photoelectric cell and amplified. The stylus tip has a radius of 2 μm. During each measurement the tip traces a length of 5.6 mm and different roughness parameters like Ra (defined as the arithmetic average of all points of the profile also called the center line average height), Rz (arithmetic average of vertical distances between the highest peak and deepest valley within a sampling length) and Rmax (maximum individual roughness depth) were recorded. For the laser cladded coatings the roughness measurements were carried out both parallel and perpendicular to the laser track. A total of six random scans were recorded from different locations on each sample and their average and standard deviation values were reported. Water contact angle for the surfaces of Ti–6Al–4V, laser cladded 25 wt.% SiO2-HA, and 100 wt.% HA coating was measured by the static sessile droplet (3 μL and residence time of 15 s) method using a contact angle goniometer (Cheminstruments, Inc., Fairfield, Ohio).
1645
cells attached on the coated surfaces was expressed as percentage of Ti–6Al–4V control. To study actin cytoskeleton development and focal adhesion formation, immunofluorescent staining was performed. Briefly, 4 h after cell seeding, cells were fixed in a 4 wt.% paraformaldehyde solution for 20 min. After fixation, cells were washed twice with a wash buffer solution (0.05% Tween-20 in phosphate buffered saline (PBS)), then stored in a 0.1 vol.% Triton X-100 PBS solution for 5 min at RT. Blocking solution (1 wt.% bovine serum albumin (BSA) in PBS), was then added to the samples and kept at RT for 30 min. After removal of the blocking solution, samples underwent incubation in primary and secondary antibody solutions at RT. The primary dilute solution contained anti-vinculin (Chemicon, USA) in the blocking solution, and was performed for 1 h. Alexa Fluor 488 goat anti-mouse IgG1 secondary antibody and Alexa Fluor 594 conjugated phalloidin solution were then applied to the samples for 45 min. A DAPI stain for cell nuclei was applied before imaging. Fluorescent images were captured with a Nikon Eclipse Ti microscope (Nikon, Japan). 2.6. Cell proliferation Proliferation of the osteoblast-like cells on Ti–6Al–4V, 25 wt.% SiO2-HA and 100 wt.% HA samples was performed in 24-well plates. Cells were seeded at an initial density of 7.5 × 10 3 cells/cm 2 and incubated for up to 7 days. The relative cell growth rate for 25 wt.% SiO2-HA and 100 wt.% HA samples was expressed as a percentage of Ti–6Al–4V control. Following incubation, the WST-1 reagent was added to each of the wells in a 1 to 10 dilution (reagent to cell culture medium). Samples were then incubated for 4 h at 37 °C, after which, 100 μL of solution from each sample was transferred into a 96-well plate. The absorbance was recorded at 450 nm using a Wallac 1420 Victor 2 Multilabel Counter (Perkin Elmer, USA). 2.7. Cell morphology
2.4. Cell culture Cell culture assays were performed using the mouse pre-osteoblast MC3T3-E1 cell line (ATCC, Manassas, VA, USA). Cells were cultured in α-modified minimum essential medium (α-MEM, Gibco, USA) supplemented with 10% fetal bovine serum (FBS), and 2% penicillin/ streptomycin (P/S) at 37 °C in a 5% CO2 atmosphere. The culture medium was replaced every 3 days and confluent cells were trypsinized and replated (0.25% trypsin-EDTA, Invitrogen, USA) to maintain the cell line.
Scanning electron microscopy (SEM) was used to study cell morphology on the various substrates. After 1 day and 7 days of culture, the samples were fixed both by a primary fixation solution, 3% glutaraldehyde in 0.1 M cacodylate buffer, and a secondary fixation solution, 2% osmium tetroxide in 0.1 M cacodylate buffer, for 1 h each, at RT. Samples were then washed with buffer and dehydrated with a series of increasing concentration of ethanol (25%, 50%, 70%, 95%, and 100%), critical point dried, sputter-coated with gold, and examined with a LEO 1525 scanning electron microscope at an accelerating voltage of 5 kV.
2.5. Cell attachment 2.8. Statistical analysis Two days prior to cell seeding, designated wells in 24-well plates were coated with 12 wt.% poly(2-hydroxyethyl methacrylate) (polyHEMA) (Polysciences, Inc.) in ethanol solution. The nonadhesive polyHEMA coating ensures cells attach and grow exclusively on the samples instead of the surrounding areas on tissue culture plates. Samples (Ti–6Al–4V, 25 wt.% SiO2-HA and 100 wt.% HA coated Ti alloy) were then placed in the wells and sterilized under ultraviolet light in an enclosed biosafety cabinet for 12 h. Cell attachment and viability were both examined qualitatively by a live/dead staining assay and quantitatively by counting the number of attached cells. MC3T3-E1 cells were seeded at 1 × 10 4 cells/cm 2 density on all sterilized samples and incubated for 4 h at 37 °C. Samples were then stained via a live/dead assay containing 2 μM calcein AM and 4 μM ethidium homodimer-1 (EthD-1). After 20 min of incubation at room temperature (RT), samples were viewed using a Nikon Eclipse E600 fluorescence microscope (Nikon, Japan). Viable cells were stained with calcein AM (green), while non-viable cells were stained with EthD-1 (red). Cell numbers in three randomly chosen areas on the surface of each sample were counted quantitatively. The total number of live
All cell culture experiments were performed simultaneously on at least triplicate samples. Each experiment was repeated once. Data were reported as mean ± standard error of mean (n ≥ 3). Statistical comparisons were performed using Student's t-test (JMP, USA), while p b 0.05 was considered statistically significant. 3. Results 3.1. Surface characterization The surface morphology of the untreated Ti–6Al–4V and the coatings (100 wt.% HA and 25 wt.% SiO2-HA) before SBF immersion is illustrated in Fig. 1. It can be observed that by adding SiO2 surface features at finer length scales (Fig. 1c) are obtained as compared to the pure HA precursor (Fig. 1b). The cross-sectional microstructure of the coatings (100 wt.% HA and 25 wt.% SiO2-HA) presented as inset in Fig. 1b and c also clearly demonstrated the rough morphology of the coatings and the sound bonding between
1646
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
HA coatings) compared to Ti–6Al–4V control, as noticed from the data in Table 2. It can also be noted that the coatings have a higher surface roughness when measured perpendicular to the laser track compared to the measurements made parallel to the track. This is due to the ridges between two consecutive parallel laser tracks in one specific direction that increased the surface roughness in that direction compared to Ti–6Al–4V control. However, within the laser cladded coatings (100 wt.% HA and 25 wt.% SiO2-HA coatings) the surface roughness did not vary much thereby indicating no distinct relation with the varying composition of precursor material within the range of present work. Surface wettability was significantly different among the three samples. Water contact angle on Ti–6Al–4V, 100 wt.% HA, and 25 wt.% SiO2-HA is 59 ± 2, 53 ± 2, and 46 ± 1°, respectively. 3.2. Bioactivity
Fig. 1. Low magnification optical microscopic images of the surface texture of the (a) untreated Ti–6Al–4V and Ti–6Al–4V samples coated with (b) 100 wt.% HA and (c) 25 wt.% SiO2-HA. The insets in panel (b) and panel (c) correspond to the higher magnification SEM images of the cross-section of the respective coatings.
the coating and the substrate material. A substantial increase in surface roughness (measured in terms of Ra, Rz, and Rmax) was observed for laser cladded coatings (100 wt.% HA and 25 wt.% SiO2-
The in vitro bioactivity of the 100 wt.% HA and 25 wt.% SiO2-HA coatings and untreated Ti–6Al–4V control was assessed by the precipitation of an apatite (HA, [Ca10(PO4)6(OH)2]) like mineral layer on the surface following immersion in SBF. As stated earlier, HA is a natural mineral component of the human bone [22], hence, this layer provides the appropriate chemistry and acts as a bone bonding interface. SEM analysis of the surfaces of 100 wt.% HA and 25 wt.% SiO2-HA coatings and untreated Ti–6Al–4V after immersion in SBF for 7 days is presented in Fig. 2. The untreated Ti–6Al–4V (Fig. 2a) barely indicated any deposition or precipitation of HA phase on its surface. In contrast, the sample laser processed with 100 wt.% HA and 25 wt.% SiO2-HA precursors clearly demonstrated (Fig. 2b and c) a thick deposition of apatite like layer. XRD spectra collected on samples before SBF immersion and after 7 days of immersion time are presented in Fig. 3. Some minor phases, such as TiO and CaO were not marked in Fig. 3b. Fig. 3a indicated that, compared to bare Ti, some new phases, specifically CaTiO3, Ca3(PO4)2, Ca2SiO4, CaO, and TiO2, were formed within the coatings by laser cladding 25 wt.% SiO2-HA and 100 wt.% HA precursors on Ti–6Al–4V substrate. However, it is obvious that with the same set of laser processing parameters, different phases were evolved within the coatings for 25 wt.% SiO2-HA and 100 wt.% HA precursors. The main phases evolved by cladding 25 wt.% SiO2-HA precursor include CaTiO3, Ca3(PO4)2, Ca2SiO4, and minor phase TiO2, but for 100 wt.% HA precursor samples, TiO2 is the main phase together with some CaTiO3 and minor Ca3(PO4)2 and CaO. However, no HA was detected in any sample laser processed using the set of parameters employed in the present work. Compared to the XRD pattern before SBF immersion (Fig. 3a), it is observed that HA was precipitated on all the sample surfaces after 7 days of immersion in SBF (Fig. 3b). The characteristic peaks of crystalline HA-like structures indicate that very small amount of HA precipitated on Ti–6Al–4V sample after a 7-day immersion in SBF. On the contrary, 100 wt.% HA coated sample appeared to have more HA precipitation than the Ti alloy control. But for 25 wt.% SiO2-HA sample, the HA peaks observed in the XRD patterns after 7 days of SBF immersion are very strong and correspond to the characteristic HA peaks at 2θ ~22°, 25.6° and 31.7° (representing the planes (200), (002), and (112), respectively). Compared to Ti alloy control, both 100 wt.% HA and 25 wt.% SiO2-HA samples, have higher surface free
Table 2 Surface roughness and contact angle of the samples. Samples
Ti–6Al–4V 25 wt.% SiO2-HA 100 wt.% HA
Parallel to the laser track
Perpendicular to the laser track
Ra (μm)
Rz (μm)
Rmax (μm)
Ra (μm)
Rz (μm)
Rmax (μm)
– 2.6 ± 0.3 2.5 ± 0.3
– 15 ± 3 14 ± 1
– 18 ± 2 18 ± 2
– 3.3 ± 0.3 4.0 ± 0.4
– 16 ± 2 20 ± 2
– 22 ± 5 28 ± 4
Ra (μm)
Rz (μm)
Rmax (μm)
Contact angle (θH2O/degrees)
0.41 ± 0.04 – –
3.4 ± 0.3 – –
4.8 ± 0.6 – –
59 ± 2 46 ± 1 53 ± 2
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
1647
Fig. 3. XRD patterns of Ti–6Al–4V, 25 wt.% SiO2-HA and 100 wt.% HA coated sample (a) before, and (b) after immersion of 7 days in SBF.
