Available online at www.sciencedirect.

com

Acta Biomaterialia 5 (2009) 32053214 www.elsevier.com/locate/actabiomat

Mechanical and in vitro biological performances of hydroxyapatitecarbon nanotube composite coatings deposited on Ti by aerosol deposition
Byung-Dong Hahn a, Jung-Min Lee a, Dong-Soo Park a,*, Jong-Jin Choi a, Jungho Ryu a, Woon-Ha Yoon a, Byoung-Kuk Lee a, Du-Sik Shin b, Hyoun-Ee Kim b
Functional Ceramics Group, Functional Materials Division, Korea Institute of Materials Science (KIMS), 66 Sangnam-Dong, Changwon, Gyeong-Nam 641-010, Republic of Korea b School of Materials Science and Engineering, Seoul National University, San 56-1 Sillim-Dong, Gwanak-gu, Seoul 151-742, Republic of Korea Received 27 November 2008; received in revised form 25 March 2009; accepted 5 May 2009 Available online 13 May 2009
a

Abstract Hydroxyapatite (HA)carbon nanotube (CNT) composite coatings on Ti plate, produced by aerosol deposition using HACNT powders, were developed for biomedical applications. For the deposition process HACNT powder mixtures with CNT contents of 1 and 3 wt.% were used. Dense coatings with a thickness of 5 lm were fabricated, irrespective of the content of CNTs. No pores or microcracks were observed in the coatings. The coatings had good adhesion to the substrate, exhibiting a high adhesion strength, ranging from 27.3 to 29.0 MPa. Microstructural observation using eld-emission gun scanning electron microscopy and transmission electron microscopy showed that CNTs with a typical tubular structure were found in the HACNT composite coatings. Nanoindentation tests revealed that the mechanical properties, such as the hardness and elastic modulus, were signicantly improved by the addition of the CNTs to the HA coating. In addition, the proliferation and alkaline phosphatase (ALP) activity of MC3T3-E1 pre-osteoblast cells grown on the HACNT composite coatings were higher than those on the bare Ti and pure HA coating. The ALP activity of the composite coatings considerably improved as the CNT content increased. These results suggest that CNTs would be an eective reinforcing agent to enhance both the mechanical and biological performances of HA coatings. 2009 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Keywords: Hydroxyapatite; Carbon nanotube; Aerosol deposition; Nanoindentation; Biological performance

1. Introduction Hydroxyapatite (Ca10(PO4)6(OH)2, HA), which belongs to the class of calcium phosphate-based biomaterials, has been widely used for a variety of biomedical applications in dentistry and orthopedics, due to its chemical resemblance to the mineral component of bone, excellent biocompatibility and osteoconductivity [13]. However, the peculiar brittleness and low fracture toughness of HA have restricted its usage in applications such as high load-bearCorresponding author. Tel.: +82 55 280 3345. E-mail addresses: pds1590@kims.re.kr, jh980214@hotmail.com (D.-S. Park).
*

ing implants [4]. One way to overcome this problem is to coat HA on a metallic implant surface. The HA-coated implant can combine the high mechanical strength of the metal with the excellent biocompatibility and bioactivity of the ceramic and is therefore suitable for implants in high load-bearing applications [5,6]. Recently, in order to improve the mechanical properties of the HA coating itself, HA composite coatings in which HA is combined with other materials used as a second phase, such as ethylene-based polymer [7], yttria-stabilized zirconia [8] and alumina [9], have been extensively studied. However, considerable amounts of the second phase are required for a signicant improvement in the mechanical properties of the composite coatings. Moreover, the bioinertness or poor bio-

1742-7061/$ - see front matter 2009 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2009.05.005

