J. Am. Ceram. Soc.
, 89 [8] 2466–2472 (2006)
Journal
Processing and Performance of Hydroxyapatite/Fluorapatite Double
Layer Coating on Zirconia by the Powder Slurry Method
Department of Dental Biomaterials, School of Dentistry, Dankook University, Cheonan 330-714, Korea
Byung-Ho Yoon, Young-Hag Koh,* and Hyoun-Ee Kim*,w
School of Materials Science and Engineering, Seoul National University, Seoul 151-742, Korea
A hydroxyapatite/fluorapatite (HA/FA) double layer was coat-
ed on ZrO2 by a powder slurry method. The FA layer between
the HA layer and the ZrO2 substrate was effective in suppress-
ing the reaction between HA and ZrO2. The rheological prop-
erties of the slurry and the thermal treatment conditions were
optimized. Addition of small amounts of tri-ethyl phosphate dis-
persant and polyvinylbutyral binder (up to 5 wt%) was effective
in reducing the slurry viscosity, even at high loadings of HA
powder. Each layer (HA, FA) was deposited on ZrO2 repeatedly
to induce a uniform layer, and the final heat treatment was car-
ried out above 12001C to consolidate the coating layer. During
this process, without the FA-intermediate layer, a severe reac-
tion between HA and ZrO2 occurred to form tri-calcium phos-
phate and CaZrO3 products. However, the presence of the FA
layer between HA and ZrO2 effectively suppressed the reaction
up to 13001C. The obtained HA/FA double coating layer was
micro-porous and relatively rough, but was firmly adhered to the
ZrO2 substrate, having a bonding strength of approximately 25
MPa after heat treatment above 12001C. The osteoblast-like
cells cultured on the HA/FA coating layer spread and prolifer-
ated favorably, having a cell proliferation rate comparable to
that of a plastic control and HA bulk ceramic.
I. Introduction
B ECAUSE of its excellent biocompatibility, hydroxyapatite
[HA, Ca10(PO4)6(OH)2] ceramic has been used in orthoped-
ics and dental surgeries in the form of bulk ceramics as well as
powders or granules.1–3 However, the low mechanical properties
of bulk HA restricted its wider applications in the load-bearing
compartments. Therefore, HA coatings on metals (Ti, Ti-based
alloys, and stainless steel) or bioinert ceramics (ZrO2 and Al2O3)
have been introduced to achieve a high mechanical strength as
well as a suitable biocompatibility of the system.4–6
The currently used coating methods, such as plasma spraying,
sol–gel synthesis, physical vapor deposition, and immersion
technique, have their own advantages as well as shortcomings.
The coating layer by the plasma-spraying technique is thick
(B50–200 mm) and relatively dense; however, phase instability
and non-uniformity are some of its weaknesses.7,8 The films
synthesized by sol–gel and physical vapor deposition methods
are relatively uniform and the fabrication method is simple;
however, the layer is too thin (B1 mm) to be applied for long-
term usage.9,10 Apatite layers obtained by an immersion tech-
G. Fischman—contributing editor
Manuscript No. 21608. Received February 26, 2003; approved March 21, 2006.
This research was supported by a grant (code no.: 05K1501-01510) from ‘‘Center for
Nanostructured Materials Technology’’ under ‘‘21st Century Frontier R&D Programs’’ of
the Ministry of Science and Technology, Korea.
*Member, American Ceramic Society.
w
Author to whom correspondence should be addressed. e-mail: kimhe@snu.ac.kr
2466
DOI: 10.1111/j.1551-2916.2006.01114.x
r 2006 The American Ceramic Society
Hae-Won Kim*
nique were quite porous and weak, raising concerns about the
longevity and stability of the layer.11 In this respect, relatively
thick coating layers (410 mm) retaining high chemical and ther-
mal stability are necessary for long-term biological applications.
Moreover, a simple and low-cost fabrication method is required
for real applications. A previous study proposed that a thick HA
coating might be fabricated by the HA powder slurry by tailor-
ing the slurry conditions and thermal treatment history.11 The
most important considerations were the properties of the coat-
ing slurry containing polymeric additives and the thermal treat-
ment conditions, such as the burnout of polymers and the
densification of the coating layer.
Many attempts to fabricate HA composite with ZrO2 re-
vealed that an extensive reaction occurred between them.12 The
reaction products (a- or b-tri-calcium phosphate (TCP) and Ca-
ZrO3) were found to be undesirable for mechanical properties as
well as for biological stability.13 Therefore, in an HA/ZrO2
coating system, the degradation of the HA coating layer re-
mains a primary concern. Recently, fluorine (F)-substituted HA
composites with ZrO2 were successfully fabricated without re-
