of 6

Structure, bonding state and in-vitro study of Ca–P–Ti film deposited on Ti6Al4V by RF magnetron sputtering

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Structure, bonding state and in-vitro study of Ca–P–Ti film deposited on Ti6Al4V by RF magnetron sputtering
  Structure, bonding state and in-vitro study of Ca–P–Ti film depositedon Ti6Al4V by RF magnetron sputtering J.D. Long a  , S. Xu a, *, J.W. Cai a  , N. Jiang a  , J.H. Lu  b , K.N. Ostrikov a  , C.H. Diong a  a   Natural Sciences, Plasma Sources and Applications Centre, NIE, Nanyang Technological University, 1 Nanyang Walk, Singapore 637616, Singapore  b  National University Medical Institute, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260, Singapore Abstract Experimental investigation of functionally graded calcium phosphate-based bio-active films on Ti–6Al–4Vorthopaedic alloy prepared inan RF magnetron sputtering plasma reactor is reported. The technique involves concurrent sputtering of Hydroxyapatite (HA) and Ti targets,which results in remarkably enhanced adhesion of the film to the substrate and stability of the interface. The films have been characterizedusing X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). The XPS data show that the films are composed of O, Ca, P andTi, and reveal the formation of O j  P groups and hybridization of O–Ca–P. The XRD pattern shows that the Ca–P thin films are of crystalline calcium oxide phosphate (4CaO  P 2 O 5 ) with preferred orientation varying with processing parameters. High-resolution opticalemission spectra show that the emission of CaO is dominant. The CaO, PO and CaPO species are strongly influenced by depositionconditions. The introduction of Ti element during deposition provides a stable interface between bio-inert substrates Ti–6Al–4V and bioactive HA coating. In-vitro cell culturing tests suggest excellent biocompatibility of the Ca–P–Ti films. D  2002 Elsevier Science B.V. Allrights reserved.  Keywords:  Hydroxyapatite; Multiple-target sputtering; Bonding state; In-vitro cell culturing 1. Introduction Bioactive surface coatings have recently become prom-isingly attractive for development of ultra-high-precisionmedical instruments, biosensors, and novel implant materi-als [1]. Biomaterials have a direct impact on the quality of life for people suffering from debilitating disease or reduc-tion in bodily function. One of the greatest challenges facingscientists and engineers working at the clinical interface isthe development of a new range of biomaterials that willcontribute to the promotion of such repair processes in thehuman body. Calcium phosphate (Ca–P) bioceramics, in particular, hydroxyapatite [HA, Ca 10 (PO 4 ) 6 (OH) 2 ], arewidely used in clinical implant devices due to its excellent stability, mechanical properties, and biocompatibility [2,3].HA coatings have proven to be able to improve tissue protein attachment and growth on Ti–6Al–4V orthopaedicalloy often used as artificial joints [3].Several techniques for fabrication of HA-coated implantssuch as plasma spraying [1,4–6] and pulsed laser depositionhave recently been explored. These, and some other existingschemes provide a wide range of calcium phosphate coat-ings with excellent mechanical and tribological properties.However, recent development in the field indicates that HA–substrate interface strength, coating composition con-trol and other factors still deserve further improvement.Furthermore, in vivo studies reveal that failure often hap- pens at the interface between the HA coating and theorthopaedic alloy soon after implantation [7–9].On the other hand, RF magnetron sputtering depositionin low-temperature plasma environment has proven to beable to provide high degree of process and film compositioncontrol [10]. Thus, it is worthwhile to explore the advan-tages that this deposition scheme can bring to fabrication of advanced calcium phosphate-based bioceramics. In thisarticle, we report on the dramatic improvement of adhesion properties of Ca–P-based films to the orthopaedic alloy Ti– 6Al–4V by using concurrent sputtering of HA and Titargets in low-pressure RF magnetron discharge in argon.Likewise, varying the discharge control parameters, one canefficiently engineer the chemical structure, composition, andcrystallinity of the films, and reproduce a number of calcium phosphate-based bio-ceramics, including hydroxyapatite.This paper is organized as follows: experimental details 0928-4931/02/$ - see front matter   D  2002 Elsevier Science B.V. All rights reserved.PII: S0928-4931(02)00029-2 * Corresponding author. Tel.: +65-6790-3818; fax: +65-6896-9432.  