1 Introduction

Calcium phosphate ceramics particularly hydroxyapatite has gained much interest among material scientists as well as medical practitioners as synthetic bone substitutes. Even though these are already being used as fillers for treating various bone defects, their application is limited in non-load bearing uses due to brittle fracture of ceramic components [1–6]. On the other hand, because of the better physico-chemical, mechanical and biological properties, titanium and its alloys are extensively used for load bearing applications such as bone plates, screws, and artificial joints. A number of titanium based implants with bioactive ceramic coating around the bone contacting areas are available in the market. This is to accelerate the osteo-integration, which help the patient to faster rehabilitation. An adherent bioactive ceramic coating was preferred to have a faster bone–implant bonding, there by reducing the post implantation healing time which provides long term in vivo functionality [7–13]. Among the various techniques such as sol–gel coatings, electro deposition, dip coating, ion beam and various other vapor deposition processes, plasma spraying is the commercially viable technique for clinical application. Despite its ability to produce various bioactive ceramic coating with high coating adhesion strength, there are many demerits for plasma spray such as variable crystalline nature and chemical inhomogeneity in coating [14–16].

Pulsed Laser Deposition (PLD) technique has been evolved as an alternative to aforesaid methods and has an added advantage of preservation of stoichiometry of target phase. PLD technique also possesses the ability to form desired film thickness, morphology and composition by varying the deposition parameters. This method additionally provides deposition of different target materials of unique physico-chemical and biological properties over single substrate for functionally graded coatings [16–20]. Many in vivo studies have reported that HA-coated Ti had enhanced implant fixation within a period of 3–6 months [9–13]. However, certain difficulties such as maintenance of phase purity and mechanical strength in vivo, have been encountered in many of the coating techniques including the plasma-sprayed coatings. Titanium has also shown to possess inherent biocompatibility as it forms a thin TiO2 layer of several nanometres which easily forms under normal atmospheric conditions [21, 22]. Reports have shown that an interlayer of titania (TiO2) formed by methods like sol–gel can improve the coating adhesion between the HA and Ti substrate [22–24]. In the present work an attempt has been made to improve the mechanical property of hydroxyapatite coating on titanium substrate by incorporating a titania interlayer by making use of PLD for multi-target deposition, since the deposited interlayer helps to improve the coating adhesion strength as well as maintains the target stoichiometry.

2 Materials and methods

2.1 Preparation of target materials

2.1.1 Preparation of dense HA target

Hydroxyapatite (HA) powder was synthesized by a wet precipitation method involving calcium nitrate and ammonium dihydrogen phosphate (Rankem, India) in stoichiometric proportions at a pH of 11 at 80°C [25]. The precipitate was aged in the mother liquor for 24 h, washed with distilled water to get rid of surface impurities such as nitrates and ammonium ions. The filter cake obtained was freeze-dried, calcined at 300°C for 2 h, pulverised and sieved to get HA particles having a uniform size in the range of 125 μm. HA powder was then compacted to 30 mm diameter and 5 mm thick discs by isostatic pressing at a pressure of 80 MPa (Cold Isostatic Press, EPSI, Belgium). The discs were sintered at 1100°C for 2 h. The density of the sintered disc was measured by Archimedes method and was found to be 3.07 g/cm3. Sintered discs were mechanically polished (Buehler-Ecomet 3 Variable speed Grinder-Polisher, Buehler Ltd, U.S.A) with a series of SiC papers having grit size ranging from 320 to 800 to get uniform surface. The discs were then cleaned ultrasonically in distilled water, rectified spirit and finally in acetone. The cleaned substrates were then dried by blowing hot air around 70°C.

2.1.2 Preparation of titania (TiO2—rutile) target

Commercial Titanium dioxide (Rutile) (SD Fine Chem. Ltd., India) powder was compacted by isostatic press similar to that of HA but by adding 5% poly vinyl alcohol as binder. TiO2 discs were made of the same dimension as that of HA disc. The compacted discs were sintered at 1230°C for 10 h. Surface was then polished and cleaned as stated for HA discs.

2.2 Preparation of substrate material—titanium disc

Commercially available Ti6Al4V sheet (ASTM F1108, Manhar Metal Supply Corporation, Mumbai, India) were machined into 15 mm × 20 mm size and were mechanically polished through a series of SiC papers to have uniform surface. These substrates are cleaned ultrasonically in distilled water, rectified spirit and finally in acetone.