Fig. 2. Low magnification SEM images of the surfaces of (a) untreated Ti–6Al–4V (b) 100 wt.% HA coated, and (c) 25 wt.% SiO2-HA coated samples following immersion in SBF for 7 days.
energy as calculated in our previous work [40], increased wettability, and possess more intense peaks of HA. In addition, a clear presence of the HA peak at 2θ = 45.4° corresponding to the plane (311) was detected in 25 wt.% SiO2-HA sample. Further, it is also worth noticing that for 25 wt.% SiO2-HA samples, after 7 days of immersion in SBF, the apatite peak (at 2θ ~22° and 31.7°) broadened with a reduction in intensity. Fig. 4 presents the weight increase in laser cladded samples and Ti–6Al–4V control following immersion in SBF for 1 day, 3, 5, and 7 days, respectively. It is obvious that all laser treated samples (25 wt.% SiO2-HA or 100 wt.% HA precursor) exhibited a higher weight increase than the bare Ti alloy over the entire range of immersion periods employed in the present work. Compared to the 100 wt.% HA coated sample, the 25 wt.% addition of SiO2 in the precursor leads to a higher increase in weight after all immersion times. Increase in weight for laser cladded samples after all immersion times might be attributed to the
Fig. 4. Weight increase as a function of SBF immersion time for 1 day, 3 days, 5 days, and 7 days.
1648
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
presence of new phases, such as CaTiO3, Ca3(PO4)2 and Ca2SiO4 within the coating. 3.3. Cell attachment Live/dead staining results (Fig. 5a–c) indicated that cells attached well on the substrates after 4 h of incubation with very few dead/ disrupted cells. This suggested good initial adhesion and high viability of cells exposed to modified surfaces. Quantitatively, the total number of cells attached after 4 h on the 25 wt.% SiO2-HA and 100 wt.% HA coated samples was significantly higher than the Ti–6Al–4V control (Fig. 5d). Actin cytoskeleton and focal adhesion play important roles in regulating the cell shape and maintaining cell adhesion. The actin labeling (Fig. 6) revealed similar cytoskeletal organization and focal adhesion distributions in cells grown on all samples after 4 h seeding. Most cells exhibited a polygonal morphology with good cytoskeletal development (red stains). Focal adhesion protein, vinculin (green) can be seen within the cytoplasm as well as at the tips of stretching cytoplasmic projections. 3.4. Cell proliferation The absorbance results of WST-1 assays used to study cell proliferation are shown in Fig. 7. The inset shows the proliferation rate with regard to untreated Ti–6Al–4V. It is clear that the absorbance for all the samples presented an increasing trend as a function of time, indicating cells were proliferating on all the tested substrates. Within all the time points studied, the cell proliferation rates on the laser cladded samples were comparable or even higher than the untreated Ti alloy control. Specifically, at day 1, proliferation rate on the 25 wt.% SiO2-HA coated sample (114.7 ± 2.2%) was comparable with the Ti
alloy control, but significantly lower than the 100 wt.% HA coated sample (127.7 ± 5.7%). At day 4, proliferation rates were comparable between 25 wt.% SiO2-HA and 100 wt.% HA samples and both were significantly higher than Ti alloy. At day 7, 25 wt.% SiO2-HA coated samples induced the highest proliferation rate of 173.5 ± 18.7%, while that for 100 wt.% HA coated sample is 148.7 ± 10.9%, and both were significantly higher than Ti alloy. 3.5. Cell morphology Cell morphology after 1 day and 7 days of proliferation was assessed by SEM, and resulting images are shown in Fig. 8. From Fig. 8c and e, it can be seen that the normal individual MC3T3-E1 osteoblasts on laser cladded sample (25 wt.% SiO2-HA and 100 wt.% HA) surfaces mostly had a triangular form, with an average size of around 20 μm in width and 60–80 μm in length. This exhibits that good adhesion was obtained by laser cladding 25 wt.% SiO2-HA and 100 wt.% HA precursors on Ti–6Al–4V substrate. Their good adhesion can be characterized by the presence of lamellipodia, i.e. large front projections, which strongly adhere to the substrate and draw the cellular body by cytoplasmic contraction. It is also found that several multiple microvilli exist on the surface of the cells, and fine cytoplasmic extensions in multiple directions were formed. Those cultured on 25 wt.% SiO2-HA and 100 wt.% HA samples revealed greater spreading with large lamellipodia, indicating active cell migration. On untreated Ti–6Al–4V sample, however, the osteoblasts showed a polygonal form (Fig. 8a). In addition, the numbers of filopodia and cytoplasmic extensions were also less on cells grown on bare Ti alloy. Fig. 8b, d and f illustrates the cell morphology after 7 days of proliferation. It is clear that after proliferation for 7 days, the cells distributed evenly and compactly on the sample surfaces, which is in accordance with the results of Fig. 7 and exhibits a proliferation
Fig. 5. Live/dead staining of MC3T3-E1 cells after 4 h of incubation on (a) Ti–6Al–4V, (b) 25 wt.% SiO2-HA coating, and (c) 100 wt.% HA coating; and (d) cell numbers attached to Ti control, 25 wt.% SiO2-HA coating, and 100 wt.% HA coating within observation areas after 4 h seeding; results are expressed as percentages of cells attached on Ti control (n = 3). * denotes that the cell number on untreated Ti–6Al–4V is significantly lower than laser cladded Ti (p b 0.05), scale bars are 100 μm.
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
1649
Fig. 7. Histograms of WST-1 assay results representing proliferation activities of MC3T3-E1 cells after 1 day, 4 and 7 days of growth; inset results are expressed as percentages of activities on bare Ti, * denotes that the absorbance for untreated Ti–6Al– 4V sample is significantly lower than that of laser cladded 25 wt.% SiO2-HA and 100 wt.% HA at day 1, day 4, and day 7 (p b 0.05). In inset, # denotes that at day 1, cell proliferation rate on 100 wt.% HA is significantly higher than the other two samples, and * denotes that at day 4 and day 7, cell proliferation rate on untreated Ti is significantly lower than the laser cladded samples (p b 0.05).
4. Discussion
Fig. 6. Fluorescent images for MC3T3-E1 cells adhered on (a) Ti–6Al–4V, (b) 25 wt.% SiO2-HA coating, and (c) 100 wt.% HA coating after 4 h, showing focal contacts (green), actin filaments (red), and the nucleus (blue). Scale bars in large images and in insets are 100 μm and 50 μm, respectively.
trend for all samples. The inset is the image with higher magnification indicating excellent cell spreading and cell–cell interaction after 7 days of proliferation.
To improve longevity and minimum failure of an orthopedic implant, one of the most challenging obstacles is to control osseointegration. Therefore, a variety of coatings, including oxides, bioglass and bioceramics on Ti–6Al–4V substrates have been developed by sol–gel, dip coating, and laser deposition processes. In order to examine how a coating affects its biocompatibility, one should investigate the behavior of cells on coated implant surface. Since osteoblasts, the bone forming cells, play a critical role in osseointegration of an orthopedic implant in vivo, the behavior of such cell type is commonly studied in vitro for the development of new orthopedic biomaterials or surface modifications of existing ones. The purpose of the present study was to investigate the interaction of osteoblasts with laser cladded HA and SiO2-HA coatings with varying chemical compositions and different surface energies. From Fig. 1 it can be observed that 25 wt.% SiO2-HA precursor resulted in a surface with finer microstructure as compared to 100 wt.% HA precursor. This significant change in the microstructure with the addition of SiO2 can be explained based on the work by Yoo [41] and Hussaina [42]. The authors reported that the addition of SiO2 can accelerate the material transfer and result in faster microstructural evolution [41], and also control the microstructure by suppressing abnormal grain growth [42]. Hence, it is clear from their results that SiO2 addition can reduce the crystallization time which in turn, can result in surface features at finer length scales. Further, as illustrated in Fig. 3a, no pure HA was observed after laser processing. The chemical and phase compositions of laser cladded 100 wt.% HA coating were different from that of 25 wt.% SiO2-HA coating. The reactions between the precursor and substrate material during processing using a high energy laser beam are important in predicting the phases evolved within the coating. In laser cladding process, the laser beam is focused on the surface of Ti alloy pre-pasted with precursors 100 wt.% HA and 25 wt.% SiO2-HA. The high energy of laser beam melts the precursor and the local portion of the substrate surface and creates a molten pool which is maintained during the laser irradiation. Due to the high temperature of laser beam, interactions between Ti alloy substrate and HA or SiO2-HA precursor can be induced. It was shown that multiple phases can be formed between these compositions during high temperature processing,
1650
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
Fig. 8. SEM images for cell morphology after 1 and 7 days of proliferation on Ti–6Al–4V (a and b), 25 wt.% SiO2-HA coating (c and d), and 100 wt.% HA coating (e and f). Scale bars in (a), (c) and (e) are 10 μm; and in (b), (d) and (f) are 100 μm. Scale bars in inset are 5 μm.