E-mail address:  syxu@nie.edu.sg (S. Xu).www.elsevier.com/locate/msecMaterials Science and Engineering C 20 (2002) 175–180  and characterization techniques are described in Section 2.The results on film composition, bonding states, crystalstructure, optical emission, interface bonding strength, and biocompatibility tests are reported in Section 3. The resultsare summarized in Section 4. 2. Experimental details The deposition of HA thin films has been carried out inan RF magnetron sputtering system. The argon discharge is produced using a water cooled, RF powered magnetronelectrode located at the lower end of a 25-l cylindricalvacuum vessel [10]. The purposely designed magnetronelectrode is made of Cu/ferromagnetic material and con-tained a large number of permanent magnets with specific polarities throughout the electrode. This arrangement com- bined with relatively large sheath potentials, enables one toachieve high sputtering yields. A 10 cm in diameter, high purity HA and/or Ti target is located on the RF poweredelectrode. An electrically floating substrate heater/holder is fixed at   f 6 cm above the target. A base pressure of  f 2  10  5 Torr is obtained by a 450 l/s turbo pump.Thereafter, high-purity Ar gas, which acts as sputtering gas,is introduced into the chamber. To maintain a gas pressureconstant during the process, the gas flow rates are monitoredwith MKS mass flow controllers. The film deposition has been conducted with 200–700 W RF power, in the operat-ing gas pressure range of 5–50 mTorr. Ti–6Al–4V plates(1.5  1. 5cm 2 ) are used as substrates. Prior to deposition,the substrates were first immersed in HF+HNO 3  acid toremove the surface oxide, followed by degreasing in ace-tone, and finally blow dried with purified nitrogen. Thesubstrate temperature was varied from 200 to 650  j C.The chemical composition and bonding states of thedeposited thin films have been studied ex situ by VGEscalab 220i-XL spectrometer (XPS) and Mg K  a  (1253.6eV) X-ray source. The crystalline structure of the film isanalyzed with Siemens D5005 X-ray diffractometer (XRD)in a lock coupled ( h  –2 h ) mode with an incident X-raywavelength of 1.540 A˚(Cu K  a  line). The interface bondingstrength measurements are performed using a CSEM MSTMicroscratch Tester. A very high-resolution (0.023 nm)Roper Scientific Acton Research optical emission spectro-scope is used to investigate the atomic and radical species inthe discharge. Finally, the in-vitro cell culture is conducted by staining the cell with rhodamine-conjugated phalloidinand visualized using a Leica DMRXA fluorescent micro-scope. 3. Results and discussion The surfaces of the deposited HA/Ti samples appear to beextremely smooth and uniform. With an  a -step system, thefilms thickness are measured to be in the range of 0.5–5 um. 3.1. Composition and bonding state analysis The composition and chemical states of the samples have been analyzed by XPS. To remove the adventitious contam-inants, the sample surface has been sputter-cleaned using 2keVAr ions prior to the XPS analysis. Fig. 1 shows an XPSwide survey scan of a sample synthesized with  W  RF =570 W,  P  o =18 mTorr and  V  B =70 V, where  W  RF ,  P  o , and  V  B  are theRF input power, gas pressure, and bias voltage, respectively.It is seen that all anticipated elements Ca, P, O and Ti are present in the film. The percentage atomic concentrations of elements in the film are calculated using transmissionfunctions and sensitivity factors for each of the constituent elements, and appear to be of Ca (25.96%), P (12.78%), O(53.84%) and Ti (3.38%). These numbers approximatelyrecover the stoichiometric value of HA. In addition, a smallamount of F originated from a Teflon component of thetarget holder is also found in the film.The subsequent XPS narrow scan spectra of Ca, P, Oand Ti in Ca–P thin film is shown in Fig. 2. From thewidth and shape of the XPS curves, we infer that multi-component peaks exist. The spectra have been processedwith VG software to fit the