2.3 Pulsed laser deposition

A custom designed pulsed laser deposition system was used for the present study (fabricated by Excel Instruments, Mumbai, India). The system consists of a stainless steel vacuum chamber evacuated by a turbo molecular pump backed with a rotary vane pump. A pirani and cold cathode gauge were used to measure the total pressure of the vacuum chamber. Chamber was also equipped with high temperature substrate holder where the substrate can be heated up to a maximum of 800°C which is programmed with the attached temperature controller and readout unit. The deposition system utilized high power Q-switched Nd-YAG laser (Quanta System S.P.A, Italy). In these experiments the radiation of the third (355 nm) harmonics with a repetition rate of 10 Hz and pulse width of 6–8 ns were used. The output laser beam of 2 W power was focussed to a spot size of about 2 mm with a spherical lens on to the HA target fixed at an angle of 45o with respect to incident beam. The target and substrate holder were set parallel and the distance between them was fixed at 35 mm. Before deposition the chamber was evacuated to a pressure of the order of 10−6 mbar and then filled with oxygen to an optimized working gas pressure of 1 mbar for the deposition of TiO2 and 1 × 10−3 mbar for the deposition of HA. The oxygen pressure was controlled by a needle valve. The targets were rotated by means of attached DC motor during deposition so as to eliminate depletion of material at one point and there by maintains a uniform fresh surface for each incoming laser pulse. The substrate temperature was previously optimized for better coating properties and kept at 400°C. Different targets were ablated for different time periods depending on the deposition parameters and to have different coating thickness. HA coating with and without TiO2 were then analysed for various physico-chemical and mechanical characteristics.

2.4 Material characterization

2.4.1 Profilometry

The surface roughness of Ti6Al4V substrate, coating thickness of TiO2 layer and that of TiO2-HA coating on titanium substrate were measured by profilometry technique. The surface profile was obtained using Talysurf CLI 1000 (Taylor Hobson, UK) with the software Talymap Gold (version 4.1). A surface of 15 mm × 10 mm containing both coated and uncoated region was scanned at 50 μm/s speed, 0.5 μm spacing, and 100 Hz sampling rate. The thickness of the coating was also determined by step height measurement.

2.4.2 X-ray diffraction analysis

The phase and crystallinity analysis of both targets and the deposited films were done by X-ray diffraction technique. The samples were scanned using Cu Kα radiation at a voltage of 40 kV and a current of 30 mA (Siemens D-5005 X-ray Diffractometer, Germany) in the 2θ range 10–70° in the case of targets and 20–50° for the coated specimens to avoid intense substrate peaks.

2.4.3 Fourier transform infrared spectroscopy (FTIR)

FTIR analysis was performed on a Thermo Nicolet 5700 spectrometer (USA) and the spectra were collected in the diffuse reflectance (DRIFT) mode. Spectra of targets and the coating obtained were compared. Samples were prepared by mixing the target powder with optical grade KBr powder; while pure KBr was used as the background. The samples of the coating were collected by scratching the surface. The spectra were recorded at an average number of 64 scans between 400 and 4000 cm−1 and at a resolution of 4 cm−1.

2.4.4 Scanning electron microscopy and energy-dispersive X-ray analysis (SEM–EDS)

The coatings obtained were compared with the targets for the elemental composition and microstructural morphology by an environmental scanning electron microscope (ESEM-Quanta 200, Germany) equipped with an energy-dispersive X-ray analysis device (EDAX; OXFORD, X-ray microanalysis software). SEM was also used to examine the failure mode of HA coating after performing the scratch test. The samples were examined as such at low vacuum mode without any coating.

2.4.5 Inductively coupled plasma-optical emission spectrometry (ICP-OES)

The elemental analysis for determining Ca/P ratio of the HA target as well as the coating was done by an ICP-OES (Perkin Elmer 5300 DV, USA). Sample solutions were prepared by dissolving in 5% HNO3. The analysis was carried out with a calibration plot of known standards and the experiments were done in triplicate.

2.4.6 Micro-scratch testing

The mechanical properties of the different coatings were compared with a micro-scratch tester (Micro-combi tester; CSM Instruments, Switzerland). The scratch test was performed with reference to ISO 20502:2005(E) [26]. A diamond tip of 100 μm (Rockwell) was used for the scratch test and the scratch track was examined with an attached optical microscope and further by ESEM. The scratch tip was moved under a progressive loading rate of 3.95 and 9.97 N/min respectively for HA and TiO2-HA coatings through a scratch length of 10 mm. Both frictional force and acoustic emission was measured to derive a conclusion of coating failure.

3 Results and discussion

3.1 Characterisation of targets and substrate

The sintered HA discs were analysed for phase and composition by XRD, FTIR and ICP while that of TiO2 by XRD prior to use as targets for the PLD. Figure 1 shows the XRD patterns of the Ti6Al4V substrate, HA and TiO2 targets. All the three patterns correspond to JCPDS patterns 44-1294 (Ti-substrate), 9-0432 (HA-target), and 21-1276 (TiO2-rutile target). The crystallite size was found to be in the range of 40–45 nm for the sintered HA target and 5–10 nm in the case of TiO2 target as measured by the Scherer equation. Both the target samples were identified as highly crystalline and phase pure without any additional phases within the detection limit.

Fig. 1
figure 1

XRD patterns of the Ti6Al4V substrate, sintered TiO2 and HA targets along with standard patterns

Full size image

FTIR spectra of HA target is shown in Fig. 2. The spectrum of HA target is similar to that of a synthetic hydroxyapatite illustrating various peaks correspond to hydroxyl and phosphate groups. Hydroxyl stretching is observed at 3570.3 cm−1 in the spectra of sintered HA target. All the theoretically predicted vibrational modes for phosphate group ν1 to ν4 were present in the prepared HA. The different sites present in the region 1090–970 cm−1 confirms hydroxyapatite rather than a carbonated hydroxyapatite which if present should have only a single peak for phosphate ν3 band centred around 1046 cm−1. Band at 962.4 cm−1 is the phosphate ν1 band and ν4 bands is observed in the region 660–520 cm−1 and the presence of three sites at 629.9, 600.7, and 571.3 cm−1 strongly indicate non-carbonated HA rather than carbonated HA which would have only two sites. There also present the usually weak phosphate ν2 bands in the region 475–450 cm−1 [27]. Presence of these characteristic groups ensures the use of phase pure synthetic HA as the target for PLD.

Fig. 2
figure 2

FTIR spectrum of sintered HA target used for PLD

Full size image

The Ca/P ratio of the sintered HA as determined by ICP was 1.64 ± 0.02 which is close to the expected stoichiometric (1.67) HA. These results obtained from XRD, FTIR and ICP clearly ensures use of stoichiometric, highly crystalline HA and TiO2 as targets for PLD.

The Ti6Al4V alloy being used as the substrate for HA deposition, were analyzed for their average surface roughness to ensure the use of uniform surface profile and was found to be 0.25 μm.

3.2 Characterization of coating

The coating thickness of individual layers was assessed separately; thickness of TiO2 coating alone was found to be around 200 nm. The HA coating with TiO2 interlayer is found to have an approximate thickness of 1 micron. Figure 3 shows the XRD patterns of the three coatings TiO2, HA and TiO2-HA. The patterns were compared with JCPDS files and shows that TiO2 coating matches with the rutile phase (JCPDS No. 21-1276) with a (110) orientation. The intensity of the TiO2 rutile peaks was comparatively lower because of the intense peaks of titanium substrate while the broader nature of peaks was due to the nano crystallite size. XRD pattern of HA coating exactly matches with JCPDS file No. 9-432 with the same (211) orientation of HA target and indicates phase pure composition of coating without any additional calcium phosphate phases at least within the detection level. The pattern of HA coating with TiO2 interlayer also seems to be same as that with out interlayer but the increased peak intensity and sharpness obtained were attributed to improved crystallinity of HA with TiO2 interlayer.

Fig. 3
figure 3

XRD patterns of pulsed laser deposited TiO2, HA and TiO2-HA

Full size image

FTIR analysis results for the HA and TiO2-HA preparations are depicted in Fig. 4. Both the spectra obtained are similar except the increased intensity obtained for TiO2-HA coating compared to HA coating. After laser ablation, the spectra obtained revealed no significant changes in the positions of various functional group frequencies. Hydroxyl (–OH) peak was still present around 3570 cm−1, indicating that the –OH group in the HA structure was well preserved after laser ablation but the peak intensity lost to some extent. The peak area and intensity of phosphate group were relatively different from the target probably because of the sample preparation i.e., from bulk HA and from surface of coating and else due to the less crystalline nature of HA coating compared to target powder. Although the spectra of coating gives only less resolved peak splitting, phosphate bands is having the same splitting pattern in both the target and coating indicates the similar site symmetry of phosphate group [20, 27, 28].

Fig. 4
figure 4

FTIR spectra of pulsed laser deposited HA coating on Ti6Al4V substrate (HA) without TiO2 and with TiO2 interlayer (TiO2-HA)

Full size image

The Ca/P ratio for the HA and TiO2-HA coatings determined by elemental analysis performed by ICP was found to be 1.64 ± 0.02 indicates preservation of Ca/P stoichiometry in both the coatings.