which can be characterized as Ca3TiO7, CaTiO3, Ca4Ti3O10 and Ca5Ti4O3 [20,43]. The possible reactions between HA and Ti can be characterized as follows [44]:
Besides the aforementioned equations, there is another possible reaction between SiO2-HA and Ti: 2CaO þ SiO2 ¼ Ca2 SiO4 ðΔG ¼ −122:17 kJ=molÞ:
Ti þ O2 ¼ TiO2 ðΔG ¼ −864:09 kJ=molÞ
ð1Þ
HA þ TiO2 ¼ CaTiO3 þ 3TCP þ H2 O ðΔG ¼ −7312:39 kJ=molÞ
ð2Þ
TCP þ TiO2 ¼ CaTiO3 þ α Ca2 P2 O7 ðΔG ¼ −41:51 kJ=molÞ
ð3Þ
α Ca2 P2 O7 þ Ti ¼ CaTiO3 þ CaO
ð4Þ
þ P2 O3 ðgÞ↑ ðΔG ¼ −1247:46 kJ=molÞ:
ð5Þ
All the above equations clearly suggest that the chemical compositions of the laser cladded coating depend on the precursor compositions and the reactions among them. Same phases such as TiO2, CaTiO3, and Ca3(PO4)2 (TCP) were detected both in laser cladded 100 wt.% HA and 25 wt.% SiO2-HA coatings based on-the aforementioned reactions (1)–(4). Due to the absence of SiO2 in 100 wt.% HA precursor, Eq. (5) will not occur, and instead it led to the retention of CaO in laser cladded 100 wt.% HA coating. On the contrary, in laser
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
cladded 25 wt.% SiO2-HA coating, reaction (5) occurred, and led to the formation of additional phase of Ca2SiO4. Numerous studies have shown that the roughness, and chemical composition of the surface, which in turn determine the surface static charge and the wettability (surface free energy), are the main factors affecting the cell activity on implants [45]. However, it is still unclear what kind of roles those surface characteristic parameters play, and how they act on each other. For instance, Webster [46] proposed that surface topography is the main factor influencing the cell response rather than the specific chemistry of a material surface. But the results of Anne Ochsenbein [45] are definitely inconsistent with that observation. In their work, they pointed out that the mutual action between different but interrelated surface characteristic parameters and their co-adjustment of cell responses is definitely more complicated and cannot be described by the surface structure alone. In the present work, we draw the same conclusion as Anne Ochsenbein. As illustrated in Table 2, although the laser cladded samples have similar surface roughness they possess distinctly different water contact angles, indicating that the chemical phase has strongly influenced the wettability, and likely to play an important role with respect to protein absorption, cell attachment and spreading. As stated earlier, the major difference in chemical composition between 25 wt.% SiO2-HA and 100 wt.% HA samples is the presence of Ca2SiO4. We, therefore, deduced that it is because of the formation of Ca2SiO4 that leads to a lower contact angle (Table 2) and improved bioactivity (more HA formation and higher weight increase in SBF) for 25 wt.% SiO2-HA than that of 100 wt.% HA coating. The work by Gou et al. showed that Ca2SiO4 possessed good bioactivity in SBF environment [47]. This is because the addition of SiO2 in HA can develop the grain boundaries and reduce the grain size [48,49], which subsequently improves the wettability of the surface. The detailed discussion has been reported in our previous work [50]. Also as previously described, for 25 wt.% SiO2-HA samples, after 7 days of immersion in SBF, the apatite peak (at 2θ ~ 22° and 31.7°) broadened with reduction in intensity. This indicates that there is a transformation from a highly crystallized HA phase to a near amorphous or fine crystallite sized HA phase. This rapid transformation can again be attributed to the high biomineralization rate owing to the improved wettability as a result of phases such as Ca2SiO4. Cell attachment and viability results show that laser cladding HA and SiO2-HA can increase the number of cells adhering to the substrate for increased cell viability. This result is in accordance with the work by Hao and Feng [51,52]. Similar results were observed from cell proliferation experiments. A significantly higher proliferation rate was achieved by laser cladding process compared to untreated Ti alloy substrate. Moreover, after laser cladding process, the cultured samples revealed greater spreading with large lamellipodia, indicating active cell migration. These results can be attributed to the presence of new phases (Ca3(PO4)2, CaTiO3 and Ca2SiO4), and three dimensional surface texture which in turn increase the surface energy and, subsequently, the wettability of the surface. In our previous work [40], surface free energy has been calculated, and the values for Ti– 6Al–4V, 100 wt.% HA and 25 wt.% SiO2-HA are 34.19 ± 0.05, 49.47 ± 0.18, and 56.20 ± 0.97 mJ/m 2, respectively. It was reported that surfaces with a higher surface free energy are more adhesive toward cell binding proteins and bone cells than those with a low surface free energy [46]. Comparing the results of 25 wt.% SiO2-HA coating to those of 100 wt.% HA coating, it is obvious that the cell viability and proliferation rate for 25 wt.% SiO2-HA coating are slightly higher due to the addition of SiO2 in the precursor. The results are in accordance with the work by Zou et al. [53], in which they stated that silicon substitution into hydroxyapatite was observed to affect human osteoblast (HOB) cell adhesion and different populations of cells respond differently to the silicate supplement. Further Si substitution can improve cell adhesion and proliferation was also given by Botelho
1651
C.M. et al. [54]. In their work, the authors reported that surface charge on Si-HA is different from that on pure HA and thereby affects cell adhesion. As already shown in multiple previous investigations on laser cladding 100 wt.% HA and 25 wt.% SiO2-HA coatings and untreated Ti– 6Al–4V substrates with different chemical compositions, further evidence was provided here that the bioactivity and biocompatibility were improved by the laser cladding process. The wettability derived from chemical composition is considered as the predominant parameter influencing the interactions at the interface of the coating and the bio system. 5. Conclusions Titanium surface treatments by laser cladding 100 wt.% HA and 25 wt.% SiO2-HA coatings have been performed to optimize the cell interaction with the substrate. XRD results showed that new phases were formed by laser cladding process, which subsequently improved the wettability of the surface. In biological tests, all laser cladded samples presented good but differential cell reactions. 25 wt.% SiO2HA samples showed a significantly higher HA precipitation rate after immersion in SBF. A significant higher cell attachment and proliferation rate were observed for both laser cladded 100 wt.% HA and 25 wt.% SiO2-HA samples. Compared to 100 wt.% HA sample, 25 wt.% SiO2-HA samples presented a slightly higher cell interaction due to the addition of SiO2. The observed biological improvements for the laser cladded samples are therefore due to the change in surface morphology and the chemical composition of the surface coatings. Acknowledgment Support to Yuling Yang during this work at the University of Tennessee by the National Science Foundation of China for Young Scholars (grant no. 50801012), the Fundamental Research Funds for the Central University (grant no. N100405001), and the Science Foundation of Liaoning Province (grant no. 20102072) is highly acknowledged. Yuling Yang also thanks China Scholarship Council (CSC) and Northeastern University (NEU) of China for providing financial support as a visiting scholar at the University of Tennessee. The authors thank Ms. Lu Huang for her help with the cell culture. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
T.M. Lee, E. Chang, C.Y. Yang, Biomaterials 25 (2004) 23–32. A. Goransson, E. Janssn, P. Tengvall, A. Wennerberg, Biomaterials 24 (2003) 197–205. N. Mirhosseini, P.L. Crouse, L. Li, D. Garrod, Appl. Surf. Sci. 253 (2003) 7798–8002. D.M. Brunette, Titanium in Medicine, Springer, London, 2001. Toth, G. Szabo, L. Kovacs, K. Vargha, J. Barabas, Z. Nemeth, Smart Mater. Struct. 11 (2002) 813–818. T. Li, J. Lee, T. Kobayashi, H. Aoki, J. Mater. Sci. Mater. Med. 7 (1996) 355–357. A. Milev, G.S.K. Kannangara, B. Ben-Nissan, Mater. Lett. 57 (2003) 1960–1965. S.T. Ding, C.P. Ju, J.H. Chern Lin, J. Mater. Sci. Mater. Med. 11 (2000) 183–190. K.A. Khor, A. Vreeling, Z.L. Dong, P. Cheang, Mater. Sci. Eng. A 266 (1999) 1–7. X. Liu, C. Ding, Z. Wang, Biomaterials 22 (2001) 2007–2012. L. Pawlowski, J. Therm. Spray Technol. 8 (1999) 279–294. J.M. Fernandez-Pradas, P. Serra, J.L. Morenza, P.N. De Aza, Biomaterials 23 (2002) 2057–2061. A. Kurella, N.B. Dahotre, J. Biomater. Appl. 20 (2005) 5–50. S.R. Paital, Z. Cao, W. He, N.B. Dahotre, Biofabrication 2 (2010) 1–14. S.R. Paital, W. He, N.B. Dahotre, J. Mater. Sci. Mater. Med. 21 (2010) 2187–2200. S.R. Paital, W. He, C. Daniel, N.B. Dahotre, J. Mater. 62 (2010) 76–83. W.M. Steen, Laser Material Processing, Springer-Verlag, London, 2003. F. Lusquiños, J. Pou, M. Boutinguiza, F. Quintero, R. Soto, B. León, M. Pérez-Amor, Appl. Surf. Sci. 247 (2005) 486–492. D. Wang, C. Chen, J. Ma, T. Lei, Appl. Surf. Sci. 253 (2007) 4016–4020. Y.C. Wang, Y.M. Li, H.L. Yu, J. Ding, X.H. Tang, J.G. Li, H.Z. Yao, Surf. Coat. Technol. 200 (2005) 2080–2084. P.N. De Aza, J.M. Fernandez-Pradas, P. Serra, Biomaterials 25 (2004) 1983–1990. L.L. Hench, J. Wilson, An Introduction to Bioceramics, World Scientific, London, 1993. H. Oonishi, L.L. Hench, J. Wilson, F. Suginara, E. Tsuji, S. Kushitani, H. Iwaki, J. Biomed. Mater. Res. 44 (1999) 31–43.