Gaussian peak componentsmixed with Lorentzian shapes. For O 1s spectra, the peak fitting routine yields two components at the binding ener-gies (BE) 531.4 and 532.1 eV. These two peaks respec-tively correspond to the states of oxygen in hydroxide and phosphates. Proper peak assignments have to include boththe Ca 2p and P 2p spectra. The two peaks displayed in Ca2p spectra confirm the presence of two chemical states of Ca with BEs at 347.5 and 348.2 eV, respectively. Com- pared to the binding energies measured using the target material HA, both peaks are identified. The first peak (alsothe main component of the spectra) exactly recovers the binding energy of Ca in calcium phosphate, revealing that most of Ca in the film is in form of the calcium phosphate. Fig. 1. XPS wide scan spectra of a film prepared at   W  RF =570 W,  P  o =18mTorr and  V  B =70 V.  J.D. Long et al. / Materials Science and Engineering C 20 (2002) 175–180 176  Two components at 132.8 and 133.7 eV are obtained in the peak fit of P 2p spectra. The 132.8-eV peak is attributed tothe 2p3/2 state of P whereas the 133.7-eV peak is ascribedto the spin-orbit splitting state 2p1/2 as suggested in Ref.[11]. The P 2p spectra imply the presence of the phosphate phase in the film. Additional peak fitting performed on Ti2p3/2 spectra yields two peaks at 458.5 and 459.2 eV,showing that titanium exists in the forms of TiO 2  or CaTiO 3  in the Ca–P thin film.Fig. 3 shows the Ca/P ratio as a function of bias voltagemeasured with XPS, for films prepared at working gas pressure 70 mTorr and 550 W RF power. It is seen that the Ca/P ratio increases with the applied bias. Below 100 V,the Ca/P ratio is measured to be 1.45–1.56, very close to theexpected stoichiometric value of HA (1.67). For a higher  bias voltage, the Ca/P ratio increase rapidly, from  f 1.6 at 100 V to 7.5 at 200 V. This result shows conclusively thetendency of the Ca/P increase with the bias voltage.The effect of substrate temperature on the Ca/P ratio isinvestigated for a fixed pressure of 70 mTorr, zero bias and550 W RF power. Fig. 4 shows the Ca/P ratio as a functionof the substrate temperature. The Ca/P ratio graduallydecreases from 1.4 to 1.0 when the temperature is elevatedfrom 300 to 650  j C. 3.2. Crystal structure analysis The crystalline structural analysis of the films has been performed using XRD. Data are collected from 20 j  to 65 j with a step of 0.02 j , sufficient to cover the main reflections.Fig. 5 shows the XRD patterns of three samples prepared at 0-, 70- and 150-V bias and at 18 mTorr. In the case of absence of a bias, it is clear seen that a strong diffraction peak at (130) is present, suggesting that the crystallinegrowth is along a preferred orientation of (130) plane. Withincreasing of the bias to 70 V, another strong peak corre-sponding to (132) plane appears. Further increasing the biasto 150 V, these peaks disappear and a new peak correspond-ing to the (143) plane is formed. Generally, the Ca–P–Tifilms feature calcium oxide phosphate (4CaO  P 2 O 5 ) as themain crystalline phase, and the preferred orientation can beefficiently controlled by varying the RF power, gas pressure,and bias. It is established that crystalline calcium phosphate Fig. 3. Ca/P ratio versus bias voltage other parameters are the same as inFig. 1.Fig. 2. The XPS narrow scan spectra for O, Ca, P and Ti.  J.D. Long et al. / Materials Science and Engineering C 20 (2002) 175–180  177  can be directly formed in the process of plasma sputteringdeposition. This phenomenon may be due to enhancednucleation at high RF sheath potential, which results in anextraordinary grain growth. 3.3. Optical emission study To investigate the atomic and radical species during thedeposition process, a high-resolution OES has beenemployed to monitor the emission. Emission signals arecollected from a region just below the substrates. Fig. 6shows typical OE spectra of dominant plasma species in the process of fabrication for the case of 18 mTorr and 70 Vdischarge. It has been observed that the intensity of opticalemission of neutral and ionised species is maximal near the bottom-electrode target. The maximum of the optical inten-sity has been recorded in the wavelength range of 380–600nm. It is remarkable that the OE spectra feature several linesattributed to CaO, Ca, CaPO, PHO, as well as other ele-ments and radicals persistent in natural bone tissues. Amongthem, the CaO spectral line at 422.41 nm appears to be of the strongest intensity. The identity and intensity of emis-sion species in turn depended on the gas flow rate, pressure, bias and RF power. 