- Email: [email protected]
Influence of thickness on the properties of hydroxyapatite coatings deposited by KrF laser ablation
Biomaterials 22 (2001) 2171}2175
In#uence of thickness on the properties of hydroxyapatite coatings deposited by KrF laser ablation J.M. FernaH ndez-Pradas*, L. Cle`ries, E. MartmH nez, G. Sardin, J. Esteve, J.L. Morenza Departament de Fn& sica Aplicada i O" ptica, Universitat de Barcelona, Av. Diagonal 647, E-08028 Barcelona, Spain Received 28 December 1999; accepted 20 November 2000
Abstract The growth of hydroxyapatite coatings obtained by KrF excimer laser ablation and their adhesion to a titanium alloy substrate were studied by producing coatings with thicknesses ranging from 170 nm up to 1.5 m, as a result of di!erent deposition times. The morphology of the coatings consists of grain-like particles and also droplets. During growth the grain-like particles grow in size, partially masking the droplets, and a columnar structure is developed. The thinnest "lm is mainly composed of amorphous calcium phosphate. The coating 350 nm thick already contains hydroxyapatite, whereas thicker coatings present some alpha tricalcium phosphate in addition to hydroxyapatite. The resulting coating to substrate adhesion was evaluated through the scratch test technique. Coatings fail under the scratch test by spallating laterally from the diamond tip and the failure load increases as thickness decreases, until not adhesive but cohesive failure for the thinnest coating is observed. 2001 Elsevier Science Ltd. All rights reserved. Keywords: Hydroxyapatite; Calcium phosphate coatings; Laser ablation; Adhesion.
1. Introduction Hydroxyapatite (HA) is a ceramic material used in bone reconstruction that besides exhibiting biocompatible properties, is bioactive. However, it encounters some limitations as bulk material when applied in load-bearing sites because of its brittleness. The solution is then to coat with HA the metallic implants that will be used in loadbearing situations. The actual coating methods include plasma spraying, sputtering and also laser ablation. The last technique has allowed the deposition of HA coatings with good structural properties [1}11]. The feasibility of these coatings on implants has been evaluated by performing in vitro studies in both undersaturated and saturated conditions [12}16] and bone cell culture experiments [17}19]. In this way, the pulsed-laser-deposited HA coatings have shown to be stable and at the same time bioactive. However, a thin ceramic coating for its application on implants must also have good adhesion to the substrate. A preliminary study has revealed the importance of the morphology on the adhesion of the HA coatings to the
* Corresponding author. Tel.: #34-93-4021134; fax: #34-934021138. E-mail address: [email protected] (J.M. FernaH ndez-Pradas).
titanium alloy substrates [20]. Moreover, the coating must guarantee a good impermeability of the substrate to ful"ll its task of surface-protective-barrier. Normally, the adhesion properties of the coatings improve as their thickness decrease. However, very thin coatings may not ful"ll the requirement of impermeability. In the present work, coatings with di!erent increasing thickness were deposited, and the in#uence of this parameter on their morphological, structural and mechanical properties was studied.
2. Experimental The coatings were deposited by laser ablation using a pulsed KrF excimer laser of 248 nm wavelength. The laser beam was focused with a #uence of 2.3 J/cm onto a rotating HA target 1.5 g/cm dense, inside a vacuum chamber where a water vapor atmosphere of 45 Pa was kept. Either 2000, 4000, 9000 or 18,000 laser shots at 10 Hz were "red onto the target surface, in order to obtain samples with di!erent thickness. As a result of ablation, the target material was ejected as a plasma plume, constituting the coating as it reached and accumulated onto a Ti}6Al}4 V substrate heated to 5753C. Prior to deposition, the substrates were polished with
0142-9612/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 0 0 ) 0 0 4 0 8 - 7
2172
J.M. Ferna& ndez-Pradas et al. / Biomaterials 22 (2001) 2171}2175
SiC paper up to a "nal roughness of 0.03 m and cleaned in a sequence of ultrasonic baths of trichlorethylene, acetone and ethanol. Sample cross section was prepared by cleaving the rear part of the substrate and cracking. Scanning electron microscopy (SEM) was used to examine the surface and cross-section morphologies of the coatings. Their crystalline properties were determined by X-ray di!ractometry (XRD). The adhesion of the coatings to the substrate was evaluated through the scratch test technique. Scratches were produced by duplicate onto each coating by linearly increasing the load on a 50 m-radius spherical diamond
stylus with a scanning speed of 20 m/s and at loading rates of 0.49, 1.60 or 7.3 N/mm to reach "nal loads of 0.9, 3.0 or 18.0 N, respectively. Scratch-induced damage of the coating}substrate system was evaluated through the friction force versus load recordings and through SEM.