1652
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
[24] M.H. Mankani, S.A. Kuznetsov, B. Shannon, R.K. Nalla, R.O. Ritchie, Y. Qin, P.G. Robey, Am. J. Pathol. 168 (2006) 542–550. [25] M. Mastrogiacomo, A. Muraglia, V. Komlev, F. Peyrin, F. Rustichelli, A. Crovace, R. Cancedda, Orthod. Craniofac. Res. 894 (2005) 277–284. [26] F.C.M. Driessens, Bull. Soc. Chim. Belg. 89 (1980) 663–689. [27] L.L. Hench, J. Am. Ceram. Soc. 74 (1991) 1487–1510. [28] J.C. Elliott, Structure and Chemistry of the Apatites and Other Calcium Orthophosphates, 18, Elsevier, Amsterdam, 1994. [29] A.E. Porter, S.M. Best, W. Bonfield, J. Biomed. Mater. Res. 68A (2003) 133–141. [30] A.E. Porter, N. Patel, J.N. Skepper, S.M. Best, W. Bonfield, Biomaterials 25 (2004) 3303–3314. [31] A.E. Porter, N. Patel, R. Brooks, S.M. Best, N. Rushton, W. Bonfield, J. Mater. Sci. Mater. Med. 16 (2005) 1–9. [32] I.R. Gibson, S.M. Best, W. Bonfield, J. Biomed. Mater. Res. 44 (1999) 422–428. [33] Q. Qiu, P. Vincent, B. Lowenberg, M. Sayer, J.E. Davies, Cell Mater. 3 (1993) 351–360. [34] S.D. Langstaff, M. Sayer, T.J.N. Smith, S.M. Pugh, S.A.M. Hesp, W.T. Thompson, Biomaterials 20 (1999) 1724–1741. [35] S.D. Langstaff, M. Sayer, T.J.N. Smith, S.M. Pugh, Biomaterials 22 (2001) 135–150. [36] E.S. Thian, J. Huang, S.M. Best, Z.H. Barber, W. Bonfield, Mater. Sci. Eng. C 27 (2007) 251–256. [37] E.S. Thian, J. Huang, S.M. Best, Z.H. Barber, W. Bonfield, Biomaterials 26 (2005) 2947–2956. [38] E.S. Thian, J. Huang, S.M. Best, Z.H. Barber, R.A. Brooks, N. Rushton, W. Bonfield, Biomaterials 27 (2006) 2692–2698.
[39] A. Oyane, H.M. Kim, T. Furuya, T. Kokubo, T. Miyazaki, T. Nakamura, J. Biomed. Mater. Res. 65A (2003) 188–195. [40] Y.L. Yang, S.R. Paital, N.B. Dahotre, J. Mater. Sci. Mater. Med. 21 (2010) 2511–2521. [41] Y.S. Yoo, H. Kim, D.Y. Kim, J. Eur. Ceram. Soc. 17 (1997) 805–811. [42] S. Hussaina, M. Anis, A. Maqsooda, M.S. Awan, J. Cryst. Growth. 297 (2006) 403–410. [43] F. Lusquinos, J. Pou, J.L. Arias, M. Boutinguiza, B. Leon, M. Perez-Amor, F.C.M. Driessens, J.C. Merry, I. Gibson, S. Best, W. Bonfield, J. Appl. Phys. 90 (2001) 4231–4236. [44] M. Roy, V.K. Balla, A. Bandopadhyay, S. Bose, Acta Biomater. 4 (2008) 324–333. [45] A. Ochsenbein, C. Feng, S. Winter, M. Traisnel, J. Breme, H.F. Hildebrand, Acta Biomater. 4 (2008) 1506–1517. [46] T.J. Webster, R.W. Siegel, R. Bizios, Biomaterials 20 (1999) 1231–1237. [47] Z. Guo, J. Chang, J. Eur. Ceram. Soc. 24 (2004) 93–99. [48] M.A. Zubair, C. Leach, J. Eur. Ceram. Soc. 30 (2010) 107–112. [49] K.S. Guan, B.J. Lu, Y.S. Yin, Surf. Coat. Technol. 173 (2003) 219–223. [50] Y.L. Yang, S.R. Paital, N.B. Dahotre, Mater. Technol. Adv. Perform. Mater. 25 (2010) 137–142. [51] L. Hao, T.H. Wang, J. Lawrence, L. Li, Appl. Surf. Sci. 247 (2005) 602–606. [52] Feng, J. Weng, B.C. Yang, S.X. Qu, X.D. Zhang, Biomaterials 24 (2003) 4663–46670. [53] S. Zou, D. Ireland, R.A. Brooks, N. Rushton, S. Best, J. Biomed. Mater. Res. B, Appl. Biomater. 90B (2008) 123–130. [54] C.M. Botelho, M.A. Lopes, I.R. Gibson, S.M. Best, J.D. Santos, J. Mater. Sci. Mater. Med. 13 (2002) 1123–1127.
Contents lists available at ScienceDirect
Materials Science and Engineering C j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m s e c
Osteoblast interaction with laser cladded HA and SiO2-HA coatings on Ti–6Al–4V Yuling Yang a, b, Kaan Serpersu b, Wei He b, c,⁎, Sameer R. Paital b, Narendra B. Dahotre d, 1 a
Department of Physics, Northeastern University, Shenyang 110004, China Department of Materials Science and Engineering, The University of Tennessee, Knoxville, TN 37996, USA c Department of Mechanical, Aerospace and Biomedical Engineering, The University of Tennessee, Knoxville, TN 37996, USA d Department of Materials Science and Engineering, University of North Texas, Denton, TX 76207, USA b
a r t i c l e
i n f o
Article history: Received 30 August 2010 Received in revised form 22 June 2011 Accepted 19 July 2011 Available online 27 July 2011 Keywords: Laser cladding HA/SiO2-HA coating Titanium Osteoblast
a b s t r a c t In order to improve the bioactivity and biocompatibility of titanium endosseous implants, the morphology and composition of the surfaces were modified. Polished Ti–6Al–4V substrates were coated by a laser cladding process with different precursors: 100 wt.% HA and 25 wt.% SiO2-HA. X-ray diffraction of the laser processed samples showed the presence of CaTiO3, Ca3(PO4)2, and Ca2SiO4 phases within the coatings. From in vitro studies, it was observed that compared to the unmodified substrate all laser cladded samples presented improved cellular interactions and bioactivity. The samples processed with 25 wt.% SiO2-HA precursor showed a significantly higher HA precipitation after immersion in simulated body fluid than 100 wt.% HA precursor and titanium substrates. The in vitro biocompatibility of the laser cladded coatings and titanium substrate was investigated by culturing of mouse MC3T3-E1 pre-osteoblast cell line and analyzing the cell viability, cell proliferation, and cell morphology. A significantly higher cell attachment and proliferation rate were observed for both laser cladded 100 wt.% HA and 25 wt.% SiO2-HA samples. Compared to 100 wt.% HA sample, 25 wt.% SiO2-HA samples presented a slightly improved cellular interaction due to the addition of SiO2. The staining of the actin filaments showed that the laser cladded samples induced a normal cytoskeleton and well-developed focal adhesion contacts. Scanning electron microscopic image of the cell cultured samples revealed better cell attachment and spreading for 25 wt.% SiO2-HA and 100 wt.% HA coatings than titanium substrate. These results suggest that the laser cladding process improves the bioactivity and biocompatibility of titanium. The observed biological improvements are mainly due to the coating induced changes in surface chemistry and surface morphology. © 2011 Elsevier B.V. All rights reserved.
1. Introduction To date, a number of materials have been successfully used as implants for hard tissue replacement such as bone. Among them one class of materials has the ability to integrate with host tissue and promote strong bonding at the tissue–material interface. This genre of materials has been dubbed as bioactive materials. In contrast, the other class of materials that remain completely neutral without any toxicity or reaction with the surrounding tissue is called the bioinert materials. Titanium and its alloys, especially Ti–6Al–4V, are the most commonly used bioinert materials for both dental and orthopedic applications [1,2], owing to their excellent mechanical properties, biocompatibility, corrosion resistance, and tissue compatibility. As bio-implants, they adsorb protein from surrounding biological fluid to
⁎ Corresponding author at: 209 Dougherty Engineering Bldg., Department of Materials Science and Engineering, The University of Tennessee, Knoxville, TN 37996, USA. Tel.: + 1 865 974 5275; fax: + 1 865 974 4115. E-mail address: [email protected] (W. He). 1 Formerly at Department of Materials Science and Engineering, The University of Tennessee, USA. 0928-4931/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.msec.2011.07.009
create a protein layer that will support cell growth [3]. Titanium and Ti–6Al–4V being completely neutral also make them the ideal choice for patients who may develop toxic reactions to other metal alloys. Although, titanium is bioinert, an oxide layer forms almost immediately on the surface interacting with the surrounding biological fluid. This naturally forming oxide layer occurs as a result of titanium high reactivity with oxygen [4]. Due to its thin layer, and amorphous and porous structures, the oxide layer possesses a risk of dissolving titanium ions into the body plasma and therefore is not a sound barrier for such dissolutions [5]. However, the field of applications can be expanded if the bioactivity and biocompatibility of the titanium implants are improved through a bioceramic coating. The techniques most commonly used to provide a bioceramic coating on the titanium implants are dip coating [6], sol–gel [7], plasma-spraying [8–10], simultaneous vapor deposition [11], and laser deposition [12] procedures. The criteria for judging the success of these coating processes are whether they can achieve a high crystallinity in the coatings, good adherence between the coating and the metal substrate, control over coating thickness and the ability to coat porous and complex-shaped implants. Of these methods, laser cladding (LC) has gained extensive use due to its ability to achieve appropriate surface textures and
1644
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
surface chemistry and thereby improve the biocompatibility of metallic surfaces at the tissue–implant interface [13–16]. Further, the LC technique has also demonstrated distinct advantages which can be listed as follows [17]: (1) localized heating which reduces thermal distortion and the size of the heat-affected zone; (2) controlled levels of dilution; (3) controlled shape of the clad within certain limits; (4) controlled surface roughness; (5) fine microstructures and metallurgical bonding; (6) high deposition rate and flexibility of the process. Therefore, LC is justifiably becoming a popular choice as a method for modifying surface properties [18–20], and lasers have been used to produce ceramic coatings on metallic materials for biomedical applications. In this way, the benefits of the bioceramics are combined with the mechanical performance of titanium implants. The most common examples of this type of combination are titanium coated with hydroxyapatite (HA) [8,9,11], and titanium with pseudowollastonite [21]. HA (chemical formula Ca10(PO4)6(OH)2) due to its chemical and crystallographic resemblance to bone and tooth minerals [22], has been widely studied as an artificial bone and tooth replacement material in orthopedic and dental implants. They are most commonly used as a coating of hard tissue implants, bone fillers, and for drug delivery. However, HA coated implants have the disadvantage that, in comparison with bioactive glasses and glass ceramics, the rate of bonding of bone tissue with bone implants is relatively low [23] and this has implications for prolonging the time required for patient recovery. Approaches toward improving the integration rates of HA with bone have included the incorporation of biological entities such as growth factors, proteins, and cells into the HA implant [24,25]. As an alternative to improve the biocompatibility of the implant, HA may also be chemically doped with small amounts of elements which are commonly found in physiological bone [26]. It was reported that trace elements of silicon and titanium in calcium HA ceramics and coatings can influence both the biological response of implant materials [27], and the crystallographic, mechanical and chemical properties of manufactured ceramics and coatings [28]. Best and co-workers have put their efforts in studying the development of silicate-substituted HA (Si-HA) [29–31], and demonstrated a significant increase in the amount of bone apposition and organization around Si-HA implants, illustrating their potential as bone graft materials [32]. Other researchers have reported the results of firing a stoichiometric HA precipitate to which SiO2 has been added [33–35]. In these reports, SiO2 was incorporated either by diffusion from an adjacent substrate [33] or by incorporation of SiO2 into the precipitate using the thermal decomposition of a metallorganic additive [34,35]. There are also several reports by Thian and co-workers [36–38] where they have synthesized biocompatible silicon substituted hydroxyapatite composite coatings on Ti substrates by a magnetron co-sputtering technique. The authors reported that the presence of a thin Si-HA coating synthesized by this process stimulated the osteoblast attachment, proliferation and differentiation. So far, few published papers have reported laser cladding of SiO2-HA on metallic alloys and corresponding bone cell response to such coatings. In the present study, HA and SiO2-HA coatings were prepared on Ti–6Al–4V substrates using a laser cladding (LC) technique. Here, a highly intense laser beam was used to melt the precursor (HA and SiO2-HA) and the surface of the Ti–6Al–4V substrate to achieve a micro-textured, multi-phase coating as well as metallurgical bonding at the interface. The in vitro bioactivity was assessed by immersing the samples in a simulated body fluid (SBF). The in vitro biocompatibility, including cell attachment, cell viability, and cell proliferation was investigated by culturing of mouse MC3T3-E1 pre-osteoblast cell line.
2. Materials and methods 2.1. Precursor materials and laser cladding Ti alloy (Ti–6Al–4V) plates, cut from the rolled sheets with a thickness of 3 mm and dimensions of 50 × 50 mm 2 were used as substrates. They were polished using 30 μm grit silicon carbide (SiC) emery paper and sequentially rinsed with acetone to obtain a clean surface free from rust and grease. The HA (Ca10(PO4)6(OH)2) and silica (SiO2) powders, obtained from Fisher Scientific, were used as the precursor materials. The HA and SiO2 precursor powders had a spherical morphology with a unimodal distribution in the 10–30 μm range. The precursors (25 wt.% SiO2-HA or 100 wt.% HA) were mixed in a water-based organic solvent (LISI W 15853) obtained from Warren Paint and Color Company (Nashville, TN, USA) and mechanically stirred for 25 min to get a viscous slurry. The slurry was then sprayed onto the polished and clean substrate coupons using an air pressurized spray gun. The sprayed coupons were dried in air to remove moisture. The thickness of the precursor deposit was maintained at 80 μm for all samples. The laser source employed in this work was a pulsed Nd:YAG laser equipped with a fiber optic beam delivery system that operates in the infrared region with a wavelength of 1064 nm. A high power laser radiation obtained from the above source is directed to the precursor sprayed surface of the substrate and as the laser beam heats up the precursor and the surface of the substrate a molten pool is created on the metallic substrate. Rapid quenching of the molten pool takes place as the laser beam is scanned away from the irradiated area. The scanning is carried in a way such that overlapping (0.1 mm or ~ 11%) laser cladding tracks were laid to obtain a continuous coating. Thus, a coating with metallurgical bonding with the substrate is achieved. The laser processing parameters employed in the present work are listed in Table 1.
2.2. Bioactivity In vitro bioactivity studies were performed by soaking the samples in a simulated body fluid (SBF) solution with ionic composition similar to that of the human plasma. The SBF solution is prepared by dissolving the reagent grade chemicals in the following order: NaCl (8.026 g), NaHCO3 (0.352 g), KCl (0.225 g), K2HPO4·3H2O (0.230 g), MgCl2·6H2O (0.311 g), CaCl2 (0.293 g) and Na2SO4 (0.072 g) in distilled water (700 mL). The fluid was then buffered to pH 7.4 with tri-hydroxymethylaminomethane [(CH2OH)3CNH2] (6.063 g) and hydrochloric acid (1 M, 40 mL) at 37 °C [39]. To study the bioactivity, the in vitro assays were carried out for a set of three samples from each processing condition. The samples were soaked in 10 mL of SBF solution in plastic containers. Soaking periods were varied for 1, 3, 5 and 7 days. The solution was refreshed every 24 h to maintain a pH value of 7.4, and the temperature was maintained at 37 °C during the soaking course. The samples following SBF immersion were dried in a vacuum desiccator for more than 12 h to remove any entrapped water molecules. The weight of samples before and after immersion in the SBF solution was measured to evaluate the HA deposition rate in SBF. The weight change measurements were carried out using a microbalance.
Table 1 Laser parameters used in this study. Pulse width (ms) Pulse repetition (Hz) Pulse energy (J) Spot diameter (μm) Laser scan speed (cm/min) Line spacing (mm)
1.0 30 4.0 900 75 0.1
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
2.3. Surface characterization After laser cladding and SBF immersion experiments, the surface chemical compositions of the samples before and after SBF immersion were characterized using a Philips Norelco X-ray Diffractometer (XRD) with Cu-Kα radiation of wavelength 0.15418 nm. The XRD system was operated at 20 kV and 10 mA in a 2θ range of 20°–100° using a step size of 0.02° and a count time of 1 s. The morphological evolutions of the surface of the coatings were characterized using low magnification Leica optical microscope. The cross-sectional microstructure of the coatings and the morphological evolutions of the surface of coatings following immersion in SBF are characterized using a LEO 1525 scanning electron microscope (SEM). In the current work, the surface roughness measurements of the laser cladded coatings and untreated Ti–6Al–4V substrates were measured using a Mahr Federal profilometer. The profilometer is equipped with a stylus based tip which traverses the surface following surface irregularities. The motion of the stylus tip is recorded by a photoelectric cell and amplified. The stylus tip has a radius of 2 μm. During each measurement the tip traces a length of 5.6 mm and different roughness parameters like Ra (defined as the arithmetic average of all points of the profile also called the center line average height), Rz (arithmetic average of vertical distances between the highest peak and deepest valley within a sampling length) and Rmax (maximum individual roughness depth) were recorded. For the laser cladded coatings the roughness measurements were carried out both parallel and perpendicular to the laser track. A total of six random scans were recorded from different locations on each sample and their average and standard deviation values were reported. Water contact angle for the surfaces of Ti–6Al–4V, laser cladded 25 wt.% SiO2-HA, and 100 wt.% HA coating was measured by the static sessile droplet (3 μL and residence time of 15 s) method using a contact angle goniometer (Cheminstruments, Inc., Fairfield, Ohio).
1645
cells attached on the coated surfaces was expressed as percentage of Ti–6Al–4V control. To study actin cytoskeleton development and focal adhesion formation, immunofluorescent staining was performed. Briefly, 4 h after cell seeding, cells were fixed in a 4 wt.% paraformaldehyde solution for 20 min. After fixation, cells were washed twice with a wash buffer solution (0.05% Tween-20 in phosphate buffered saline (PBS)), then stored in a 0.1 vol.% Triton X-100 PBS solution for 5 min at RT. Blocking solution (1 wt.% bovine serum albumin (BSA) in PBS), was then added to the samples and kept at RT for 30 min. After removal of the blocking solution, samples underwent incubation in primary and secondary antibody solutions at RT. The primary dilute solution contained anti-vinculin (Chemicon, USA) in the blocking solution, and was performed for 1 h. Alexa Fluor 488 goat anti-mouse IgG1 secondary antibody and Alexa Fluor 594 conjugated phalloidin solution were then applied to the samples for 45 min. A DAPI stain for cell nuclei was applied before imaging. Fluorescent images were captured with a Nikon Eclipse Ti microscope (Nikon, Japan). 2.6. Cell proliferation Proliferation of the osteoblast-like cells on Ti–6Al–4V, 25 wt.% SiO2-HA and 100 wt.% HA samples was performed in 24-well plates. Cells were seeded at an initial density of 7.5 × 10 3 cells/cm 2 and incubated for up to 7 days. The relative cell growth rate for 25 wt.% SiO2-HA and 100 wt.% HA samples was expressed as a percentage of Ti–6Al–4V control. Following incubation, the WST-1 reagent was added to each of the wells in a 1 to 10 dilution (reagent to cell culture medium). Samples were then incubated for 4 h at 37 °C, after which, 100 μL of solution from each sample was transferred into a 96-well plate. The absorbance was recorded at 450 nm using a Wallac 1420 Victor 2 Multilabel Counter (Perkin Elmer, USA). 2.7. Cell morphology
2.4. Cell culture Cell culture assays were performed using the mouse pre-osteoblast MC3T3-E1 cell line (ATCC, Manassas, VA, USA). Cells were cultured in α-modified minimum essential medium (α-MEM, Gibco, USA) supplemented with 10% fetal bovine serum (FBS), and 2% penicillin/ streptomycin (P/S) at 37 °C in a 5% CO2 atmosphere. The culture medium was replaced every 3 days and confluent cells were trypsinized and replated (0.25% trypsin-EDTA, Invitrogen, USA) to maintain the cell line.
Scanning electron microscopy (SEM) was used to study cell morphology on the various substrates. After 1 day and 7 days of culture, the samples were fixed both by a primary fixation solution, 3% glutaraldehyde in 0.1 M cacodylate buffer, and a secondary fixation solution, 2% osmium tetroxide in 0.1 M cacodylate buffer, for 1 h each, at RT. Samples were then washed with buffer and dehydrated with a series of increasing concentration of ethanol (25%, 50%, 70%, 95%, and 100%), critical point dried, sputter-coated with gold, and examined with a LEO 1525 scanning electron microscope at an accelerating voltage of 5 kV.