action problems.14 The excellent chemical stability of fluorapa-
tite [FA, Ca10(PO4)6F2] with respect to ZrO2 has effectively
prevented the decomposition reaction. In practice, FA has at-
tracted attention because of its slow bio-resorption as well as its
comparable bone fixation and ingrowth compared with HA.15,16
In this study, an FA layer was coated between the HA layer
and ZrO2 substrate to improve the integrity and stability of the
HA coating on ZrO2. The fabrication method of the HA/FA
double-layer coating on ZrO2 is addressed in relation to slurry
properties as well as coating and thermal treatment conditions.
The mechanical and biological performances of the coating sys-
tem are briefly assessed in terms of bonding strength and in vitro
cell proliferation behavior, respectively.
II. Experimental Procedures
(1) Preparation of Coating Slurry
A commercially available HA powder (Alfa Aesar, Ward Hill,
MA) was used to make an HA slurry for coating. The HA
powder was mixed with a controlled amount of tri-ethyl phos-
phate (TEP; PO(C2H5O)3, Aldrich, St. Louis, MO) in 100 mL
ethanol, and the mixture was stirred vigorously with a magnetic
spin bar. After stirring for 24 h, polyvinylbutyral (PVB, Aldrich)
was added slowly to the slurry and the mixture was stirred again
for 48 h. TEP and PVB were used as a dispersant and a binder,
respectively, and the amounts were changed from 0.5 to 10 wt%
with respect to HA. The amount of HA powder was varied from
10 to 50 wt% relative to the ethanol.
The FA powder was fabricated using TCP (Ca3(PO4)2, Merck
Co., Darmstadt, Germany) and calcium fluoride (CaF2, Ald-
rich) by reacting them at 10001C for 3 h, as follows:
3Ca3 ðPO4 Þ2 þ CaF2 ¼ Ca10 ðPO4 Þ6 F2 (1)
August 2006 Hydroxyapatite/Fluorapatite Double Layer Coating on Zirconia
Procedures for preparing the FA slurry were the same as the
case of HA as described above.
(2) HA/FA Coating on ZrO2
The substrate (ZrO2) was prepared as a disk shape after molding
and sintering a high-purity ZrO2 powder (with 3 mol% Y2O3,
Cerac Inc., Milwaukee, WI) at 14001C for 3 h. To fabricate an
HA/FA double coating layer on ZrO2, the FA layer was coated
first, and then HA was coated on the FA layer. After cleaning
and drying, the ZrO2 disk was immersed into the FA slurry at a
dipping/pulling speed of 0.5 mm/min, and then dried at room
temperature and subsequently at 801C for 3 h. The dried sample
was heat treated at 8001C for 3 h and further at various tem-
peratures (10001–13001C at 1001C intervals) for 3 h to prepare
an FA-intermediate layer onto ZrO2. Binder burnout tempera-
ture was selected as 8001C based on thermal analysis data, and a
slow heating rate (21C/min) was adopted up to 8001C to avoid
an abrupt burning out of the polymeric precursors. A relatively
thick and uniform FA layer was obtained by repeating this
coating process. The FA-coated sample was dipped into the HA
slurry and dried, heated under the same conditions, and then the
procedure was repeated four times. The coating cycle of each
layer was changed to obtain an optimum coating structure. For
the purpose of comparison, an HA-coated ZrO2 disk without an
FA layer was also prepared.
(3) Characterization and Bonding Strength Test
The rheological properties of the HA slurries were investigated
at a constant temperature of 251C using a rheometer (Rheostress
600; Thermo Haake Co., Karlsruhe, Germany). The viscosity of
the slurries was determined at various shear rates (from 0.1 to 10
s
1) for 60 s. To avoid solvent evaporation, silicone oil was em-
bedded around the slurry after molding, and the viscosity was
measured as a function of time at a fixed shear rate. The effects
of the additions of dispersant (TEP; 0.5–10 wt%) and binder
(PVB; 0.5–10 wt%) as well as the loadings of HA powder (10–50
wt%) on the viscosity were investigated.
To observe the sintering behaviors of the HA and FA pow-
ders, each powder was compacted separately in a metal die at a
low pressure (B0.5 MPa), further pressed isostatically at 150
MPa, and then sintered at temperatures of 9001–14001C for 1 h
in air. The density of the sintered body was measured by the
Archimedes method. The phase and the morphology of the
coating layer were evaluated with an X-ray diffractometer
(XRD) and a scanning electron microscope (SEM), respectively.
The adhesion strength of the coating layer was tested with an
adhesion testing apparatus (Sebastian V, Quad Group, Spo-
kane, WA).17 A stud, pre-coated by the manufacturer using an
epoxy of a proprietary composition, was adhered to the coating
layer by curing the epoxy at 1501C for 1 h. The 2.69 mm diam-
eter stud was pulled at a loading rate of B2 mm/min until the