3. Results and discussion 3.1. Morphology Top and cross-section SEM views of the obtained coatings are shown in Fig. 1. The coatings are composed
Fig. 1. Scanning electron micrographs of the: (a) 2000-shots, (b) 4000-shots, (c) 9000-shots, and (d) 18,000-shots coatings.
J.M. Ferna& ndez-Pradas et al. / Biomaterials 22 (2001) 2171}2175
2173
Table 1 Deposition time, thickness, growth rate and critical load in the scratch test for the studied coatings No. shots
2000
4000
9000
18,000
Deposition time (min) Thickness (m) Growth rate (nm/shot) Critcal load (N)
3.3 0.17 0.08 5.7
6.7 0.35 0.09 2.9
15 0.75 0.08 1.7
30 1.5 0.08 0.3
Cohesive failure. Adhesive failure.
of a bed of grains, which results from the nucleation and growth of the vaporized species, and micron-size droplets which are a product of the ablation process. As more shots are accumulated, grains grow in size, become elongated and mask the presence of droplets. Some droplets in the 18,000 shots coating have a faceted polycrystalline appearance. The cross-section view of the 2000-shots sample shows complete coverage of the substrate with approximately a few hundred nm thick coating constituted by 100-nmwide columns located in the areas not covered by the droplets (Fig. 1a). The cross-sectional views of the thicker coatings show that the columnar structure evolves growing in height and width. Some droplets seem to have been trapped between columns during growth. It is also possible to identify a few hundred nm thick layer with a glassy appearance interposed between the substrate and the columnar coating on all the samples. The nature of this layer could be attributed to a titanium oxide, since the high temperatures and the presence of both a waterreactive atmosphere and the oxygen from the HA structure, provide conditions for the oxidation of the titanium substrate. However, there is not any apparent correlation between the thickness of this layer and the time of deposition. The coating thicknesses measured from the cross-sectional views and the calculated growth rates are depicted in Table 1 with the number of shots and the associated deposition times. These results indicate that the coatings grow at a constant rate. This behavior is consistent with the results of a previous work where constant ablation rates were found for #uences around 2 J/cm [21]. 3.2. Structure The X-ray di!ractograms of the obtained coatings are shown in Fig. 2. The 2000-shots coating does not give any signal from a crystalline phase. All the peaks in the X-ray di!ractogram of the 4000-shots coating correspond to HA. The 9000- and the 18,000-shots coatings present -tricalcium phosphate (-TCP) peaks in addition to the HA peaks. The presence of any type of titanium oxide was not detected in any sample.
Fig. 2. X-ray di!ractograms for the: (a) 2000-shots, (b) 4000-shots, (c) 9000-shots, and (d) 18,000-shots coatings.