2.5. Cell attachment 2.8. Statistical analysis Two days prior to cell seeding, designated wells in 24-well plates were coated with 12 wt.% poly(2-hydroxyethyl methacrylate) (polyHEMA) (Polysciences, Inc.) in ethanol solution. The nonadhesive polyHEMA coating ensures cells attach and grow exclusively on the samples instead of the surrounding areas on tissue culture plates. Samples (Ti–6Al–4V, 25 wt.% SiO2-HA and 100 wt.% HA coated Ti alloy) were then placed in the wells and sterilized under ultraviolet light in an enclosed biosafety cabinet for 12 h. Cell attachment and viability were both examined qualitatively by a live/dead staining assay and quantitatively by counting the number of attached cells. MC3T3-E1 cells were seeded at 1 × 10 4 cells/cm 2 density on all sterilized samples and incubated for 4 h at 37 °C. Samples were then stained via a live/dead assay containing 2 μM calcein AM and 4 μM ethidium homodimer-1 (EthD-1). After 20 min of incubation at room temperature (RT), samples were viewed using a Nikon Eclipse E600 fluorescence microscope (Nikon, Japan). Viable cells were stained with calcein AM (green), while non-viable cells were stained with EthD-1 (red). Cell numbers in three randomly chosen areas on the surface of each sample were counted quantitatively. The total number of live
All cell culture experiments were performed simultaneously on at least triplicate samples. Each experiment was repeated once. Data were reported as mean ± standard error of mean (n ≥ 3). Statistical comparisons were performed using Student's t-test (JMP, USA), while p b 0.05 was considered statistically significant. 3. Results 3.1. Surface characterization The surface morphology of the untreated Ti–6Al–4V and the coatings (100 wt.% HA and 25 wt.% SiO2-HA) before SBF immersion is illustrated in Fig. 1. It can be observed that by adding SiO2 surface features at finer length scales (Fig. 1c) are obtained as compared to the pure HA precursor (Fig. 1b). The cross-sectional microstructure of the coatings (100 wt.% HA and 25 wt.% SiO2-HA) presented as inset in Fig. 1b and c also clearly demonstrated the rough morphology of the coatings and the sound bonding between
1646
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
HA coatings) compared to Ti–6Al–4V control, as noticed from the data in Table 2. It can also be noted that the coatings have a higher surface roughness when measured perpendicular to the laser track compared to the measurements made parallel to the track. This is due to the ridges between two consecutive parallel laser tracks in one specific direction that increased the surface roughness in that direction compared to Ti–6Al–4V control. However, within the laser cladded coatings (100 wt.% HA and 25 wt.% SiO2-HA coatings) the surface roughness did not vary much thereby indicating no distinct relation with the varying composition of precursor material within the range of present work. Surface wettability was significantly different among the three samples. Water contact angle on Ti–6Al–4V, 100 wt.% HA, and 25 wt.% SiO2-HA is 59 ± 2, 53 ± 2, and 46 ± 1°, respectively. 3.2. Bioactivity
Fig. 1. Low magnification optical microscopic images of the surface texture of the (a) untreated Ti–6Al–4V and Ti–6Al–4V samples coated with (b) 100 wt.% HA and (c) 25 wt.% SiO2-HA. The insets in panel (b) and panel (c) correspond to the higher magnification SEM images of the cross-section of the respective coatings.
the coating and the substrate material. A substantial increase in surface roughness (measured in terms of Ra, Rz, and Rmax) was observed for laser cladded coatings (100 wt.% HA and 25 wt.% SiO2-
The in vitro bioactivity of the 100 wt.% HA and 25 wt.% SiO2-HA coatings and untreated Ti–6Al–4V control was assessed by the precipitation of an apatite (HA, [Ca10(PO4)6(OH)2]) like mineral layer on the surface following immersion in SBF. As stated earlier, HA is a natural mineral component of the human bone [22], hence, this layer provides the appropriate chemistry and acts as a bone bonding interface. SEM analysis of the surfaces of 100 wt.% HA and 25 wt.% SiO2-HA coatings and untreated Ti–6Al–4V after immersion in SBF for 7 days is presented in Fig. 2. The untreated Ti–6Al–4V (Fig. 2a) barely indicated any deposition or precipitation of HA phase on its surface. In contrast, the sample laser processed with 100 wt.% HA and 25 wt.% SiO2-HA precursors clearly demonstrated (Fig. 2b and c) a thick deposition of apatite like layer. XRD spectra collected on samples before SBF immersion and after 7 days of immersion time are presented in Fig. 3. Some minor phases, such as TiO and CaO were not marked in Fig. 3b. Fig. 3a indicated that, compared to bare Ti, some new phases, specifically CaTiO3, Ca3(PO4)2, Ca2SiO4, CaO, and TiO2, were formed within the coatings by laser cladding 25 wt.% SiO2-HA and 100 wt.% HA precursors on Ti–6Al–4V substrate. However, it is obvious that with the same set of laser processing parameters, different phases were evolved within the coatings for 25 wt.% SiO2-HA and 100 wt.% HA precursors. The main phases evolved by cladding 25 wt.% SiO2-HA precursor include CaTiO3, Ca3(PO4)2, Ca2SiO4, and minor phase TiO2, but for 100 wt.% HA precursor samples, TiO2 is the main phase together with some CaTiO3 and minor Ca3(PO4)2 and CaO. However, no HA was detected in any sample laser processed using the set of parameters employed in the present work. Compared to the XRD pattern before SBF immersion (Fig. 3a), it is observed that HA was precipitated on all the sample surfaces after 7 days of immersion in SBF (Fig. 3b). The characteristic peaks of crystalline HA-like structures indicate that very small amount of HA precipitated on Ti–6Al–4V sample after a 7-day immersion in SBF. On the contrary, 100 wt.% HA coated sample appeared to have more HA precipitation than the Ti alloy control. But for 25 wt.% SiO2-HA sample, the HA peaks observed in the XRD patterns after 7 days of SBF immersion are very strong and correspond to the characteristic HA peaks at 2θ ~22°, 25.6° and 31.7° (representing the planes (200), (002), and (112), respectively). Compared to Ti alloy control, both 100 wt.% HA and 25 wt.% SiO2-HA samples, have higher surface free
Table 2 Surface roughness and contact angle of the samples. Samples
Ti–6Al–4V 25 wt.% SiO2-HA 100 wt.% HA
Parallel to the laser track
Perpendicular to the laser track
Ra (μm)
Rz (μm)
Rmax (μm)
Ra (μm)
Rz (μm)
Rmax (μm)
– 2.6 ± 0.3 2.5 ± 0.3
– 15 ± 3 14 ± 1
– 18 ± 2 18 ± 2
– 3.3 ± 0.3 4.0 ± 0.4
– 16 ± 2 20 ± 2
– 22 ± 5 28 ± 4
Ra (μm)
Rz (μm)
Rmax (μm)
Contact angle (θH2O/degrees)
0.41 ± 0.04 – –
3.4 ± 0.3 – –
4.8 ± 0.6 – –
59 ± 2 46 ± 1 53 ± 2
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
1647
Fig. 3. XRD patterns of Ti–6Al–4V, 25 wt.% SiO2-HA and 100 wt.% HA coated sample (a) before, and (b) after immersion of 7 days in SBF.
Fig. 2. Low magnification SEM images of the surfaces of (a) untreated Ti–6Al–4V (b) 100 wt.% HA coated, and (c) 25 wt.% SiO2-HA coated samples following immersion in SBF for 7 days.
energy as calculated in our previous work [40], increased wettability, and possess more intense peaks of HA. In addition, a clear presence of the HA peak at 2θ = 45.4° corresponding to the plane (311) was detected in 25 wt.% SiO2-HA sample. Further, it is also worth noticing that for 25 wt.% SiO2-HA samples, after 7 days of immersion in SBF, the apatite peak (at 2θ ~22° and 31.7°) broadened with a reduction in intensity. Fig. 4 presents the weight increase in laser cladded samples and Ti–6Al–4V control following immersion in SBF for 1 day, 3, 5, and 7 days, respectively. It is obvious that all laser treated samples (25 wt.% SiO2-HA or 100 wt.% HA precursor) exhibited a higher weight increase than the bare Ti alloy over the entire range of immersion periods employed in the present work. Compared to the 100 wt.% HA coated sample, the 25 wt.% addition of SiO2 in the precursor leads to a higher increase in weight after all immersion times. Increase in weight for laser cladded samples after all immersion times might be attributed to the
Fig. 4. Weight increase as a function of SBF immersion time for 1 day, 3 days, 5 days, and 7 days.