3.4. Interface bonding strength measurement  The scratch-test method has been used for assessing theadhesive strength between deposited film and substrate. Thescratch tests are carried out using a CSEM MST Micro-scratch Tester. The scratch length is set to be 8 mm and theapplied load is varied from 0.3 to 8 N with load increment of 2 N/min. The critical loads are determined by the combi-nation of optical observation and the frictional force meas-urement. Fig. 7 displays the variation of the critical loads Fig. 5. Influence of bias voltage on crystalline structure.Fig. 6. OES spectra during the concurrent HA/Ti sputtering process.Fig. 7. Interface bonding strength as a function of argon gas pressure.Fig. 4. The effect of the substrate temperature on Ca/P ratio.  J.D. Long et al. / Materials Science and Engineering C 20 (2002) 175–180 178  under different working pressures for the cases of zero and70-V bias films. It is seen that the critical load exhibitshigher value at 10 mTorr and gradually decreases with theincrease of the working gas pressure. The critical load variesin the range of 2.5 and 4.5 N. for the films synthesized at different conditions. Thus, the bias voltage has an insignif-icant effect on the interface bonding strength. This result implies that a low working gas pressure may enhance theinterface bonding strength. We also observed that, with theintroduction of Ti, the bonding strength, hence the adhesionof the film to the substrate and stability of the interface, isremarkably improved. 3.5. Cell culture To investigate the biocompatibility of coated HA film,special samples have been prepared on the bio-glass cover slips, co-deposited with the usual Ti–6Al–4V substrate.The property of the coated Ca–P–Ti film is investigated for its ability to support the adhesion of mammalian cells in anin vitro cell cultured. COS7 cells have been inoculated at 5  10 5 cells/well into 24-well tissue culture plates in whichthe HA-coated cover slips are placed. The cells have beencultured for 48 h and then fixed in 3.7% formaldehyde and permeablized in 0.2% saponin solutions. The cells are thenstained with rhodamine-conjugated phalloidin and visual-ized using a Leica DMRXA fluorescent microscope. Thecells are examined under phase contrast. Fig. 8 shows theCOS7 cell morphology after 48-h culturing, revealingnormal adhesion of the cells to the HA-coated cover slips.Furthermore, cell morphology and cytoskeletal structuresappear to be similar to those of cells cultured on uncoatedcover slips. 4. Conclusion In conclusion, Ca–P-based films with various structural/ mechanical properties have been deposited by the multiple-target RF magnetron sputtering technique. The main crys-talline phase is calcium oxide phosphate (4CaO  P 2 O 5 ). Thecrystallization and preferred orientation of crystal growthdepends on the working pressure and bias voltage. A stoi-chiometric, highly crystalline HA film can be obtained bycareful selection of processing parameters, especially theworking gas pressure and bias voltage. The interface bond-ing strength has higher value under low working pressures.Finally, in-vitro cell culturing tests sound promising for developing the implant coating technique and forthcomingin-vivo tests. References [1] C.C. Berndt, G.N. Haddad, A.J.D. Farmer, K.A. Gross, Thermalspraying for bioceramic applications, Mater. Forum 14 (1990) 161– 173.[2] Y.C. Tsui, C. Doyle, T.W. Clyne, Plasma sprayed hydroxyapatite coat-ings on titanium substrates: Part 2. Optimization of coating properties,Biomaterials 19 (1998) 2031–2043.[3] C.K. Wang, J.N. Chern Lin, C.P. Ju, H.C. Ong, P.R.H. Chang, Struc-tural characterization of pulsed laser-deposited hydroxyapatite film ontitanium substrate, Biomaterials 18 (1997) 1331–1338.[4] F. Brossa, A. Cigada, R. Chiesa, L. Paracchini, C. Consonni, Adhe-Fig. 8. COS7 cell morphology after 48 h of culturing.  J.D. Long et al. / Materials Science and Engineering C 20 (2002) 175–180  179
Related Search
Related Docs
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks