-
Ion-beam-sputtering/mixing deposition of calciumphosphate
coatings. I. Effects of ion-mixing beams
C.-X. Wang,1,2 Z.Q. Chen,1 M. Wang,2 Z.Y. Liu,3 P.L. Wang
31Department of Dental Materials, College of Stomatology, West
China University of Medical Sciences, Chengdu610041, Sichuan,
China2School of Mechanical and Production Engineering, Nanyang
Technological University, Nanyang Avenue, Singapore639798,
Singapore3Laboratory of Radiation Physics and Technology, Institute
of Nuclear Science and Technology, Sichuan University,Chengdu
610064, Sichuan, China
Received 26 November 1999; revised 11 December 2000; accepted 14
December 2000
Abstract: Ion-beam-sputtering/mixing deposition wasused to
produce thin calcium phosphate coatings on tita-nium substrate from
the hydroxyapatite target. The mixingbeam could be either Ar+ or N+
ions. It was found thatas-deposited coatings were amorphous. No
distinct peak ofthe hydroxyl group was observed in FTIR spectra of
thecoatings, but new spectral peaks, brought about during
thedeposition process, were present for CO3
2−. Scanning elec-tron microscopy revealed that the deposited
coatings had auniform and dense structure. The
calcium-to-phosphorousratio of these coatings varied between 2.0
and 3.0. Compared
with the calcium phosphate coatings produced by Ar+
beam-mixing deposition, the calcium phosphate coatingsproduced
by N+ beam-mixing deposition exhibited a higherdissolution rate in
the physiologic saline solution andshowed a lower proliferation
rate of osteoblast cells. © 2001John Wiley & Sons, Inc. J
Biomed Mater Res 55: 587–595,2001
Key words: ion-beam-sputtering/mixing deposition; cal-cium
phosphate coating; titanium; dissolution rate; cell cul-ture
INTRODUCTION
Due to their good biocompatibility and excellentmechanical
properties, calcium phosphate ceramicssuch as hydroxyapatite- (HA)
coated titanium and itsalloys are among the most promising implant
materi-als for orthopedic and dental applications.1–3 Cur-rently,
commercially available coated metal implantsare manufactured by
using the plasma-spray tech-nique for depositing HA coatings.
However, prob-lems, such as low bond strength between the
coatingand the substrate and nonuniformity across the thick-ness of
the coating, often are encountered withplasma-sprayed coatings.4 A
variety of other coatingtechniques, such as electrochemical
deposition,5 radio-frequency magnetron sputtering,6 excimer laser
depo-sition,7 pulsed laser deposition,8 and dipping9 also
have been investigated for producing HA coatings onmetallic
substrates. In addition to bioactivity, a cal-cium phosphate
coating satisfactory for clinical usemust be dense, hard, adherent,
and tough.10 Since themid-1970s, many surface modification
techniquesbased on ion implantation, such as ion-beam deposi-tion
(IBD), ion-beam-assisted deposition (IBAD), ion-beam mixing (IBM),
and techniques that are based onplasma-assisted ion implantation,
such as plasma-source ion implantation (PSII) and
plasma-immersionion implantation (PIII), have been developed and
noware used widely to modify the surfaces of materialssuch as
metals, ceramics, and polymers.11 Ion-beam-sputtering deposition
has been investigated as amethod for producing biocompatible
ceramic coatingson metallic implants because it produces thin
coatingswith high density and superior adhesion.12,13 In
thisprocess, ionized argon gas was used to sputter atomsfrom a
ceramic target. The sputtered atoms built up onthe metallic
substrate that was placed in the path ofthe sputtered material.
Both argon ion beam and nitrogen ion beam can beused for
ion-beam bombardment. It has been foundthat grafting −NH2
+ amidogen radicals on Al2O3 ce-
Correspondence to: C.-X. Wang, NTU, Singapore;
e-mail:[email protected]
Contract grant sponsor: National 863 Program of
NewTechnology
Contract grant sponsor: National Natural Science Foun-dation of
China
© 2001 John Wiley & Sons, Inc.
-
ramic using an ion implantation technique can im-prove the
biocompatibility of this ceramic.14 By usinga nitrogen ion beam to
bombard Ca-P coatings, it ispossible to further improve the bone
bonding betweenthe coating and the bone to hasten the wound
healing.Moreover, during the bombardment process, nitrogenions are
able to penetrate the whole coating layer andbond with the titanium
substrate to form TixNy,
15
which improves the properties of the coatings. In
thisinvestigation, the ion-beam-sputtering/mixing depo-sition
technique was used to produce thin calciumphosphate coatings on
titanium substrate. The influ-ences of argon and nitrogen
ion-mixing beams on thestructure and properties of the resultant
coatings arereported in this paper.
MATERIALS AND METHODS
Deposition method
A commercially available pure titanium substrate
(platesdimensions 20 × 20 × 1 mm) was mechanically polished
andultrasonically cleaned with acetone and alcohol. HA powderwas
synthesized using the wet method. HA powder disks,which were to be
used as the ion-beam-sputtering target,were cold pressed and
sintered at 1150°C for 3 h. X-raydiffraction patterns of HA disks
matched those of the stan-dard synthetic hydroxyapatite (PDFSM #
9-0432), indicatingthat the disks had well-crystallized HA with a
microstruc-ture of purely random grain orientation.
The ion-beam-sputtering/mixing deposition system con-sisted
mainly of one Kaufman ion source and one Freemanion source, one
target holder, and one rotatable sampleholder in the path of both
ion beams (Fig.1). The depositionchamber was evacuated to a base
pressure of 2.8∼3.7 × 10−4
Pa. Prior to deposition, etching of the substrates with 800
eVand 40 mA/cm2 argon ions was performed for 30 min toclean the
surface of the titanium substrate. The energetic ionbeam was
produced by ionizing high-purity argon gas(99.999% pure). After
cleaning, the stage was rotated so thatthe substrates were placed
in the path of the sputtered at-
oms. The monolayer calcium phosphate coatings were
sput-ter-deposited by an Ar+ beam with 1200 eV and 40 mA/cm2
for 60 min. The Ar+ beam sputtering- and Ar+ beam
mixing-deposited calcium phosphate coatings were produced firstby
sputtering the HA target with the 1400 eV and 40 mA/cm2 Ar+ beam
for 1.5 h and then by using the second Ar+
beam with 60 keV to homogenize the coating. The dosage ofthe Ar+
mixing beam was 2.0 × 1016 ions/cm2. For the Ar+
beam sputtering- and N+ beam mixing-deposited calciumphosphate
coatings, after sputtering the HA target with the1400 eV and 40
mA/cm2 Ar+ beam for 1.5 h, the N+ beamwith 60KeV was used to
homogenize the coating, and itsdosage also was 2.0 × 1016 ions/cm2.
The coated samplesused for Fourier transform infrared spectroscopy
(FTIR) hadKCl crystal substrates and their coatings were
sputtering/mixing-deposited from an HA target as well.
Characterization of coatings
X-ray diffraction (XRD) was employed to analyze thestructure of
as-deposited coatings. A Rigaku D/max-tA X-ray diffractometer with
Cu Ka radiation at 40 keV and 50∼80mA was used. FTIR analysis was
performed on a NicoletFTIR 20SXB machine for characterizing various
functionalgroups of the coatings, especially the hydroxyl and
phos-phate groups. The FTIR spectra were obtained using
thetransmittance mode from 4000 to 400 cm−1. The surface
mor-phology of coatings was examined by using a scanning elec-tron
microscope (SEM; Hitachi X-650, Hitachi, Japan) and ascanning
tunneling microscope (STM; Explorer STM/AFM,TopoMetrix Co., USA).
To prevent charging, the samples forSEM observations were coated
with a thin layer of carbon.The elemental composition of coatings
was determined byenergy dispersive X-ray analysis (EDX).
Transmission electron microscopy (TEM) also was used toexamine
the microstructure of as-deposited coatings. Be-cause of the
brittle nature of thin Ca-P coatings, it was verydifficult to
prepare suitable specimens for TEM observationdirectly from
coatings with titanium substrate by using anion beam thinning
apparatus. An alternative method wasused, and TEM specimens
successfully were produced. First,the Ca-P coatings were deposited
on KCl crystal substratesfrom a HA target. The coated samples then
were put into abeaker containing distilled water. As a result, the
calciumphosphate coatings could be peeled off after the
dissolutionof the KCl crystals. The peeled-off Ca-P coatings were
ex-amined under a TEM (JEOL TEM-100CX, Japan).