These results indicate that in the thinnest coating HA, if present, would be in a much lower ratio than in the thickest ones. So, this coating would be mainly constituted by amorphous calcium phosphate. However, when deposition continues and hence the sample is kept at high temperature, the heat enables crystallization. It is probable that -TCP is also present in the thinnest coatings although in an insu$cient amount to be detected. The lack of titanium oxide peaks suggests that the interface layer observed by SEM is amorphous. 3.3. Adhesion The scratch test results are reproducible for all the coatings in the two scratch traces. At a certain scratching distance, which corresponds to a load of 0.33 N, the 18,000-shots coating breaks by spallating laterally (Fig. 3a). The 9000-shots coating fails at a higher load (1.64 N) and the spallated coating area is smaller than that for the preceding coating (Fig. 3b). The failure load for the 4000-shots coating increases up to 2.85 N and the spallated coating area is much smaller (Fig. 3c). A magni"cation of the spallated parts (Fig. 3d) indicates that it is only the coating that detaches from the underlying titanium oxide and titanium alloy substrate. The 2000-shots coating does not detach from the substrate even when subjected to the highest load (18 N) (Fig. 3e), but is plastically deformed much in the same manner as a bare titanium alloy substrate does (Fig. 3f ). However, it
2174
J.M. Ferna& ndez-Pradas et al. / Biomaterials 22 (2001) 2171}2175
Fig. 3. Scanning electron micrographs after the scratch test showing: the "rst adhesive failure for (a) the 18,000-shots coating, (b) the 9000-shots coating and (c,d) the 4000-shots coating; and the traces on (e) the 2000-shots coating, and (f) a bare titanium alloy substrate.
su!ers cohesive failure at 5.7 N which provokes a slight coating debris. The critical load values of failure are depicted in Table 1. Therefore, substrate coverage on a polished substrate is attained already with a 2000-shots coating (0.17 m thick). Moreover, the thinnest coatings withstand higher loading forces than the thickest ones. It is a known fact that the critical load for coatings decreases with increasing coating thickness [22,23], and this fact has been partly attributed to the accumulation of stresses [24]. As thickness decreases, more in#uence of the substrate on the scratch characteristics is found: the thinnest coating does not fail adhesively as the thickest ones do, and only deforms plastically because the substrate is ductile. For such a thin "lm, the adhesion to the substrate overcomes its own cohesive strength.
4. Conclusion The hydroxyapatite coatings obtained by excimer laser ablation have a columnar structure and are the result of the growth in size of grain-like particulates and the progressive accumulation of droplets. The coatings start to grow as amorphous calcium phosphate. However, hydroxyapatite is synthesized yet in the "lm 350 nm thick. Alpha tricalcium phosphate is also detected in thicker coatings, although it could be present already from the beginning of coating growth. Coatings fail under the scratch test by spallating laterally from the diamond tip and the failure load increases as thickness decreases, until only plastic deformation and cohesive failure for the thinnest coating is observed.
J.M. Ferna& ndez-Pradas et al. / Biomaterials 22 (2001) 2171}2175
Acknowledgements This work is a part of a research program "nanced by DGESIC of the Spanish Government (Project No. MAT98-0334-C02-1) and DGR of the Catalan Government.
[12]
[13]
References [14] [1] Cotell CM. Pulsed laser deposition and processing of biocompatible hydroxylapatite thin "lms. Appl Surf Sci 1993;69:140}8. [2] Sardin G, Varela M, Morenza JL. Deposition of hydroxyapatite coatings by laser ablation. In: Brown PW, Constanz B, editors. Hydroxyapatite and related materials. London: CRC Press, 1994. p. 225}30. [3] Torrisi L. Structural investigations on laser deposited hydroxyapatite "lms. Thin Solid Films 1994;237:12}5. [4] Singh RK, Qian F, Nagabushnam V, Damodaran R, Moudgil BM. Excimer laser deposition of hydroxyapatite thin "lms. Biomaterials 1994;15:522}8. [5] JelmH nek M, Ols\ an V, JastrabmH k L, Studnicka V, Hnatowicz V, KvmH tek J, HavraH nek V, DostaH lovaH T, Zergioti I, Petrakis A, Hontzopoulos E, Fotakis C. E!ect of processing parameters on the properties of hydroxylapatite "lms grown by pulsed laser deposition. Thin Solid Films 1995;257:125}9. [6] Antonov EN, Bagratashvili VN, Popov VK, Sobol EN, Davies MC, Tendler SJB, Roberts CJ, Howdle SM. Atomic force microscopic study of the surface morphology of apatite "lms deposited by pulsed laser ablation. Biomaterials 1997;18:1043}9. [7] Wang CK, Chern Lin JH, Ju CP, Chang RPH. Structural characterization of pulsed laser-deposited hydroxyapatite "lm on titanium substrate. Biomaterials 1997;18:1331}8. [8] Mayor B, Arias J, Chiussi S, Garcia F, Pou J, LeoH n Fong B, PeH rez-Amor M. Calcium phosphate coatings grown at di!erent substrate temperature by pulsed ArF-laser deposition. Thin Solid Films 1998;317:363}6. [9] Arias JL, GarcmH a-Sanz FJ, Mayor MB, Chiussi S, Pou J, LeoH n B, PeH rez-Amor M. Physicochemical properties of calcium phosphate coatings produced by pulsed laser deposition at di!erent water vapour pressures. Biomaterials 1998;19:883}8. [10] FernaH ndez-Pradas JM, Sardin G, Cle`ries L, Serra P, Ferrater C, Morenza JL. Hydroxyapatite thin "lms by excimer laser ablation. Thin Solid Films 1998;317:393}6. [11] FernaH ndez-Pradas JM, Cle`ries L, Sardin G, Morenza JL. Hydroxyapatite coatings grown by pulsed laser deposition
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
[24]
2175
with a beam of 355 nm wavelength. J Mater Res 1999;14: 4715}9. Antonov EN, Bagratashvili VN, Popov VK, Sobol EN, Howdle SM. Determination of the stability of laser deposited apatite coatings in phosphate bu!ered saline solution using Fourier transform infrared (FTIR) spectroscopy. Spectrochim Acta A 1996;52:123}7. Tucker BE, Cotell CM, Auyeung RCY, Spector M, Nancollas GH. Pre-conditioning and dual constant composition dissolution kinetics of pulsed laser deposited hydroxyapatite thin "lms on silicon substrates. Biomaterials 1996;17:631}7. Cle`ries L, FernaH ndez-Pradas JM, Sardin G, Morenza JL. Dissolution behaviour of calcium phosphate coatings obtained by laser ablation. Biomaterials 1998;19:1483}9. Cle`ries L, FernaH ndez-Pradas JM, Sardin G, Morenza JL. Application of dissolution studies to further characterize pulsed laser deposited calcium phosphate coatings. Biomaterials 1999; 20:1401}5. Cle`ries L, FernaH ndez-Pradas JM, Morenza JL. Behavior in simulated body #uid of calcium phosphate coatings obtained by laser ablation. Biomaterials 2000;21:1861}5. DostaH lova T, HimmlovaH L, JelmH nek M, BaH rtova J. Some biological and physical properties of laser deposited hydroxyapatite based "lms. Cells Mater 1995;5:255}60. Parker TL, Parker KG, Howdle SM, Antonov EN, Bagratashvili VN, Popov VK, Sobol EN, Roberts CJ. Biocompatibility of laser-deposited hydroxyapatite coatings: correlation of coating parameters with cell behaviour. Cell Engng 1996;1:91}6. Cle`ries L, FernaH ndez-Pradas JM, Morenza JL. Bone growth on and resorption of calcium phosphate coatings obtained by pulsed laser deposition. J Biomed Mater Res 2000;49:43}52. Cle`ries L, MartmH nez E, FernaH ndez-Pradas JM, Sardin G, Esteve J, Morenza JL. Mechanical properties of calcium phosphate coatings deposited by laser ablation. Biomaterials 2000;21: 967}71. FernaH ndez-Pradas JM, Cle`ries L, Serra P, Sardin G, Morenza JL. Evolution of the deposition rate and target morphology during the pulsed laser deposition of hydroxyapatite coatings. Appl Phys A, in press. Burnett PJ, Rickerby DS. The scratch adhesion test: an elasticplastic indentation analysis. Thin Solid Films 1988;157:233}54. Wolke JGC, de Groot K, Jansen JA. Dissolution behaviour of radio-frequency magnetron sputtered Ca-P coatings in goats. J Mater Sci 1998;33:3371}6. Wolke JGC, van der Waerden JPCM, de Groot K, Jansen JA. Stability of radiofrequency magnetron sputtered calcium phosphate coatings under cyclically loaded conditions. Biomaterials 1997;18:483}6.