3206

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214

activity of these phases often results in the HA composite coatings having much lower biological properties than HA itself [8]. Therefore, for the fabrication of composite coatings having good mechanical properties without any deterioration in their bioactivity it is necessary to use a second phase with high bioactivity or to keep the amount of the second phase as low as possible, so as not to lower the biological performance of the coatings signicantly. Carbon nanotubes (CNTs), which were rst discovered by Iijima in 1991 [10], have been widely investigated owing to their excellent thermal, electrical and mechanical properties resulting from their unique atomic structure [11,12]. CNTs consist of a rolled-up sheet of graphite and generally exhibit a high aspect ratio in the range of 100010,000. It was reported that CNTs exhibit a remarkable elastic modulus of more than 1 TPa and tensile strength reaching 30 GPa [13,14]. These excellent mechanical properties of CNTs have stimulated their use as a reinforcing agent for high strength composites [15,16]. Since the bioactivity of CNTs was reported [1719] their use in biomedical applications has been anticipated. The combination of biological and excellent mechanical properties of CNTs has opened a new research eld of CNT reinforced HA composite coatings [20,21]. Balani et al. fabricated an HACNT composite coating using plasma spraying and reported that its fracture toughness was improved by 56% compared with that of a monolithic HA coating [20]. Although plasma spraying is the most popular method in commercial use to produce HA coatings on implants, plasma sprayed HA coatings have several disadvantages, such as their low density, susceptibility to microcracks and phase decomposition on high temperature exposure, thereby limiting the long-term stability of the implant [22]. Another reported method of producing HACNT composite coatings is electrophoretic deposition, which is known to be a low cost and simple method of producing HA coatings on metallic implants [23]. Kaya et al. investigated the mechanical properties of an electrophoretically deposited HA2 wt.% CNT coating and reported that it showed a 400% improvement in adhesion strength over a monolithic HA coating [21]. However, the adhesion strength value of the HACNT composite coating was as low as 2.8 MPa, which might make it unsuitable for real applications. Aerosol deposition (AD) is a newly developed coating method which is capable of depositing dense coating layers on various substrates such as metals, ceramics and plastics [24,25]. This technique uses solid particles as the starting materials and the coating layer is formed by the collision of the particles with a substrate. AD has received considerable attention due to its low processing temperature (near room temperature), high adhesion strength and precise control of the coating composition, making it very suitable for producing HA coatings for use in biomedical applications. In the present work we deposited HACNT composite coatings on Ti substrates using AD with the aim of improv-

ing the mechanical and biological properties of HA coatings. For the coating process a powder mixture of HA and well-dispersed CNT was prepared. The mechanical properties of the coatings were measured using nanoindentation as a function of the amount of CNTs added to the coatings. The in vitro biological properties of the coatings were also evaluated in terms of cell proliferation and dierentiation. 2. Experimental procedure 2.1. Materials A commercially available nanocrystalline HA (Samjo Industrial Co. Ltd, Cheongwon, Korea) powder with an average particle diameter of about 15 nm was used as the starting material. The as received HA powder was heated at 1100 C for 2 h in order to obtain a powder with a particle size appropriate for coating. In addition, commercially available multi-walled CNT (CM-95, Hanwha Nanotech Co. Ltd, Seoul, Korea) with a diameter of 1015 nm and length of 1020 lm was also used for fabricating the HACNT composite coatings by AD. Since CNTs tend to agglomerate due to the strong van der Waals attractions between them they must be dispersed uniformly within the HA matrix for them to act as an eective reinforcement in HACNT composite coatings. The dispersion technique used in this study was acid treatment, which introduced COOH groups on the surface of the CNTs by chemical oxidation, resulting in good dispersion of the CNTs in the solvent [26]. The CNTs were reuxed in an acid solution of H2SO4 and HNO3 (1:3 by volume) at 130 C for 30 min, following by washing to pH 7 using distilled water. The acid-treated CNTs were then added to an aqueous suspension of the dispersed HA powder and ultrasonically mixed for 2 h. The concentrations of CNTs added to the HA suspension were 1 and 3 wt.% in this study. After drying the HACNT powder mixtures were prepared for the composite coating. They were called, respectively, HA1 wt.% CNT and HA3 wt.% CNT. 2.2. Preparation of HA coating The HACNT composite coating was deposited at room temperature using AD. Full details of the apparatus utilized in the AD system have been described elsewhere [24]. The substrates used for the coating deposition were commercially pure Ti plates with dimensions of 10 10 0.5 mm. The Ti plates were rinsed in distilled water and ultrasonically washed in acetone for 10 min before coating. A slit-type nozzle with a 10 0.5 mm2 rectangular opening was employed and the ow rate of air used as a carrier gas was 30 l min1. Coatings with a thickness of 5 lm were deposited over the entire surface of the Ti plates by scanning the Ti plates on a motorized XY stage for 1 min.

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214

3207

2.3. Characterization The powders used for coating were characterized by transmission electron microscopy (TEM) (JEM-2100F, JEOL Co., Tokyo, Japan) and X-ray diraction (XRD) (D-MAX 2200, Rigaku Co., Tokyo, Japan). Field-emission gun scanning electron microscopy (FEG-SEM) (JSM6700F, JEOL Co., Tokyo, Japan) in conjunction with energy dispersive X-ray spectroscopy (EDS) and TEM were used to observe the surface and cross-sectional morphologies of the lms. The TEM sample preparation of the coatings was carried out using a focused ion beam (FIB) (NOVA 200, FEI Co., Hillsboro, OR, USA). The surface roughness of the coatings was measured by means of a laser surface prolometer (Perthometer PGK 120, Mahr GmbH, Gottingen, Germany). The surface parame ter used for the evaluation of surface roughness was the arithmetic average roughness (Ra). Four scans were performed for each coating specimen with a scanning distance of 5.6 mm and a scanning speed of 30 mm min1. The tensile adhesion strength between the coating and substrate was measured by the standard adhesion test (ISO 137794) using a universal testing machine (Instron series IX automated materials testing system, Instron Corp., Norwood, MA, USA). Five replicates of each sample were tested. Coatings of 5 lm thickness were deposited on a circular Ti substrate with dimensions 25 mm in diameter and 0.5 mm in thickness, and the supporting xtures for a tensile testing machine were bonded on the coatings top side and the substrates bottom side with thermal cure, high strength epoxy (3 M Scotch-Weld Epoxy Adhesive 2214, 3 M Corp., St Paul, MN, USA). A tensile force was then applied by the machine until the coating failed. The nanoindentation test was conducted with a standard Berkovich indenter using a Nanoindenter XP (MTS Systems Corp., Nano Instruments Innovation Center, Oak Ridge, TN, USA). The continuous stiness measurement (CSM) method was employed to evaluate the mechanical properties of the coatings. The hardness and elastic modulus of the coatings were derived from the loadingunloading curves using the OliverPharr method [27]. Twelve indentations were made for each test on each sample. The loadindentation depth proles were recorded automatically during the indentation. The maximum indentation depth chosen for the present test was 0.5 lm. 2.4. In vitro cell tests

50 lg ml1 streptomycin at 37 C in a humidied atmosphere of 5% CO2 and 95% air. For characterization of cell attachment the cells were seeded on specimens and were grown in culture medium. After 5 h incubation the cells were xed in 2.5% glutaraldehyde, dehydrated gradually and coated with gold to be examined by SEM. Cytotoxicity was quantied by a standard measurement of lactate dehydrogenase (LDH) release using a LDH assay kit (Roche Diagnostics GmbH, Mannheim, Germany) according to the instructions of the manufacturer. The cells (2 104 cells ml1), suspended in 200 ll a-MEM, were seeded on the specimens in 24-well plates. The plates were incubated for 24 h. After incubation the supernatant from each well was collected for LDH measurement. The percentage cytotoxicity was calculated according to the equation used by Korzeniewski et al. [28]. For cell proliferation tests on coatings with dimensions of 10 10 mm a Cell Titer 96 Aqueous One Solution Cell Proliferation Assay (Promega Corp., Madison, WI, USA) was used. The cells (3 104 cells ml1) were seeded on the coating specimens in 24-well plates containing a-MEM and placed for 5 days in an incubator set at 37 C. The cells were then washed with phosphate-buered saline (PBS) solution and placed in culture medium containing MTS solution (3-(4,5-dimethlythiazol-2-yl)-5-(3carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) and returned to the incubator at 37 C for 3 h. The absorbance was determined at a wavelength of 490 nm using a microplate reader (Model 550, Biorad, Hercules, CA, USA). Cell dierentiation was assessed by measuring the alkaline phosphatase (ALP) activity of cells cultured on the coatings. The cells (1 104 cells ml1) were seeded on the coating specimens and cultured for 10 days, washed with PBS and detached using trypsinethylene diamine tetraacetic acid. The amounts of protein in the cell lysates were quantied using a protein assay kit (Biorad, Hercules, CA, USA) and the ALP activities were assayed colorimetrically using p-nitrophenyl phosphate as the substrate (ALP yellow liquid substrate for ELISA, Sigma, St. Louis, MO, USA). The absorbance of the reaction product, p-nitrophenol, was measured at 405 nm using a microplate reader. Data are expressed as means SD and statistical signicance was set at P < 0.001 as determined by the Students t-test. 3. Results and discussion

MC3T3-E1 pre-osteoblast cells (CRL-2593, ATCC, Manassas, VA, USA) were used to examine cellular proliferation and dierentiation on the HACNT composite coatings. Five replicates of each specimen were also used for testing. Prior to the in vitro experiments the cells were cultured in a-minimum essential medium (a-MEM) (Join Bio Innovation, Seoul, Korea) supplemented with 10% heat-inactivated fetal bovine serum (HI-FBS) (GIBCO, USA), 2 mM L-glutamine, 50 IU ml1 penicillin and

3.1. HACNT powder mixture for coating In order for CNTs to be used as the reinforcement of a composite it is very important to disperse them uniformly within a matrix. However, well-dispersed CNTs are not easy to achieve through a simple mixing method. In the present study the surfaces of the CNTs were modied by acid treatment in order to obtain a well-dispersed, stable

3208

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214

CNT suspension. Fig. 1 shows optical micrographs of aqueous suspensions containing the CNTs with and without acid treatment. The CNT suspension without acid treatment was not stable and separated into two layers after 10 min (Fig. 1a). On the other hand, the acid-treated CNTs were successfully dispersed in distilled water and no sedimentation was observed in the suspension after 24 h, as shown in Fig. 1b. Surface treatment of the CNTs using acid was found to be eective in producing a stable CNT suspension. Fig. 2 shows a TEM micrograph of the HA3 wt.% CNT powder, conrming that the CNTs with a diameter of about 15 nm were well mixed with the HA particles. Furthermore, the selected area electron diraction (SAED) pattern for the HA particle indicated that the particle was single crystal. 3.2. Microstructures and phases of the coatings deposited by aerosol deposition The polished cross-sectional FEG-SEM images of the HA and HA3 wt.% CNT composite coatings are shown in Fig. 3. Irrespective of addition or not of the CNTs, the coatings with a uniform thickness of about 5 lm were very compact and dense. No pores or microcracks were detected in the coatings. Both of the coatings exhibited a similar deposition rate of 5 lm min1 on the coating area of 10 10 mm. These results imply that the addition of CNTs did not signicantly aect the deposition behavior. Furthermore, the coatingsubstrate interfaces were clear and no pores or microcracks could be observed near the interface, suggesting that the adhesive strength of the coating to the substrates was high. Fig. 4a shows a FEG-SEM image of the fracture surface of the HA3 wt.% CNT composite coating. Some CNTs, as
Fig. 2. TEM micrograph of HA3 wt.% CNT powder mixture and SAED pattern for the indicated region.

Fig. 1. Optical micrographs showing the aqueous suspensions containing CNTs (a) without and (b) with acid treatment.

indicated by arrows, were distributed in the HA matrix. The EDS spectra, showing the constituent elements of the HA coatings with and without CNTs, are presented in Fig. 4b and c, respectively. The presence of the CNTs was conrmed by the peaks for carbon in the EDS spectrum of the HACNT composite coating. FEG-SEM micrographs of the surface morphologies of the HA and HA3 wt.% CNT coatings are shown in Fig. 5. The coatings had a similar microstructure with a rough surface. The rough surface of HA bioactive coatings deposited on the surface of metallic implants might improve mechanical xation of the living bone to the implant, due to the increased contact area between the implant and bone [29]. In addition, a rough coating surface has been reported to be helpful in the enhancement of the biological performance of HA-coated implants [30,31]. The surface roughness (Ra) values of the coatings determined by a laser prolometer are presented in Table 1. The Ra of the Ti plate used as substrate was 0.48 lm. The Ra values of the HA, HA1 wt.% CNT and HA3 wt.% CNT coatings were 0.57, 0.59 and 0.55 lm, respectively. No signicant change in the surface roughness of the coatings was observed dependent on the amount of CNTs added. TEM analysis was conducted to conrm the existence of the CNTs in the HACNT composite coating, and the micrographs are shown in Fig. 6. The CNTs maintained their typical tubular structure in the HA matrix, as shown in Fig. 6a, and would therefore be expected to act as an excellent reinforcement in the HA matrix. Fig. 6b is the high-resolution TEM image showing the HA matrix in the HA3 wt.% CNT composite coating. The crystallite size of the HA matrix in the coating was in the range 520 nm. After the coating process the HA crystallite size was reduced to approximately one-tenth of its original value in the starting powder (Fig. 2). When high kinetic energy particles collide with a substrate in the AD process they are fragmented into smaller pieces on the nanometer

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214

3209

Fig. 3. Cross-section micrographs of the polished surfaces of the (a) pure HA and (b) HA3 wt.% CNT coatings on the Ti plate.

(a)

500 nm (b)
cps (A.U.)
P Ca

(c)
P

Ca

Ca C O O

cps (A.U.)

Ca

Energy (keV)

Energy (keV)

Fig. 4. (a) FEG-SEM micrograph of the fracture surface of the HA3 wt.% CNT coating, (b) the corresponding EDS spectrum and (c) the EDS spectrum obtained from the fracture surface of the pure HA coating.

Fig. 5. Surface morphologies of the coatings: (a) pure HA and (b) HA3 wt.% CNT coatings.

3210

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214

Table 1 Surface roughness of the coatings as a function of the amount of CNT addition. HA Ra (lm) 0.57 0.02 HA1 wt.% CNT 0.59 0.05 HA3 wt.% CNT 0.55 0.04

scale and subsequently rebond to each other to form a coating layer [24,32]. This is a typical characteristic of the AD process, which has been similarly reported for other systems such as PZT and TiO2 [24,33]. Fig. 7 shows the XRD patterns of the HA and HA3 wt.% CNT mixture in the form of a powder and coating on the Ti

Fig. 6. (a) TEM bright-eld image of the HA3 wt.% CNT coating and (b) high resolution TEM image of the HA matrix in the HACNT coating.

: HA * : Carbon

(a)
Ti

Ti

Ti

HA coating on Ti

Intensity (A.U.)

(112) (300)

(211)

(222) (113) (203)

(102) (210)

(312) (320)

(200) (111)

(212)

(213) (321) (410) (402) (004)

(002)

(202)

(310)

(201)

(322) (313)

(301)

(311)

Powder
20 30 40 50
o

60

2 Theta ( )

(b)
Ti

Ti Ti

Intensity (A.U.)

HA-3 wt% CNT coating on Ti

*
20 30 40 50
o

(002)

Powder
60

2 Theta ( )
Fig. 7. XRD patterns of the powders and the coatings on the Ti plate: (a) HA and (b) HA3 wt.% CNT.

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214

3211

plate. XRD analysis of the HA powder revealed only those peaks corresponding to stoichiometric HA (JCPDS card No. 9432). In the case of the HACNT powder, besides the HA peaks, a (0 0 2) peak for carbon, indicating the existence of the CNTs, was observed (JCPDS card No. 41 1487). The coatings deposited on the Ti plate exhibited similar XRD patterns, with weak and broadened peaks. The broadened, weak peaks might be due to the small grain size of the coatings, as shown in the TEM images in Fig. 6. 3.3. Mechanical properties of the coatings The adhesion strength of a bioactive coating to a substrate signicantly aects successful implantation and long-term stability of the implant coated with the bioactive material. Table 2 shows the adhesion strength of the coatings with the Ti plate. The coatings exhibited a high adhesion strength, ranging from 27.3 to 29.0 MPa, which was much higher than that of electrophoretically deposited HA coatings [21] and comparable with that of plasma sprayed HA coatings [34]. In addition, no specic trend was discerned between the adhesion strength and the amount of CNTs added to the coatings. The nanoindentation test, which is known to be an eective technique to measure the mechanical properties of composite materials with small volumes and dimensions, was employed to investigate such mechanical properties as the hardness and elastic modulus of coatings with dierent CNT contents. Fig. 8 shows the representative load indentation depth curves for the HA and HACNT composite coatings. All indentations were made to a depth of 0.5 lm. With increasing amount of CNTs smaller indentaTable 2 Adhesion strength of the coatings deposited on Ti plate. HA Adhesion strength (MPa) 28.5 1.7 HA1 wt.% CNT 29.0 1.1 HA3 wt.% CNT 27.3 1.0

70 60 50

HA HA-1wt% CNT HA-3wt% CNT

Load (mN)

40 30 20 10 0 0 100 200 300 400 500 600

Indentation depth (nm)

Fig. 8. Typical loadindentation depth curves for the HA and HACNT composite coatings.

tion depths were observed at the same load. This implies that the hardness and elastic modulus of the coatings were improved as the content of CNTs increased. The hardness and elastic modulus proles of the coatings as a function of the indentation depth were obtained from the loadindentation depth curves (Fig. 9). The average hardness and elastic modulus of the coatings are presented in Table 3. The hardness and elastic modulus of the coatings increased with increasing content of CNTs. When the amount of CNTs was increased to 3 wt.% the hardness and elastic modulus of the coatings were improved by 27% and 11%, respectively, compared with those of the pure HA coating. The presence of well-dispersed CNTs with high stiness and strength in the HA matrix is presumed to strengthen the HA coating considerably. The HA3 wt.% CNT composite coating showed a hardness of 9.02 GPa and elastic modulus of 137.05 GPa, which were higher than those of bulk HA. A hardness of 6.196.76 GPa and an elastic modulus of 122125 GPa for the bulk HA material were reported by other researchers [35].

(a) 12
10

(b) 200
Elastic modulus (GPa)
150

Hardness (GPa)

8 6 4 2 0 0 100 200 300 400 500

100

HA HA-1wt% CNT HA-3wt% CNT

50

HA HA-1wt% CNT HA-3wt% CNT

0 100 200 300 400 500

Displacement (nm)

Displacement (nm)

Fig. 9. Hardness (a) and elastic modulus (b) proles of the coatings as a function of indentation depth.

3212 Table 3 Hardness and elastic modulus of the coatings. HA Hardness (GPa) Elastic modulus (GPa) 7.12 0.42 123.72 6.24

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214

HA1 wt.% CNT 7.89 0.35 128.13 3.17

HA3 wt.% CNT 9.02 0.56 137.05 5.15

3.4. In vitro cell tests Fig. 10 shows representative SEM morphologies of preosteoblast MC3T3-E1 cells grown on the bare Ti surface and on the coatings after 5 h. The cell on the bare Ti did not spread out over the surface, as shown in Fig. 10a. In contrast, on the HA-coated Ti the cell was found to have numerous lopods attached to the surface (Fig. 10b). Furthermore, the cell on the HA3 wt.% CNT composite coating was more spread out and had longer lopods than those on the bare Ti and the HA-coated Ti (Fig. 10c). The results demonstrate that the CNTs in the composite coating promoted pre-osteoblast cell attachment and adhesion. The percentage cytotoxicity for the specimens is shown in Table 4. All of the specimens were found to exhibit relatively low cytotoxicity. The HA3 wt.% CNT composite coating showed lower cytotoxicity than the bare Ti and the pure HA coating. This result suggests that the CNTs incorporated into the composite coating for the present study are not cytotoxic.

Fig. 11 shows the proliferation and ALP activity of the pre-osteoblast MC3T3-E1 cells cultured on each coating for 5 and 10 days. Uncoated Ti samples were used as the control group in the cell experiment. The numbers of cells cultured for 5 days on the pure HA and HACNT composite coatings were higher than on the bare Ti. Moreover, the HA3 wt.% CNT composite coating showed a signicantly higher cell proliferation rate relative to the pure HA coating (P < 0.001). In the case of ALP activity the pure HA and HACNT composite coatings had a signicantly higher ALP activity than that of bare Ti (P < 0.001). The ALP activity of the cells on the coatings markedly increased with increasing CNT content in the coatings. The ALP expression level on the HA3 wt.% CNT composite coating was about 1.5 times higher than that on the pure HA coating. Although there has been a lot of debate about the biocompatibility of CNTs, recent studies have demonstrated that osteoblast cells adhered to and grew and proliferated well on the surface of functionalized multi-walled CNTs and, therefore, that CNTs could be used as a scaold for articial bone growth due to their biocompatible properties [18,3638]. Zanello et al. tried to control cell growth on CNTs by functionalizing the
Table 4 Percentage of cytotoxicity for the specimens. Ti Cytotoxicity (%) 57.6 0.4 HA coating 53.4 4.6 HA3 wt.% CNT 45.6 4.5

Fig. 10. SEM micrographs of pre-osteoblast MC3T3-E1 cells attached to the specimens: (a) bare Ti, (b) pure HA coating and (c) HA3 wt.% CNT composite coating.

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214

1.2 1.2

3213

Cell density ALP activity

1.0

p<0.001 p<0.001

mol PNP/30 minute/mg protein

1.0

Absorbance (490 nm)/cm 2

ALP activity

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0

0.0

Control Ti

HA

HA-1wt% CNT HA-3wt% CNT

Samples
Fig. 11. Proliferation rate and ALP activity of MC3T3-E1 cells cultured on the samples for 5 and 10 days. Bare Ti plates were used as a control.

CNTs. They showed that the neutrally charged CNTs sustained osteoblast growth and bone formation [18]. George et al. investigated the response of human lung epithelial cells and osteoblast cells to multi-walled CNTs. They reported that these cells were well attached and survived on multi-walled CNTs [38]. The results of the present work also demonstrate that CNTs possess sucient biocompatibility for use as biomaterials and the addition of CNTs to an HA coating is remarkably eective in improving the biological cellular response to the coating. 4. Conclusion CNT-reinforced HA composite coatings were successfully deposited on the surface of Ti substrates using AD. Acid treatment of their surface was an eective method of dispersing the CNTs in the HA matrix, and CNTs well mixed with HA particles were used for deposition. SEM images revealed that the coatings were dense and they had good adhesion to the substrate while retaining their structural integrity. The adhesion strengths of the coatings were measured as more than 27.3 MPa, which were much higher than that of electrophoretically deposited HA coatings and comparable with that of plasma sprayed HA coatings. TEM observation showed that CNTs with their typical tubular structure could be observed in the HACNT composite coatings. As the amount of CNTs added to the HA matrix increased the mechanical properties, such as the hardness and elastic modulus, of the coatings were markedly improved and, simultaneously, the biological performance was also signicantly enhanced. The addition of CNTs to HA could remarkably improve both the mechanical and biological properties of the HA coating. The AD method was shown to be very eective in producing dense and highly adhesive HACNT composite coatings. Therefore, HACNT

composite coatings fabricated by AD are expected to be promising for high load-bearing orthopedic implants, such as hip, knee and shoulder joints. Acknowledgement This work was supported by a grant from the Center for Advanced Materials Processing (CAMP) of the 21st Century Frontier R&D Program funded by the Ministry of Knowledge Economy (MKE), Republic of Korea. References
[1] Hench LL. Bioceramics: from concept to clinic. J Am Ceram Soc 1991;74:1487510. [2] Jarcho M. Calcium phosphate ceramics as hard tissue prosthetics. Clin Orthop Relat Res 1981;157:25978. [3] Joshi SV, Srivastava MP, Pal A, Pal S. Plasma spraying of biologically derived hydroxyapatite on implantable materials. J Mater Sci Mater Med 1993;4:2515. [4] With GD, Van Dijk HJA, Hattu N, Prijs K. Preparation, microstructure and mechanical properties of dense polycrystalline hydroxyapatite. J Mater Sci 1981;16:15928. [5] Tsui YC, Doyle C, Clyne TW. Plasma sprayed hydroxyapatite coatings on titanium substrates. Part I: mechanical properties and residual stress levels. Biomaterials 1998;19:201529. [6] Liang H, Shi B, Fairchild A, Cale T. Applications of plasma coatings in articial joints: an overview. Vacuum 2004;73:31726. [7] Sun L, Berndt CC, Gross KA. Hydroxyapatite/polymer composite ame-sprayed coatings for orthopedic applications. J Biomater Sci Polym Ed 2002;13:97790. [8] Gu YW, Khor KA, Pan D, Cheang P. Activity of plasma sprayed yttria stabilized zirconia reinforced hydroxyapatite/Ti-6Al-4V composite coatings in simulated body uid. Biomaterials 2004;25:317785. [9] Evis Z, Doremus RH. Coatings of hydroxyapatite-nanosize alpha alumina composites on Ti-6Al-4V. Mater Lett 2005;59:38247. [10] Iijima S. Helical microtubules of graphitic carbon. Nature 1991;354: 568. [11] Dresselhaus MS, Dai H. Carbon nanotubes: continued innovations and challenges. MRS Bull 2004;29:2379.

3214

B.-D. Hahn et al. / Acta Biomaterialia 5 (2009) 32053214 [26] Shaer MSP, Fan X, Windle AH. Dispersion and packing of carbon nanotubes. Carbon 1998;36:160312. [27] Oliver WC, Pharr GM. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J Mater Res 1992;7:156483. [28] Korzeniewski C, Callewaert DM. An enzyme-release assay for natural cytotoxicity. J Immunol Methods 1983;64:31320. [29] Wernnerberg A. The importance of surface roughness for implant incorporation. Int J Mach Tool Manu 1998;38:65762. [30] Buser D, Schenk RK, Steinemann S, Fiorellini JP, Fox CH, Stich H. Inuence of surface characteristics on bone integration of titanium implants. A histomorphometric study in miniature pigs. J Biomed Mater Res 1991;25:889902. [31] Deligianni DD, Katsala N, Ladas S, Sotiropoulou D, Amedee J, Missirlis YF. Eect of surface roughness of the titanium alloy Ti-6Al4V on human bone marrow cell response and on protein adsorption. Biomaterials 2001;22:124151. [32] Akedo J. Aerosol deposition of ceramic thick lms at room temperature: densication mechanism of ceramic layers. J Am Ceram Soc 2006;89:18349. [33] Ryu J, Park DS, Hahn BD, Choi JJ, Yoon WH, Kim KY, et al. Photocatalytic TiO2 thin lms by aerosol-deposition: from micronsized particles to nano-grained thin lm at room temperature. Appl Catal 2008;B83:17. [34] Yang Y, Ong JL. Bond strength, compositional, and structural properties of hydroxyapatite coating on Ti, ZrO2-coated Ti, and TPScoated Ti substrate. J Biomed Mater Res 2003;A64:50916. [35] Kumar RR, Wang M. Modulus and hardness evaluations of sintered bioceramic powders and functionally graded bioactive composites by nano-indentation technique. Mater Sci Eng 2002;A338:2306. [36] Zhao B, Hu H, Mandal SK, Haddon RC. A bone mimic based on the self-assembly of hydroxyapatite on chemically functionalized singlewalled carbon nanotubes. Chem Mater 2005;17:323541. [37] Correa-Duarte MA, Wagner N, Rojas-Chapana J, Morsczeck C, Thie M, Giersig M. Fabrication and biocompatibility of carbon nanotube-based 3D networks as scaolds for cell seeding and growth. Nano Lett 2004;4:22336. [38] George JH, Shaer MS, Stevens MM. Investigating the cellular response to nanobrous materials by use of a multi-walled carbon nanotube model. J Exp Nanosci 2006;1:112.

[12] Peigney A. Composite materials: tougher ceramics with nanotubes. Nat Mater 2003;2:156. [13] Treacy MMJ, Ebbesen TW, Gibson JM. Exceptionally high Youngs modulus observed for individual carbon nanotube. Nature 1996;381: 67880. [14] Salvetat JP, Bonard JM, Thomson NH. Mechanical properties of carbon nanotubes. Appl Phys A Mater Sci Proc 1999;63:25560. [15] Curtin WA, Sheldon BW. CNT-reinforced ceramics and metals. Mater Today 2004;7:449. [16] Zhan GD, Kuntz JD, Wan J, Mukherjee AK. Single-wall carbon nanotubes as attractive toughening agents in alumina-based nanocomposites. Nat Mater 2003;2:3842. [17] Chlopek J, Czajkowska B, Szaraniec B, Frackowiak E, Szostak K, Beguin F. In vitro studies of carbon nanotubes biocompatibility. Carbon 2006;44:110611. [18] Zanello LP, Zhao B, Hu H, Haddon RC. Bone cell proliferation on carbon nanotubes. Nano Lett 2006;6:5627. [19] Usui Y, Aoki K, Narita N, Murakami N, Nakamura I, Nakamura K, et al. Carbon nanotubes with high bone-tissue compatibility and bone-formation acceleration eects. Small 2008;4:2406. [20] Balani K, Anderson R, Laha T, Andara M, Tercero J, Crumpler E, et al. Plasma-sprayed carbon nanotube reinforced hydroxyapatite coatings and their interaction with human osteoblasts in vitro. Biomaterials 2007;28:61824. [21] Kaya C, Singh I, Boccaccini AR. Multi-walled carbon nanotubereinforced hydroxyapatite layers of Ti6Al4V medical implants by electrophoretic deposition (EPD). Adv Eng Mater 2008;10:1318. [22] Sun L, Berndt CC, Gross KA, Kucuk A. Material fundamentals and clinical performance of plasma-sprayed hydroxyapatite coatings: a review. J Biomed Mater Res 2001;B58:57092. [23] Wei M, Ruys AJ, Milthorpe BK, Sorrell CC. Precipitation of hydroxyapatite nanoparticles: eects of precipitation method on electrophoretic deposition. J Mater Sci Mater Med 2005;16:31924. [24] Hahn BD, Park DS, Choi JJ, Yoon WH, Ryu J, Kim DY. Eects of Zr/Ti ratio and post-annealing temperature on the electrical properties of PZT thick lms fabricated by aerosol deposition. J Mater Res 2008;23:22635. [25] Hahn BD, Kim KH, Park DS, Choi JJ, Ryu J, Yoon WH, et al. Fabrication of lead zirconate titanate thick lms using a powder containing organic residue. Jpn J Appl Phys 2008;47:554552.