Dissolution rate test
For the dissolution tests, the coated samples (with thetitanium
as the substrate) were incubated in a physiologicsaline solution
(0.9% NaCl) at pH 7.2 and at a temperature of37°C. At regular
intervals, the solution was analyzed for theCa2+ concentration. The
Ca2+ concentration in the solutionwas determined by flame atomic
adsorption spectrometry
Figure 1. Schematic diagram showing the ion beam
sput-tering/mixing deposition technique: 1 = lower energy ionsource
for sputtering; 2 = higher energy ion source for mix-ing; 3 =
sample holder; 4 = target holder.
588 WANG ET AL.
-
(Hitachi 180-80, Hotachi, Japan). Five replicates were
com-pleted and expressed as means ± standard deviations.
Osteoblast cell culture
For the osteoblast cell culture tests, both the Ar+ beammixing-
and the N+ beam mixing-deposited Ca-P coatingswere used. Pure
titanium metal plates also were used as thepositive control. All
samples were sterilized under dry heatat 180°C for 2 h. The
osteoblast cell line MC3T3-E1 (suppliedby the Fourth Military
University of Medical Sciences, Xi’an,China) was used for in vitro
tests. It was cultured in DMEM(Gibco BRL, USA), containing 10%
fetal bovine serum in ahumidified 5% CO2–air atmosphere at 37°C.
Cells wereseeded on each specimen at 6000 cells/plate in 96-well
tissueculture plates. The intracellular activity of MC3T3-E1
wasmeasured by the
3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)
formazan assay. Ab-sorbance at 490 nm of MTT-formazan in dimethyl
sulphox-ide (DMSO) was measured using the Bio-Kinetics
readermicroplate. The cell morphology was observed under SEM.
Statistical analysis
All measurements were collected in five replicates andexpressed
as means ± standard deviations. The differencesbetween tested
coatings were evaluated by Student’s t testand considered
statistically significant with P < 0.05.
RESULTS
Characterization of coatings
A typical XRD pattern of as-deposited monolayercalcium phosphate
coatings from the HA target is
shown in Figure 2. As can be seen from this pattern,the coatings
only exhibited a broad bump and nopeaks other than those of
titanium substrate were ob-served, indicating that as-deposited
coatings wereamorphous.
The microstructure of as-deposited coatings was ex-amined under
TEM. The TEM analysis provided in-formation on both the phases and
the structure ofcoatings on a microscopic scale. Figure 3(a) shows
thebright field TEM micrograph of the coating depositedwith the
ion-beam energy of 1.2 keV and 40 mA. Thecorresponding selected
area diffraction (SAD) patternis shown in Figure 3(b). A single
halo in the SADpattern indicated that the as-deposited coatings
wereamorphous, which confirmed XRD results.
FTIR was used mainly to characterize the hydroxyland phosphate
groups in the as-deposited coatings.According to Dasarathy et al.3
and Walters et al.,16 inthe molecular structure of HA, the
phosphate groupitself had a Td symmetry, resulting in four
internalmodes (symmetric stretch n1: 956 cm
−1; asymmetricstretch n2: 430–460 cm
−1; bending n3: 1040–1090 cm−1;
bending n4: 575-–610 cm−1); and the hydroxyl group
with C`n had vibration modes at wavenumbers of3570 and 630 cm−1.
Figure 4 and Figure 5 display typi-cal FTIR spectra of the Ar+ beam
sputtering and Ar+
beam mixing Ca-P coating and the Ar+ beam sputter-ing and N+
beam mixing Ca-P coating from HA tar-gets that were deposited on
the KCl crystal substrate,respectively. In Figure 4, spectral peaks
at 1034 and565cm−1 were observed, indicating the existence
ofPO4
3− in the as-deposited Ca-P coating. No distinctspectral peaks
of the hydroxyl group were observedexcept for a peak at 3494 cm−1
for water. New peaks at1496, 1416, and 939 cm−1 were present for
CO3
2−,which was brought about by the deposition process.From Figure
4 and Figure 5, it can be seen that therewas no shape change in
FTIR spectra for Ca-P coatings
Figure 2. XRD pattern of as-deposited Ca-P coating from the HA
target.
589EFFECTS OF ION-MIXING BEAMS
-
produced by Ar+ beam and N+ beam mixing deposi-tions. However,
there were some alterations in the ab-sorption bands at 1034 and
565 cm−1, which were forthe Ca-P coating produced by Ar+ beam
mixing de-position. These peaks moved to 1032 and 577 cm−1 forthe
Ca-P coating produced by N+ beam mixing pro-cess.
Figure 6 shows the surface morphology of Ca-Pcoatings deposited
on titanium substrate under differ-ent conditions. These coatings
exhibited a clean andsmooth surface morphology. Energy-dispersive
X-rayspot analysis (EDX) demonstrated that the obtainedcoatings
consisted of calcium and phosphorous. The
semiquantitative EDX analysis of relative amounts ofcalcium and
phosphorous revealed that the Ca/P ratioof the coatings varied
between 2.0 and 3.0. The Ca-Pratio for the coating produced by Ar+
beam and Ar+
beam mixing depositions was 2.5, and it was 3.0 forthe coating
produced by Ar+ beam and N+ beam mix-ing depositions.
Scanning tunneling microscopy (STM) is a relativelynew technique
for mapping a surface with high reso-lution. A probe can be
sharpened to a few angstromsin radius at the tip and then brought
to within about 2nm above a flat surface. Piezoelectric mounts in
theinstrument control both lateral and up-and-down
Figure 3. TEM micrograph of as-deposited Ca-P coating on the KCl
crystal substrate: (a) bright field image; (b) correspond-ing
selected area diffraction (SAD) pattern.
Figure 4. FTIR spectrum of Ca-P coating produced by Ar+ beam
sputtering and Ar+ beam mixing deposition from the HAtarget.
590 WANG ET AL.
-
movements. This technique is based on the electron-tunneling
phenomenon; that is, a current flows be-tween the probe and the
surface due to an overlap ofrespective wave functions. The minute
current, nano-amperes or less, varies exponentially with the
tip-sample separation, about tenfold per angstrom. Thetunneling
current is used to control the up-and-downmovement. As the surface
is scanned, a relief map (i.e.,image) of the surface is obtained,
with a resolutiondown to the atomic scale. A major advantage of
STMover other imaging techniques is that the specimen
does not need to be in vacuum; it can be in air orimmersed in
water or in some other fluids. STM can beused to study conducting
and semiconducting sur-faces. In the present study, STM was used to
investi-gate the surfaces of as-deposited Ca-P coatings. Figure7
shows typical STM images of as-deposited Ca-Pcoating on titanium
substrate using the HA target.These images give detailed
information of coating sur-faces at the atomic scale. The
accumulation of sput-tered atoms and the formation of the coating
layerclearly were shown. It also was observed that the
Figure 5. FTIR spectrum of Ca-P coating produced by Ar+ beam
sputtering and N+ beam mixing deposition from the HAtarget.
Figure 6. SEM micrographs of Ca-P coatings deposited on titanium
substrate: (a) coating produced by Ar+ beam sputteringand Ar+ beam
mixing deposition, dosage 2 × 1016 ions/cm2 (original magnification
×1000); (b) coating produced by Ar+ beamsputtering and N+ beam
mixing process, dosage 2 × 1016 ions/cm2 (original magnification
×1000).
591EFFECTS OF ION-MIXING BEAMS
-
grains in the coatings were very small (i.e., at nanom-eter
scale).
Dissolution rate
Figure 8 shows the Ca2+ concentration of differentCa-P coatings
in physiologic saline solution (0.9%NaCl) during a 5-h incubation
period. It can be ob-served that, compared with coatings produced
by Ar+
beam mixing deposition, coatings produced by N+
beam mixing deposition always exhibit higher Ca2+
concentration in the saline solution during the 5-h test-ing
period.
Osteoblast cell culture
The intracellular activity of MC3T3-E1 cells on eachmaterial is
shown in Figure 9. Statistically significantdifferences (P <
0.001) were observed in two testedcoatings. However, no statistical
differences (P =0.137) were observed in the same materials
incubated
for different times. After 2 days of incubation, the
in-tracellular activity of MC3T3-E1 cells on the Ca-P coat-ing
produced by Ar+ ion beam mixing depositionreached the peak rate.
After 3 days of incubation, theintracellular activity of MC3T3-E1
cells on the samematerial slowed down sharply. The same phenom-enon
was observed on the negative control sample. Asfor the other two
materials, namely, titanium and theCa-P coating produced by N+ beam
mixing deposi-tion, the intracellular activity of the osteoblastic
cellswas increasing during the incubation period. Com-pared with
the coating produced by Ar+ beam mixingdeposition, the
intracellular activity of cells incubatedon titanium and the
coating produced by N+ beammixing deposition were significantly
lower. Figure 10exhibits the morphology of MC3T3-E1 cells on
differentmaterials. No major difference was noted for thesecells
with coatings tested.
DISCUSSION
XRD and TEM analyses showed that as-depositedCa-P coatings were
amorphous. This “amorphous”
Figure 7. STM images of the Ca-P coating produced by Ar+ beam
sputtering and Ar+ beam mixing deposition, dosage 2 ×1016
ions/cm2.
Figure 8. Ca2+
concentration resulting from different Ca-Pcoatings in
physiologic saline solution (pH 7.2, 37°C).
Figure 9. Intracellular activity of MC3T3-E1 cells on differ-ent
materials as a function of incubation time.
592 WANG ET AL.
-
appearance is a direct consequence of the ion-beamdeposition
technique. Because of the high vacuum anda relatively low
temperature as compared to theplasma-spray coating method, there
was not enoughenergy for the growth of nano-crystallites in the
coat-ings during the ion-beam-sputtering/mixing deposi-tion
process, and hence the as-deposited coatings wereshown to be
“amorphous.” This observation was veri-fied by the STM examination.
The STM results re-vealed that the size of crystallites in the
as-depositedcoatings was at the nanometer scale. The accumula-tion
of the nano-crystallites also was shown in theSTM images. Using
postdeposition heat treatment forthe amorphous Ca-P coatings, the
crystallinity of theCa-P coatings could be increased.13 This means
thatthe heat treatment provides energy for the growth
ofnano-crystallites, and as a result, the crystallinity is
shown to be “increased.” In a separate study, part ofthe XRD
peaks of HA was observed after heat treat-ment, and some
crystal-like morphology was apparentin the surface of the
coatings.17
Variations for the phosphate group and the loss ofthe hydroxyl
group in FTIR spectra also were causedby the ion-beam deposition
technique. During the de-position process, components of the
target, such ashydroxide or oxygen and hydrogen, may not be
trans-ferred completely to the substrate (or anchored ontothe
substrate surface and thus remaining in the coat-ings) because of
the requirement of maintaining a lowpressure in this particular
coating method.
The wide variation in the Ca-P ratio among coatingswas
attributed to the following: on the one hand, asphosphorus is very
volatile, the low pressure duringthe sputtering process affected
the anchoring of the
Figure 10. SEM micrographs of MC3T3-E1 cells incubated on
different materials: (a) Ca-P coating produced by Ar+ ion beam
sputtering and Ar+ beam mixing deposition (original
magnification ×2000); (b) Ca-P coating produced by Ar+ beam
sput-tering and N+ beam mixing deposition (original magnification
×2000); (c) pure titanium (original magnification ×2000); (d)Ca-P
coating produced by Ar+ beam sputtering and N+ beam mixing
deposition (no cells seeded, original magnification×2000).
593EFFECTS OF ION-MIXING BEAMS
-
sputtered phosphorus on the substrate; in addition,during the
sputtering and the deposition process, oneAr+ ion sputtered
different phosphorus and calciumatoms, indicating that the
sputtering rate of calciumatoms and that of phosphorus atoms are
different. Thehigher sputtering rate of calcium atoms led to a
Ca-Pratio higher than the standard 1.67 of pure HA crystal.On the
other hand, according to Cotell,18 one possiblecause was the
substitution of the carbonate group forthe phosphate groups in the
HA molecule. The pro-cess of ion-beam-sputtering HA target had been
stud-ied using theoretical analysis.19 The results showedthat
because of the collision of sputtering ions and HAmolecules,
defects, such as lattice displacement andvacancy, were produced in
the coating. These defectsmade it possible for the substitution of
carbonategroups for the phosphate groups in the HA molecule.
The higher Ca2+ concentration in the saline solutionresulted
from the coating produced by the N+ beammixing process and suggests
that this coating is moresoluble and dissolves faster than the
coating producedby the Ar+ beam mixing process. Previous
studieshave shown that the dissolution of Ca-P coatings
isinfluenced by the crystallinity of the deposited coat-ing.20,21
In the present study, it was found that themorphology of the
coating after immersion in aqueoussolutions was the major factor
affecting the dissolutionrate of the coating: the more cracks that
were pro-duced on the coating surface, the higher the dissolu-tion
rate of the coating. Yoshinari et al.22 studied thedissolution rate
of calcium phosphate coatings afterrapid heating. They too found
that surface morphol-ogy affects the solubility of the
coatings.
It was confirmed that the MTT–formazan formationis proportional
to the number of living cells in therange from 104 to 106 cells.
After 2 days of incubation,the intracellular activity of MC3T3-E1
cells on theamorphous calcium phosphate coating produced bythe Ar+
beam mixing process reached peak intracellu-lar activity. After 3
days of incubation, the intracellu-lar activity of MC3T3-E1 cells
on the same materialslowed down drastically. The same phenomenon
wasobserved in the negative control sample. There wereno
differences in the trend of the intracellular activitybetween the
coating produced by the Ar+ beam mix-ing process and the negative
control samples. As forthe other two materials, that is, titanium
and the amor-phous calcium phosphate coating produced by N+
beam mixing process, the intracellular activity of
theosteoblastic cells increased during the incubation pe-riod.
However, compared with the coating producedby the Ar+ beam mixing
process, the intracellular ac-tivity of cells incubated on titanium
and the coatingproduced by the N+ beam mixing process were
sig-nificantly lower. Although no major difference in themorphology
of cells was noted among the materialstested during the in vitro
test, the cells appeared to be
in the early stage of confluency on the coating pro-duced by the
Ar+ beam mixing process.
It seems that no major differences exist between thecalcium
phosphate coatings produced by the Ar+
beam mixing and the N+ beam mixing processes.However, the Ca-P
coatings produced by the N+ beammixing process exhibited a higher
dissolution rate andcaused a slower cellular activity of the
osteoblasticcells. And preliminary animal tests have shown
thatthere is a faster bone bonding between the implantsproduced by
Ar+ beam mixing and N+ beam mixingprocesses than there is with
coatings produced by theAr+ beam mixing and the Ar+ beam mixing
processes .23
The possible reasons may be, first, that because themass of N+
ion is smaller than that of Ar+ ion, the sizeof nano-crystallites
in the coatings produced by Ar+
beam mixing and N+ beam mixing processes wassmaller. Second, the
higher Ca-P ratio in the coatingsproduced by Ar+ beam mixing and N+
beam mixingprocesses also contributed to the differences in
disso-lution rate and intracellular activity. Further experi-ments
that include the use of other analytical Ar+
beam mixing and N+ beam mixing processes tech-niques need to be
performed to obtain more informa-tion about the composition and
structure of the depos-ited Ca-P coatings. Such experiments may
reveal thedifferences in the calcium phosphate coatings pro-duced
by the two mixing processes.
CONCLUSIONS
Dense and homogeneous Ca-P coatings successfullywere produced on
titanium substrate using the ion-beam deposition technique. The
results showed thatthe as-deposited coatings were amorphous due to
thesmall crystallite size. In comparison with the HA ce-ramic
target, some variations in the chemical compo-sition of the
coatings were brought about in the depo-sition process, such as the
distortion of the phosphatelattice, loss of hydroxyl groups, and
the incorporationof CO3
2−. Although no structure difference was foundin Ca-P coatings
produced by Ar+ beam and N+ beammixing processes, the Ca-P coatings
produced by theN+ beam mixing process were more soluble than
theCa-P coatings produced by the Ar+ beam mixing pro-cess. In the
in vitro test, the osteoblastic cells appearedto be in the early
stage of confluency on the Ca-Pcoatings produced by the Ar+ beam
mixing process.Further experiments are required to obtain more
in-formation about the composition and structure of thedeposited
Ca-P coatings.
Professor Deng Li of College of Pharmacy, West ChinaUniversity
of Medical Sciences, is thanked for assistance inthe cell culture
experiment.
594 WANG ET AL.
-
References
1. Geesink RGT, de Groot K, Klein CPAT, Serekian P. Bonding
ofbone to apatite-coated implants. J Bone Joint Surg
1988;70:17–22.
2. Breme J, Zhou Y, Groh L. Development of a titanium
alloysuitable for an optimized coating with hydroxyapatite.
Bioma-terials 1995;16:239–244.