In#uence of thickness on the properties of hydroxyapatite coatings deposited by KrF laser ablation J.M. FernaH ndez-Pradas*, L. Cle`ries, E. MartmH nez, G. Sardin, J. Esteve, J.L. Morenza Departament de Fn& sica Aplicada i O" ptica, Universitat de Barcelona, Av. Diagonal 647, E-08028 Barcelona, Spain Received 28 December 1999; accepted 20 November 2000
Abstract The growth of hydroxyapatite coatings obtained by KrF excimer laser ablation and their adhesion to a titanium alloy substrate were studied by producing coatings with thicknesses ranging from 170 nm up to 1.5 m, as a result of di!erent deposition times. The morphology of the coatings consists of grain-like particles and also droplets. During growth the grain-like particles grow in size, partially masking the droplets, and a columnar structure is developed. The thinnest "lm is mainly composed of amorphous calcium phosphate. The coating 350 nm thick already contains hydroxyapatite, whereas thicker coatings present some alpha tricalcium phosphate in addition to hydroxyapatite. The resulting coating to substrate adhesion was evaluated through the scratch test technique. Coatings fail under the scratch test by spallating laterally from the diamond tip and the failure load increases as thickness decreases, until not adhesive but cohesive failure for the thinnest coating is observed. 2001 Elsevier Science Ltd. All rights reserved. Keywords: Hydroxyapatite; Calcium phosphate coatings; Laser ablation; Adhesion.
1. Introduction Hydroxyapatite (HA) is a ceramic material used in bone reconstruction that besides exhibiting biocompatible properties, is bioactive. However, it encounters some limitations as bulk material when applied in load-bearing sites because of its brittleness. The solution is then to coat with HA the metallic implants that will be used in loadbearing situations. The actual coating methods include plasma spraying, sputtering and also laser ablation. The last technique has allowed the deposition of HA coatings with good structural properties [1}11]. The feasibility of these coatings on implants has been evaluated by performing in vitro studies in both undersaturated and saturated conditions [12}16] and bone cell culture experiments [17}19]. In this way, the pulsed-laser-deposited HA coatings have shown to be stable and at the same time bioactive. However, a thin ceramic coating for its application on implants must also have good adhesion to the substrate. A preliminary study has revealed the importance of the morphology on the adhesion of the HA coatings to the
* Corresponding author. Tel.: #34-93-4021134; fax: #34-934021138. E-mail address: [email protected] (J.M. FernaH ndez-Pradas).
titanium alloy substrates [20]. Moreover, the coating must guarantee a good impermeability of the substrate to ful"ll its task of surface-protective-barrier. Normally, the adhesion properties of the coatings improve as their thickness decrease. However, very thin coatings may not ful"ll the requirement of impermeability. In the present work, coatings with di!erent increasing thickness were deposited, and the in#uence of this parameter on their morphological, structural and mechanical properties was studied.
2. Experimental The coatings were deposited by laser ablation using a pulsed KrF excimer laser of 248 nm wavelength. The laser beam was focused with a #uence of 2.3 J/cm onto a rotating HA target 1.5 g/cm dense, inside a vacuum chamber where a water vapor atmosphere of 45 Pa was kept. Either 2000, 4000, 9000 or 18,000 laser shots at 10 Hz were "red onto the target surface, in order to obtain samples with di!erent thickness. As a result of ablation, the target material was ejected as a plasma plume, constituting the coating as it reached and accumulated onto a Ti}6Al}4 V substrate heated to 5753C. Prior to deposition, the substrates were polished with
0142-9612/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 0 0 ) 0 0 4 0 8 - 7
2172
J.M. Ferna& ndez-Pradas et al. / Biomaterials 22 (2001) 2171}2175
SiC paper up to a "nal roughness of 0.03 m and cleaned in a sequence of ultrasonic baths of trichlorethylene, acetone and ethanol. Sample cross section was prepared by cleaving the rear part of the substrate and cracking. Scanning electron microscopy (SEM) was used to examine the surface and cross-section morphologies of the coatings. Their crystalline properties were determined by X-ray di!ractometry (XRD). The adhesion of the coatings to the substrate was evaluated through the scratch test technique. Scratches were produced by duplicate onto each coating by linearly increasing the load on a 50 m-radius spherical diamond
stylus with a scanning speed of 20 m/s and at loading rates of 0.49, 1.60 or 7.3 N/mm to reach "nal loads of 0.9, 3.0 or 18.0 N, respectively. Scratch-induced damage of the coating}substrate system was evaluated through the friction force versus load recordings and through SEM.