Scanning electron microscopy analysis performed on the deposited samples, together with elemental analysis by EDS, provided supplementary information regarding the chemical composition and the morphological structure of the deposits. Figures 5, 6 and 7 shows the micro-structural features of TiO2, HA and that of HA coating with TiO2 interlayer respectively. The EDS analysis shows the presence of Ca, P, O and Ti peaks according to their elemental composition in the coatings and indicates the presence of TiO2 and calcium phosphates on the substrate as expected. The deposited TiO2 cannot be distinguished easily because of the small coating thickness and high uniformity. But the HA coatings have grown to form thick layer with particles bound together during the film growth process. The HA coating with TiO2 interlayer (Fig. 7) is more uniform with very few groves and pinholes when compared to HA coating with out interlayer (Fig. 6). Similar EDS spectra was obtained for both type of coating and Fig. 7c is the EDS spectra indicating the presence of Ca, P, O, and Ti on the surface of the HA coating with titania interlayer. The higher magnification SEM micrograph shows that HA particles deposited are grown to form clusters and further seems to form a dense aggregated structure on the substrate with the interlayer.

Fig. 5
figure 5

SEM image of TiO2 coating on Ti6Al4V substrate (a) and the corresponding EDS pattern (b)

Full size image
Fig. 6
figure 6

SEM images of HA coating on Ti6Al4V substrate (a) at lower magnification and (b) at higher magnification

Full size image
Fig. 7
figure 7

SEM images of HA coating on Ti6Al4V substrate with TiO2 interlayer (a) at lower magnification (b) higher magnification. (c) Energy dispersive spectrum of HA coating on Ti6Al4V substrate with TiO2 interlayer

Full size image

The scratch test performed gave the acoustic and frictional force along with the penetration depth. Figure 8 shows the representative acoustic emissions produced during the scratch test for both the HA coatings i.e., with and without TiO2 interlayer. Different failure modes and points were assigned in the acoustic profile by observed through the attached optical image capturing device and are denoted in the figure as first crack, first delamination and finally as point of complete delamination of the coated layer. SEM of crack initiation (marked by arrows) and area of complete delamination are also shown respectively in Fig. 9a, b. In the case of HA coating without interlayer (Fig. 8a), the first crack was formed around 0.82 N, and there is a gradual ploughing of HA coating even at lower loads which leads to the first delamination of coating at 1.46 N. As the load increases further, the width and depth of scratch increases till the indenter reaches the metal surface. At this point (1.59 N); there is an abrupt increase in the acoustic signal and further remains stable for the metallic surface. It was also confirmed by SEM and EDS that there is still a thin calcium phosphate layer in the scratch groove which may be carried by the scratch tip and by spallation from the scratch edges till the point where complete delamination occurs. Coating failure was mainly due to the substrate coating interface and the low internal cohesion within the coating. The behaviour of acoustic emission is quite different for the coating with TiO2 interlayer Fig. 8b. Here there is no such detachment of layer at lower loads and the coating first fails within itself at about 4.48 N and on further increase of load up to 5.29 N the coating seems to be adhered well to the substrate. Beyond 6.46 N there is an increase in penetration depth with normal deflection in acoustic signal, which is due to complete removal of HA layer and substrate penetration. But the nature of acoustic emission compared to that of HA single layer conveys that the interlayer and HA adhered well each other and to the substrate so that there is no sign of abrupt coating failure. Here there is no single point observed where the coating got detached completely from the substrate; rather the coating and substrate seems to behave as a single layer with an adherent interlayer. Hence the thin TiO2 coating provided seems to act as the binding layer between the titanium and HA layer.

Fig. 8
figure 8

Acoustic emission profile of scratch test (a) HA coating on Ti6Al4V substrate (b) HA coating on Ti6Al4V substrate with TiO2 interlayer

Full size image
Fig. 9
figure 9

SEM images of HA coating after scratch test showing (a) direction of scratch test with area of crack initiation and (b) region of complete delamination indicated by arrow

Full size image

4 Conclusions

The Pulsed Laser Deposition (PLD) technique is highly advantageous for obtaining stoichiometric and adherent HA coatings over titanium substrate. The ability of PLD to deposit multi-targets is used to provide an interlayer of TiO2 and is found to be very promising for obtaining improved bonding strength of HA coating onto titanium substrate. The XRD, FTIR, SEM and ICP analysis confirmed the formation of HA coating without any other phases even with the interlayer. Comparison of the mechanical property of HA coating with TiO2-HA clearly demonstrates the improved adhesion of coating by incorporating an interlayer of titania. Thus implants making use of HA coating can be further improved by providing an adequate interlayer for their in vivo functionality and long term performance.