3. Dasarathy H, Riley C, Coble HD. Analysis of apatite
depositson substrates. J Biomed Mater Res 1993;27:477–482.
4. Lacefield WR. Hydroxyapatite coatings. In: Ducheyne P,
Lem-ons JE, editors. Bioceramics: Material characteristics versus
invivo behavior. New York: New York Academy of Sciences;1988. p
72–80.
5. Monma H. Electrochemical deposition of
calcium-deficientapatite on stainless steel substrate. J Ceram Soc
Japan, Int Edn,1993;101:718–720.
6. Wolke JGC, der Waerden VJPC, de Groot K, Jansen JA.
Stabil-ity of radiofrequency magnetron sputtered calcium
phosphatecoatings under cyclically loaded conditions. Biomaterials
1997;18:483–488.
7. Singh RK, Qian F, Nagabushnam V, Damodaran R, Moudgil
B.Excimer laser deposition of hydroxyapatite thin films.
Bioma-terials 1994;15:522–528.
8. Wang CK, Chern Lin JH, Ju CP, Ong HC, Chang RPH. Struc-tural
characterization of pulsed laser-deposited hydroxyapa-tite film on
titanium substrate. Biomaterials 1997;18:1331–1338.
9. Wen HB, de Wijin JR, Cui FZ, de Groot K. Preparation
ofbioactive Ti6Al4V surface by a simple method.
Biomaterials1998;19:215–221.
10. Wang M, Yang XY, Khor KA, Wang Y. Preparation and
char-acterization of bioactive monolayer and functionally
gradedcoatings. J Mater Sci: Mater Med 1999;10:269–273.
11. Zhang QY, Long ZH, Ren CS, Guo BH, Xu DQ, Ma TC. Multi-beam
mixing implantation system and its applications. SurfCoat Tech
1998;103/104:195–199.
12. Ong JL, Lucas LC, Lacefield WR, Rigney ED. Structure,
solu-bility and bond strength of thin calcium phosphate
coatings
produced by ion beam sputter deposition. Biomaterials
1992;13:249–254.
13. Ong JL, Lucas LC. Post-deposition heat treatments for
ionbeam sputter deposited calcium phosphate coatings. Biomate-rials
1994;15:337–341.
14. Zhao Q, Zhai GJ, Ng DH, Zhang XZ, Chen ZQ. Surface
modi-fication of Al2O3 bioceramic by NH2+ ion implantation.
Bioma-terials 1999;20:595–599.
15. Wang CX. Unpublished data.16. Walters MA, Leung YC,
Blumenthal NC, Legeros RZ, Konsker
KA. A Raman and infrared spectroscopic investigation of
bio-logical hydroxyapatite, J Inorg Biochem 1990;39:193–200.
17. Chen ZQ, Wang CX, Guan LM, Wang M. Effects of heat
treat-ment on the structure and dissolution of ion beam
sputterdeposited calcium phosphate coatings. In: Proceedings of
the8th International Conference on Processing and Fabrication
ofAdvanced Materials, Singapore; 1999.
18. Cotell MC. Pulsed laser deposition and processing of
biocom-patible hydroxyapatite thin films. Appl Surf Sci
1993;69:140–148.
19. Wang CX, Chen ZQ, Zhang HL. Design of HA/Ti
biomedicalimplants using ion beam assisted deposition. In:
Proceedingsof the 20th Annual International Conference of the IEEE,
HongKong; 1998. p 2890–2892.
20. Maxian SH, Zawadsky JP, Dunn MG. In vitro evaluation
ofamorphous calcium phosphate and poorly crystallized
hy-droxyapatite coatings on titanium implants. J Biomed MaterRes
1993;27:111–118.
21. Wolke JGC, Dijk KV, Schaeken HG, de Groot K, Jansen JA.Study
of the surface characteristics of magnetron-sputter cal-cium
phosphate coatings. J Biomed Mater Res 1994;28:1477–1484.
22. Yoshinari M, Watanabe Y, Ohtsuka Y, Derand T.
Solubilitycontrol of thin calcium phosphate coatings with rapid
heating.J Dent Res 1997;76:1485–1494.
23. Wang CX, Chen ZQ, Guan LM. Preliminarily histologicalstudy
on calcium phosphate coatings produced by ion beamsputtering/mixing
deposition. To appear.
595EFFECTS OF ION-MIXING BEAMS