3. Results and discussion 3.1. Morphology Top and cross-section SEM views of the obtained coatings are shown in Fig. 1. The coatings are composed
Fig. 1. Scanning electron micrographs of the: (a) 2000-shots, (b) 4000-shots, (c) 9000-shots, and (d) 18,000-shots coatings.
J.M. Ferna& ndez-Pradas et al. / Biomaterials 22 (2001) 2171}2175
2173
Table 1 Deposition time, thickness, growth rate and critical load in the scratch test for the studied coatings No. shots
2000
4000
9000
18,000
Deposition time (min) Thickness (m) Growth rate (nm/shot) Critcal load (N)
3.3 0.17 0.08 5.7
6.7 0.35 0.09 2.9
15 0.75 0.08 1.7
30 1.5 0.08 0.3
Cohesive failure. Adhesive failure.
of a bed of grains, which results from the nucleation and growth of the vaporized species, and micron-size droplets which are a product of the ablation process. As more shots are accumulated, grains grow in size, become elongated and mask the presence of droplets. Some droplets in the 18,000 shots coating have a faceted polycrystalline appearance. The cross-section view of the 2000-shots sample shows complete coverage of the substrate with approximately a few hundred nm thick coating constituted by 100-nmwide columns located in the areas not covered by the droplets (Fig. 1a). The cross-sectional views of the thicker coatings show that the columnar structure evolves growing in height and width. Some droplets seem to have been trapped between columns during growth. It is also possible to identify a few hundred nm thick layer with a glassy appearance interposed between the substrate and the columnar coating on all the samples. The nature of this layer could be attributed to a titanium oxide, since the high temperatures and the presence of both a waterreactive atmosphere and the oxygen from the HA structure, provide conditions for the oxidation of the titanium substrate. However, there is not any apparent correlation between the thickness of this layer and the time of deposition. The coating thicknesses measured from the cross-sectional views and the calculated growth rates are depicted in Table 1 with the number of shots and the associated deposition times. These results indicate that the coatings grow at a constant rate. This behavior is consistent with the results of a previous work where constant ablation rates were found for #uences around 2 J/cm [21]. 3.2. Structure The X-ray di!ractograms of the obtained coatings are shown in Fig. 2. The 2000-shots coating does not give any signal from a crystalline phase. All the peaks in the X-ray di!ractogram of the 4000-shots coating correspond to HA. The 9000- and the 18,000-shots coatings present -tricalcium phosphate (-TCP) peaks in addition to the HA peaks. The presence of any type of titanium oxide was not detected in any sample.
Fig. 2. X-ray di!ractograms for the: (a) 2000-shots, (b) 4000-shots, (c) 9000-shots, and (d) 18,000-shots coatings.