1648
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
presence of new phases, such as CaTiO3, Ca3(PO4)2 and Ca2SiO4 within the coating. 3.3. Cell attachment Live/dead staining results (Fig. 5a–c) indicated that cells attached well on the substrates after 4 h of incubation with very few dead/ disrupted cells. This suggested good initial adhesion and high viability of cells exposed to modified surfaces. Quantitatively, the total number of cells attached after 4 h on the 25 wt.% SiO2-HA and 100 wt.% HA coated samples was significantly higher than the Ti–6Al–4V control (Fig. 5d). Actin cytoskeleton and focal adhesion play important roles in regulating the cell shape and maintaining cell adhesion. The actin labeling (Fig. 6) revealed similar cytoskeletal organization and focal adhesion distributions in cells grown on all samples after 4 h seeding. Most cells exhibited a polygonal morphology with good cytoskeletal development (red stains). Focal adhesion protein, vinculin (green) can be seen within the cytoplasm as well as at the tips of stretching cytoplasmic projections. 3.4. Cell proliferation The absorbance results of WST-1 assays used to study cell proliferation are shown in Fig. 7. The inset shows the proliferation rate with regard to untreated Ti–6Al–4V. It is clear that the absorbance for all the samples presented an increasing trend as a function of time, indicating cells were proliferating on all the tested substrates. Within all the time points studied, the cell proliferation rates on the laser cladded samples were comparable or even higher than the untreated Ti alloy control. Specifically, at day 1, proliferation rate on the 25 wt.% SiO2-HA coated sample (114.7 ± 2.2%) was comparable with the Ti
alloy control, but significantly lower than the 100 wt.% HA coated sample (127.7 ± 5.7%). At day 4, proliferation rates were comparable between 25 wt.% SiO2-HA and 100 wt.% HA samples and both were significantly higher than Ti alloy. At day 7, 25 wt.% SiO2-HA coated samples induced the highest proliferation rate of 173.5 ± 18.7%, while that for 100 wt.% HA coated sample is 148.7 ± 10.9%, and both were significantly higher than Ti alloy. 3.5. Cell morphology Cell morphology after 1 day and 7 days of proliferation was assessed by SEM, and resulting images are shown in Fig. 8. From Fig. 8c and e, it can be seen that the normal individual MC3T3-E1 osteoblasts on laser cladded sample (25 wt.% SiO2-HA and 100 wt.% HA) surfaces mostly had a triangular form, with an average size of around 20 μm in width and 60–80 μm in length. This exhibits that good adhesion was obtained by laser cladding 25 wt.% SiO2-HA and 100 wt.% HA precursors on Ti–6Al–4V substrate. Their good adhesion can be characterized by the presence of lamellipodia, i.e. large front projections, which strongly adhere to the substrate and draw the cellular body by cytoplasmic contraction. It is also found that several multiple microvilli exist on the surface of the cells, and fine cytoplasmic extensions in multiple directions were formed. Those cultured on 25 wt.% SiO2-HA and 100 wt.% HA samples revealed greater spreading with large lamellipodia, indicating active cell migration. On untreated Ti–6Al–4V sample, however, the osteoblasts showed a polygonal form (Fig. 8a). In addition, the numbers of filopodia and cytoplasmic extensions were also less on cells grown on bare Ti alloy. Fig. 8b, d and f illustrates the cell morphology after 7 days of proliferation. It is clear that after proliferation for 7 days, the cells distributed evenly and compactly on the sample surfaces, which is in accordance with the results of Fig. 7 and exhibits a proliferation
Fig. 5. Live/dead staining of MC3T3-E1 cells after 4 h of incubation on (a) Ti–6Al–4V, (b) 25 wt.% SiO2-HA coating, and (c) 100 wt.% HA coating; and (d) cell numbers attached to Ti control, 25 wt.% SiO2-HA coating, and 100 wt.% HA coating within observation areas after 4 h seeding; results are expressed as percentages of cells attached on Ti control (n = 3). * denotes that the cell number on untreated Ti–6Al–4V is significantly lower than laser cladded Ti (p b 0.05), scale bars are 100 μm.
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
1649
Fig. 7. Histograms of WST-1 assay results representing proliferation activities of MC3T3-E1 cells after 1 day, 4 and 7 days of growth; inset results are expressed as percentages of activities on bare Ti, * denotes that the absorbance for untreated Ti–6Al– 4V sample is significantly lower than that of laser cladded 25 wt.% SiO2-HA and 100 wt.% HA at day 1, day 4, and day 7 (p b 0.05). In inset, # denotes that at day 1, cell proliferation rate on 100 wt.% HA is significantly higher than the other two samples, and * denotes that at day 4 and day 7, cell proliferation rate on untreated Ti is significantly lower than the laser cladded samples (p b 0.05).
4. Discussion
Fig. 6. Fluorescent images for MC3T3-E1 cells adhered on (a) Ti–6Al–4V, (b) 25 wt.% SiO2-HA coating, and (c) 100 wt.% HA coating after 4 h, showing focal contacts (green), actin filaments (red), and the nucleus (blue). Scale bars in large images and in insets are 100 μm and 50 μm, respectively.
trend for all samples. The inset is the image with higher magnification indicating excellent cell spreading and cell–cell interaction after 7 days of proliferation.
To improve longevity and minimum failure of an orthopedic implant, one of the most challenging obstacles is to control osseointegration. Therefore, a variety of coatings, including oxides, bioglass and bioceramics on Ti–6Al–4V substrates have been developed by sol–gel, dip coating, and laser deposition processes. In order to examine how a coating affects its biocompatibility, one should investigate the behavior of cells on coated implant surface. Since osteoblasts, the bone forming cells, play a critical role in osseointegration of an orthopedic implant in vivo, the behavior of such cell type is commonly studied in vitro for the development of new orthopedic biomaterials or surface modifications of existing ones. The purpose of the present study was to investigate the interaction of osteoblasts with laser cladded HA and SiO2-HA coatings with varying chemical compositions and different surface energies. From Fig. 1 it can be observed that 25 wt.% SiO2-HA precursor resulted in a surface with finer microstructure as compared to 100 wt.% HA precursor. This significant change in the microstructure with the addition of SiO2 can be explained based on the work by Yoo [41] and Hussaina [42]. The authors reported that the addition of SiO2 can accelerate the material transfer and result in faster microstructural evolution [41], and also control the microstructure by suppressing abnormal grain growth [42]. Hence, it is clear from their results that SiO2 addition can reduce the crystallization time which in turn, can result in surface features at finer length scales. Further, as illustrated in Fig. 3a, no pure HA was observed after laser processing. The chemical and phase compositions of laser cladded 100 wt.% HA coating were different from that of 25 wt.% SiO2-HA coating. The reactions between the precursor and substrate material during processing using a high energy laser beam are important in predicting the phases evolved within the coating. In laser cladding process, the laser beam is focused on the surface of Ti alloy pre-pasted with precursors 100 wt.% HA and 25 wt.% SiO2-HA. The high energy of laser beam melts the precursor and the local portion of the substrate surface and creates a molten pool which is maintained during the laser irradiation. Due to the high temperature of laser beam, interactions between Ti alloy substrate and HA or SiO2-HA precursor can be induced. It was shown that multiple phases can be formed between these compositions during high temperature processing,
1650
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
Fig. 8. SEM images for cell morphology after 1 and 7 days of proliferation on Ti–6Al–4V (a and b), 25 wt.% SiO2-HA coating (c and d), and 100 wt.% HA coating (e and f). Scale bars in (a), (c) and (e) are 10 μm; and in (b), (d) and (f) are 100 μm. Scale bars in inset are 5 μm.
which can be characterized as Ca3TiO7, CaTiO3, Ca4Ti3O10 and Ca5Ti4O3 [20,43]. The possible reactions between HA and Ti can be characterized as follows [44]:
Besides the aforementioned equations, there is another possible reaction between SiO2-HA and Ti: 2CaO þ SiO2 ¼ Ca2 SiO4 ðΔG ¼ −122:17 kJ=molÞ:
Ti þ O2 ¼ TiO2 ðΔG ¼ −864:09 kJ=molÞ
ð1Þ
HA þ TiO2 ¼ CaTiO3 þ 3TCP þ H2 O ðΔG ¼ −7312:39 kJ=molÞ
ð2Þ
TCP þ TiO2 ¼ CaTiO3 þ α Ca2 P2 O7 ðΔG ¼ −41:51 kJ=molÞ
ð3Þ
α Ca2 P2 O7 þ Ti ¼ CaTiO3 þ CaO
ð4Þ
þ P2 O3 ðgÞ↑ ðΔG ¼ −1247:46 kJ=molÞ:
ð5Þ
All the above equations clearly suggest that the chemical compositions of the laser cladded coating depend on the precursor compositions and the reactions among them. Same phases such as TiO2, CaTiO3, and Ca3(PO4)2 (TCP) were detected both in laser cladded 100 wt.% HA and 25 wt.% SiO2-HA coatings based on-the aforementioned reactions (1)–(4). Due to the absence of SiO2 in 100 wt.% HA precursor, Eq. (5) will not occur, and instead it led to the retention of CaO in laser cladded 100 wt.% HA coating. On the contrary, in laser
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