These results indicate that in the thinnest coating HA, if present, would be in a much lower ratio than in the thickest ones. So, this coating would be mainly constituted by amorphous calcium phosphate. However, when deposition continues and hence the sample is kept at high temperature, the heat enables crystallization. It is probable that -TCP is also present in the thinnest coatings although in an insu$cient amount to be detected. The lack of titanium oxide peaks suggests that the interface layer observed by SEM is amorphous. 3.3. Adhesion The scratch test results are reproducible for all the coatings in the two scratch traces. At a certain scratching distance, which corresponds to a load of 0.33 N, the 18,000-shots coating breaks by spallating laterally (Fig. 3a). The 9000-shots coating fails at a higher load (1.64 N) and the spallated coating area is smaller than that for the preceding coating (Fig. 3b). The failure load for the 4000-shots coating increases up to 2.85 N and the spallated coating area is much smaller (Fig. 3c). A magni"cation of the spallated parts (Fig. 3d) indicates that it is only the coating that detaches from the underlying titanium oxide and titanium alloy substrate. The 2000-shots coating does not detach from the substrate even when subjected to the highest load (18 N) (Fig. 3e), but is plastically deformed much in the same manner as a bare titanium alloy substrate does (Fig. 3f ). However, it
2174
J.M. Ferna& ndez-Pradas et al. / Biomaterials 22 (2001) 2171}2175
Fig. 3. Scanning electron micrographs after the scratch test showing: the "rst adhesive failure for (a) the 18,000-shots coating, (b) the 9000-shots coating and (c,d) the 4000-shots coating; and the traces on (e) the 2000-shots coating, and (f) a bare titanium alloy substrate.
su!ers cohesive failure at 5.7 N which provokes a slight coating debris. The critical load values of failure are depicted in Table 1. Therefore, substrate coverage on a polished substrate is attained already with a 2000-shots coating (0.17 m thick). Moreover, the thinnest coatings withstand higher loading forces than the thickest ones. It is a known fact that the critical load for coatings decreases with increasing coating thickness [22,23], and this fact has been partly attributed to the accumulation of stresses [24]. As thickness decreases, more in#uence of the substrate on the scratch characteristics is found: the thinnest coating does not fail adhesively as the thickest ones do, and only deforms plastically because the substrate is ductile. For such a thin "lm, the adhesion to the substrate overcomes its own cohesive strength.
4. Conclusion The hydroxyapatite coatings obtained by excimer laser ablation have a columnar structure and are the result of the growth in size of grain-like particulates and the progressive accumulation of droplets. The coatings start to grow as amorphous calcium phosphate. However, hydroxyapatite is synthesized yet in the "lm 350 nm thick. Alpha tricalcium phosphate is also detected in thicker coatings, although it could be present already from the beginning of coating growth. Coatings fail under the scratch test by spallating laterally from the diamond tip and the failure load increases as thickness decreases, until only plastic deformation and cohesive failure for the thinnest coating is observed.
J.M. Ferna& ndez-Pradas et al. / Biomaterials 22 (2001) 2171}2175
Acknowledgements This work is a part of a research program "nanced by DGESIC of the Spanish Government (Project No. MAT98-0334-C02-1) and DGR of the Catalan Government.
[12]
[13]
References [14] [1] Cotell CM. Pulsed laser deposition and processing of biocompatible hydroxylapatite thin "lms. Appl Surf Sci 1993;69:140}8. [2] Sardin G, Varela M, Morenza JL. Deposition of hydroxyapatite coatings by laser ablation. In: Brown PW, Constanz B, editors. Hydroxyapatite and related materials. London: CRC Press, 1994. p. 225}30. [3] Torrisi L. Structural investigations on laser deposited hydroxyapatite "lms. Thin Solid Films 1994;237:12}5. [4] Singh RK, Qian F, Nagabushnam V, Damodaran R, Moudgil BM. Excimer laser deposition of hydroxyapatite thin "lms. Biomaterials 1994;15:522}8. [5] JelmH nek M, Ols\ an V, JastrabmH k L, Studnicka V, Hnatowicz V, KvmH tek J, HavraH nek V, DostaH lovaH T, Zergioti I, Petrakis A, Hontzopoulos E, Fotakis C. E!ect of processing parameters on the properties of hydroxylapatite "lms grown by pulsed laser deposition. Thin Solid Films 1995;257:125}9. [6] Antonov EN, Bagratashvili VN, Popov VK, Sobol EN, Davies MC, Tendler SJB, Roberts CJ, Howdle SM. Atomic force microscopic study of the surface morphology of apatite "lms deposited by pulsed laser ablation. Biomaterials 1997;18:1043}9. [7] Wang CK, Chern Lin JH, Ju CP, Chang RPH. Structural characterization of pulsed laser-deposited hydroxyapatite "lm on titanium substrate. Biomaterials 1997;18:1331}8. [8] Mayor B, Arias J, Chiussi S, Garcia F, Pou J, LeoH n Fong B, PeH rez-Amor M. Calcium phosphate coatings grown at di!erent substrate temperature by pulsed ArF-laser deposition. Thin Solid Films 1998;317:363}6. [9] Arias JL, GarcmH a-Sanz FJ, Mayor MB, Chiussi S, Pou J, LeoH n B, PeH rez-Amor M. Physicochemical properties of calcium phosphate coatings produced by pulsed laser deposition at di!erent water vapour pressures. Biomaterials 1998;19:883}8. [10] FernaH ndez-Pradas JM, Sardin G, Cle`ries L, Serra P, Ferrater C, Morenza JL. Hydroxyapatite thin "lms by excimer laser ablation. Thin Solid Films 1998;317:393}6. [11] FernaH ndez-Pradas JM, Cle`ries L, Sardin G, Morenza JL. Hydroxyapatite coatings grown by pulsed laser deposition
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
[24]
2175
with a beam of 355 nm wavelength. J Mater Res 1999;14: 4715}9. Antonov EN, Bagratashvili VN, Popov VK, Sobol EN, Howdle SM. Determination of the stability of laser deposited apatite coatings in phosphate bu!ered saline solution using Fourier transform infrared (FTIR) spectroscopy. Spectrochim Acta A 1996;52:123}7. Tucker BE, Cotell CM, Auyeung RCY, Spector M, Nancollas GH. Pre-conditioning and dual constant composition dissolution kinetics of pulsed laser deposited hydroxyapatite thin "lms on silicon substrates. Biomaterials 1996;17:631}7. Cle`ries L, FernaH ndez-Pradas JM, Sardin G, Morenza JL. Dissolution behaviour of calcium phosphate coatings obtained by laser ablation. Biomaterials 1998;19:1483}9. Cle`ries L, FernaH ndez-Pradas JM, Sardin G, Morenza JL. Application of dissolution studies to further characterize pulsed laser deposited calcium phosphate coatings. Biomaterials 1999; 20:1401}5. Cle`ries L, FernaH ndez-Pradas JM, Morenza JL. Behavior in simulated body #uid of calcium phosphate coatings obtained by laser ablation. Biomaterials 2000;21:1861}5. DostaH lova T, HimmlovaH L, JelmH nek M, BaH rtova J. Some biological and physical properties of laser deposited hydroxyapatite based "lms. Cells Mater 1995;5:255}60. Parker TL, Parker KG, Howdle SM, Antonov EN, Bagratashvili VN, Popov VK, Sobol EN, Roberts CJ. Biocompatibility of laser-deposited hydroxyapatite coatings: correlation of coating parameters with cell behaviour. Cell Engng 1996;1:91}6. Cle`ries L, FernaH ndez-Pradas JM, Morenza JL. Bone growth on and resorption of calcium phosphate coatings obtained by pulsed laser deposition. J Biomed Mater Res 2000;49:43}52. Cle`ries L, MartmH nez E, FernaH ndez-Pradas JM, Sardin G, Esteve J, Morenza JL. Mechanical properties of calcium phosphate coatings deposited by laser ablation. Biomaterials 2000;21: 967}71. FernaH ndez-Pradas JM, Cle`ries L, Serra P, Sardin G, Morenza JL. Evolution of the deposition rate and target morphology during the pulsed laser deposition of hydroxyapatite coatings. Appl Phys A, in press. Burnett PJ, Rickerby DS. The scratch adhesion test: an elasticplastic indentation analysis. Thin Solid Films 1988;157:233}54. Wolke JGC, de Groot K, Jansen JA. Dissolution behaviour of radio-frequency magnetron sputtered Ca-P coatings in goats. J Mater Sci 1998;33:3371}6. Wolke JGC, van der Waerden JPCM, de Groot K, Jansen JA. Stability of radiofrequency magnetron sputtered calcium phosphate coatings under cyclically loaded conditions. Biomaterials 1997;18:483}6.