cladded 25 wt.% SiO2-HA coating, reaction (5) occurred, and led to the formation of additional phase of Ca2SiO4. Numerous studies have shown that the roughness, and chemical composition of the surface, which in turn determine the surface static charge and the wettability (surface free energy), are the main factors affecting the cell activity on implants [45]. However, it is still unclear what kind of roles those surface characteristic parameters play, and how they act on each other. For instance, Webster [46] proposed that surface topography is the main factor influencing the cell response rather than the specific chemistry of a material surface. But the results of Anne Ochsenbein [45] are definitely inconsistent with that observation. In their work, they pointed out that the mutual action between different but interrelated surface characteristic parameters and their co-adjustment of cell responses is definitely more complicated and cannot be described by the surface structure alone. In the present work, we draw the same conclusion as Anne Ochsenbein. As illustrated in Table 2, although the laser cladded samples have similar surface roughness they possess distinctly different water contact angles, indicating that the chemical phase has strongly influenced the wettability, and likely to play an important role with respect to protein absorption, cell attachment and spreading. As stated earlier, the major difference in chemical composition between 25 wt.% SiO2-HA and 100 wt.% HA samples is the presence of Ca2SiO4. We, therefore, deduced that it is because of the formation of Ca2SiO4 that leads to a lower contact angle (Table 2) and improved bioactivity (more HA formation and higher weight increase in SBF) for 25 wt.% SiO2-HA than that of 100 wt.% HA coating. The work by Gou et al. showed that Ca2SiO4 possessed good bioactivity in SBF environment [47]. This is because the addition of SiO2 in HA can develop the grain boundaries and reduce the grain size [48,49], which subsequently improves the wettability of the surface. The detailed discussion has been reported in our previous work [50]. Also as previously described, for 25 wt.% SiO2-HA samples, after 7 days of immersion in SBF, the apatite peak (at 2θ ~ 22° and 31.7°) broadened with reduction in intensity. This indicates that there is a transformation from a highly crystallized HA phase to a near amorphous or fine crystallite sized HA phase. This rapid transformation can again be attributed to the high biomineralization rate owing to the improved wettability as a result of phases such as Ca2SiO4. Cell attachment and viability results show that laser cladding HA and SiO2-HA can increase the number of cells adhering to the substrate for increased cell viability. This result is in accordance with the work by Hao and Feng [51,52]. Similar results were observed from cell proliferation experiments. A significantly higher proliferation rate was achieved by laser cladding process compared to untreated Ti alloy substrate. Moreover, after laser cladding process, the cultured samples revealed greater spreading with large lamellipodia, indicating active cell migration. These results can be attributed to the presence of new phases (Ca3(PO4)2, CaTiO3 and Ca2SiO4), and three dimensional surface texture which in turn increase the surface energy and, subsequently, the wettability of the surface. In our previous work [40], surface free energy has been calculated, and the values for Ti– 6Al–4V, 100 wt.% HA and 25 wt.% SiO2-HA are 34.19 ± 0.05, 49.47 ± 0.18, and 56.20 ± 0.97 mJ/m 2, respectively. It was reported that surfaces with a higher surface free energy are more adhesive toward cell binding proteins and bone cells than those with a low surface free energy [46]. Comparing the results of 25 wt.% SiO2-HA coating to those of 100 wt.% HA coating, it is obvious that the cell viability and proliferation rate for 25 wt.% SiO2-HA coating are slightly higher due to the addition of SiO2 in the precursor. The results are in accordance with the work by Zou et al. [53], in which they stated that silicon substitution into hydroxyapatite was observed to affect human osteoblast (HOB) cell adhesion and different populations of cells respond differently to the silicate supplement. Further Si substitution can improve cell adhesion and proliferation was also given by Botelho
1651
C.M. et al. [54]. In their work, the authors reported that surface charge on Si-HA is different from that on pure HA and thereby affects cell adhesion. As already shown in multiple previous investigations on laser cladding 100 wt.% HA and 25 wt.% SiO2-HA coatings and untreated Ti– 6Al–4V substrates with different chemical compositions, further evidence was provided here that the bioactivity and biocompatibility were improved by the laser cladding process. The wettability derived from chemical composition is considered as the predominant parameter influencing the interactions at the interface of the coating and the bio system. 5. Conclusions Titanium surface treatments by laser cladding 100 wt.% HA and 25 wt.% SiO2-HA coatings have been performed to optimize the cell interaction with the substrate. XRD results showed that new phases were formed by laser cladding process, which subsequently improved the wettability of the surface. In biological tests, all laser cladded samples presented good but differential cell reactions. 25 wt.% SiO2HA samples showed a significantly higher HA precipitation rate after immersion in SBF. A significant higher cell attachment and proliferation rate were observed for both laser cladded 100 wt.% HA and 25 wt.% SiO2-HA samples. Compared to 100 wt.% HA sample, 25 wt.% SiO2-HA samples presented a slightly higher cell interaction due to the addition of SiO2. The observed biological improvements for the laser cladded samples are therefore due to the change in surface morphology and the chemical composition of the surface coatings. Acknowledgment Support to Yuling Yang during this work at the University of Tennessee by the National Science Foundation of China for Young Scholars (grant no. 50801012), the Fundamental Research Funds for the Central University (grant no. N100405001), and the Science Foundation of Liaoning Province (grant no. 20102072) is highly acknowledged. Yuling Yang also thanks China Scholarship Council (CSC) and Northeastern University (NEU) of China for providing financial support as a visiting scholar at the University of Tennessee. The authors thank Ms. Lu Huang for her help with the cell culture. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
T.M. Lee, E. Chang, C.Y. Yang, Biomaterials 25 (2004) 23–32. A. Goransson, E. Janssn, P. Tengvall, A. Wennerberg, Biomaterials 24 (2003) 197–205. N. Mirhosseini, P.L. Crouse, L. Li, D. Garrod, Appl. Surf. Sci. 253 (2003) 7798–8002. D.M. Brunette, Titanium in Medicine, Springer, London, 2001. Toth, G. Szabo, L. Kovacs, K. Vargha, J. Barabas, Z. Nemeth, Smart Mater. Struct. 11 (2002) 813–818. T. Li, J. Lee, T. Kobayashi, H. Aoki, J. Mater. Sci. Mater. Med. 7 (1996) 355–357. A. Milev, G.S.K. Kannangara, B. Ben-Nissan, Mater. Lett. 57 (2003) 1960–1965. S.T. Ding, C.P. Ju, J.H. Chern Lin, J. Mater. Sci. Mater. Med. 11 (2000) 183–190. K.A. Khor, A. Vreeling, Z.L. Dong, P. Cheang, Mater. Sci. Eng. A 266 (1999) 1–7. X. Liu, C. Ding, Z. Wang, Biomaterials 22 (2001) 2007–2012. L. Pawlowski, J. Therm. Spray Technol. 8 (1999) 279–294. J.M. Fernandez-Pradas, P. Serra, J.L. Morenza, P.N. De Aza, Biomaterials 23 (2002) 2057–2061. A. Kurella, N.B. Dahotre, J. Biomater. Appl. 20 (2005) 5–50. S.R. Paital, Z. Cao, W. He, N.B. Dahotre, Biofabrication 2 (2010) 1–14. S.R. Paital, W. He, N.B. Dahotre, J. Mater. Sci. Mater. Med. 21 (2010) 2187–2200. S.R. Paital, W. He, C. Daniel, N.B. Dahotre, J. Mater. 62 (2010) 76–83. W.M. Steen, Laser Material Processing, Springer-Verlag, London, 2003. F. Lusquiños, J. Pou, M. Boutinguiza, F. Quintero, R. Soto, B. León, M. Pérez-Amor, Appl. Surf. Sci. 247 (2005) 486–492. D. Wang, C. Chen, J. Ma, T. Lei, Appl. Surf. Sci. 253 (2007) 4016–4020. Y.C. Wang, Y.M. Li, H.L. Yu, J. Ding, X.H. Tang, J.G. Li, H.Z. Yao, Surf. Coat. Technol. 200 (2005) 2080–2084. P.N. De Aza, J.M. Fernandez-Pradas, P. Serra, Biomaterials 25 (2004) 1983–1990. L.L. Hench, J. Wilson, An Introduction to Bioceramics, World Scientific, London, 1993. H. Oonishi, L.L. Hench, J. Wilson, F. Suginara, E. Tsuji, S. Kushitani, H. Iwaki, J. Biomed. Mater. Res. 44 (1999) 31–43.
1652
Y. Yang et al. / Materials Science and Engineering C 31 (2011) 1643–1652
[24] M.H. Mankani, S.A. Kuznetsov, B. Shannon, R.K. Nalla, R.O. Ritchie, Y. Qin, P.G. Robey, Am. J. Pathol. 168 (2006) 542–550. [25] M. Mastrogiacomo, A. Muraglia, V. Komlev, F. Peyrin, F. Rustichelli, A. Crovace, R. Cancedda, Orthod. Craniofac. Res. 894 (2005) 277–284. [26] F.C.M. Driessens, Bull. Soc. Chim. Belg. 89 (1980) 663–689. [27] L.L. Hench, J. Am. Ceram. Soc. 74 (1991) 1487–1510. [28] J.C. Elliott, Structure and Chemistry of the Apatites and Other Calcium Orthophosphates, 18, Elsevier, Amsterdam, 1994. [29] A.E. Porter, S.M. Best, W. Bonfield, J. Biomed. Mater. Res. 68A (2003) 133–141. [30] A.E. Porter, N. Patel, J.N. Skepper, S.M. Best, W. Bonfield, Biomaterials 25 (2004) 3303–3314. [31] A.E. Porter, N. Patel, R. Brooks, S.M. Best, N. Rushton, W. Bonfield, J. Mater. Sci. Mater. Med. 16 (2005) 1–9. [32] I.R. Gibson, S.M. Best, W. Bonfield, J. Biomed. Mater. Res. 44 (1999) 422–428. [33] Q. Qiu, P. Vincent, B. Lowenberg, M. Sayer, J.E. Davies, Cell Mater. 3 (1993) 351–360. [34] S.D. Langstaff, M. Sayer, T.J.N. Smith, S.M. Pugh, S.A.M. Hesp, W.T. Thompson, Biomaterials 20 (1999) 1724–1741. [35] S.D. Langstaff, M. Sayer, T.J.N. Smith, S.M. Pugh, Biomaterials 22 (2001) 135–150. [36] E.S. Thian, J. Huang, S.M. Best, Z.H. Barber, W. Bonfield, Mater. Sci. Eng. C 27 (2007) 251–256. [37] E.S. Thian, J. Huang, S.M. Best, Z.H. Barber, W. Bonfield, Biomaterials 26 (2005) 2947–2956. [38] E.S. Thian, J. Huang, S.M. Best, Z.H. Barber, R.A. Brooks, N. Rushton, W. Bonfield, Biomaterials 27 (2006) 2692–2698.
[39] A. Oyane, H.M. Kim, T. Furuya, T. Kokubo, T. Miyazaki, T. Nakamura, J. Biomed. Mater. Res. 65A (2003) 188–195. [40] Y.L. Yang, S.R. Paital, N.B. Dahotre, J. Mater. Sci. Mater. Med. 21 (2010) 2511–2521. [41] Y.S. Yoo, H. Kim, D.Y. Kim, J. Eur. Ceram. Soc. 17 (1997) 805–811. [42] S. Hussaina, M. Anis, A. Maqsooda, M.S. Awan, J. Cryst. Growth. 297 (2006) 403–410. [43] F. Lusquinos, J. Pou, J.L. Arias, M. Boutinguiza, B. Leon, M. Perez-Amor, F.C.M. Driessens, J.C. Merry, I. Gibson, S. Best, W. Bonfield, J. Appl. Phys. 90 (2001) 4231–4236. [44] M. Roy, V.K. Balla, A. Bandopadhyay, S. Bose, Acta Biomater. 4 (2008) 324–333. [45] A. Ochsenbein, C. Feng, S. Winter, M. Traisnel, J. Breme, H.F. Hildebrand, Acta Biomater. 4 (2008) 1506–1517. [46] T.J. Webster, R.W. Siegel, R. Bizios, Biomaterials 20 (1999) 1231–1237. [47] Z. Guo, J. Chang, J. Eur. Ceram. Soc. 24 (2004) 93–99. [48] M.A. Zubair, C. Leach, J. Eur. Ceram. Soc. 30 (2010) 107–112. [49] K.S. Guan, B.J. Lu, Y.S. Yin, Surf. Coat. Technol. 173 (2003) 219–223. [50] Y.L. Yang, S.R. Paital, N.B. Dahotre, Mater. Technol. Adv. Perform. Mater. 25 (2010) 137–142. [51] L. Hao, T.H. Wang, J. Lawrence, L. Li, Appl. Surf. Sci. 247 (2005) 602–606. [52] Feng, J. Weng, B.C. Yang, S.X. Qu, X.D. Zhang, Biomaterials 24 (2003) 4663–46670. [53] S. Zou, D. Ireland, R.A. Brooks, N. Rushton, S. Best, J. Biomed. Mater. Res. B, Appl. Biomater. 90B (2008) 123–130. [54] C.M. Botelho, M.A. Lopes, I.R. Gibson, S.M. Best, J.D. Santos, J. Mater. Sci. Mater. Med. 13 (2002) 1123–1127.