failure of the coating layer, and the bonding strength was de-
termined from the maximum load recorded.
(4) Cell Proliferation
For an assessment of the proliferation behavior of cells on the
coating system, an MG63 cell line was pre-incubated in flasks
containing Dulbecco’s modified Eagle’s medium (DMEM, Life
Technologies Inc., Rockville, MD) supplemented with 10% fetal
bovine serum (FBS, Life Technologies Inc.). Then, the cells were
plated at densities of 5  103 and 3  104 cells/mL for prolifer-
ation rate calculation and cell morphology observation, respec-
tively, on 24-well plates containing fabricated coating specimens
and a cell culture dish (Thermanox; NUNC, Rochester, NY) as
a control. After culturing for 3 days in an incubator humidified
with 5% CO2/95% air at 371C, the cells were detached with a
trypsin-EDTA solution, and the living cells were counted using a
hemocytometer (Superior Co., Lauda, Germany). Each set of
tests was performed in triplicate, and the data were normalized
by taking the surface area into consideration.
2467
To observe the morphology of the growing cells, SEM was
performed after fixing the cells with glutaraldehyde (2.5%), de-
hydrating with a graded series of ethanols (70%, 90%, and
100%), and critical point drying in CO2.
III. Results
(1) Rheological Properties of Slurries for Coating
The viscosities of the HA slurry, with and without the addition
of a TEP dispersant, with respect to the shear rate are shown
in Fig. 1. The loading of HA powder was fixed at 30 wt% rel-
ative to the ethanol. In pure HA slurry, the viscosity exhibited a
typical shear-thinning behavior. A small addition of TEP (0.5
wt%) decreased the viscosity of the HA slurry sharply in the 0–
10 s
1 range, illustrating the effectiveness of TEP as a disper-
sant. On increasing the TEP content up to 5 wt%, the viscosity
continued to decrease at all shear rates tested. However, there
was little, if any, change in viscosity with further additions, sug-
gesting the appropriate amount of TEP addition to be B5 wt%.
The effect of PVB binder addition on the HA slurry is shown
in Fig. 2. The loading of HA powder and the TEP was fixed to
30 and 5 wt%, respectively. When the shear rate was low (0.1
s
1), small additions of PVB (up to 5 wt%) were effective in
decreasing the viscosity, while further additions gradually in-
creased the value. When the shear rate was 1 s
1, a similar trend
was observed, even though the degree of decrease in viscosity
was not so large. At a high shear rate (10 s
1), the viscosity re-
mained constant up to 5 wt% and increased at higher additions
of PVB. This result clearly indicates that only small amounts of
PVB are effective in reducing the viscosity of the slurry.
Figure 3 shows the viscosity of the HA slurry as a function of
the HA powder loading into ethanol at a constant shear rate of 1
s
1. The added amounts of TEP and/or PVB were 5 wt% rel-
ative to the HA powders. Regardless of the TEP and/or PVB
addition, the viscosity of HA slurries increased with increasing
HA loadings. In the pure HA slurry, the viscosity increased
sharply at HA loadings of B30–50 wt%. However, the addition
of 5 wt% TEP suppressed any significant increase in the viscos-
ity even at HA powder loadings up to 50 wt%. In the slurry
containing both TEP and PVB of 5 wt%, the increase in vis-
cosity was much slower than that in pure HA slurry, but slightly
higher than that in slurry containing TEP only.
(2) Consolidification and Stability of Coating Layer
To determine the thermal treatment temperatures, the sintering
behaviors of HA and FA were monitored first, as shown in
60 100
0% TEP
0.5% TEP 10
15 2% TEP
5% TEP
10% TEP
Viscosity [Pa.s]
1
10
0.1 1 10
5
0
0 2 4 6 8 10
Shear Rate [s –1]
Fig. 1. Viscosity change of hydroxyapatite (HA) slurry with and with-
out various additions of triethyl phosphate, presented with respect to
shear rates from 0.1 to 10 s
1. Loading of HA powder was fixed at 30
wt% relative to ethanol.
2468 Journal of the American Ceramic Society—Kim et al. Vol. 89, No. 8

100
60
Shear rate (1/s)
0.1
50 1
10 90

Relative Density [%]

40
Viscosity [Pa.s]

80
30

20 HA
70 FA

10

60
0

0 5 10 900 1000 1100 1200 1300 1400

PVB Concentration [wt%]
Fig. 2. Viscosity change of hydroxyapatite (HA) slurry with respect to
the added concentration of polyvinylbutyral: calculated at different
shear rates of 0.1, 1, and 10 s
1. Loading of HA powder was fixed at
30 wt% relative to ethanol.
Sintering Temperature [°C]
Fig. 4. Sintering behaviors of hydroxyapatite and fluorapatite powders:
Sintered at various temperatures for 1 h in air.

HA was effectively suppressed or reduced, as shown in Figs.

Fig. 4. For both HA and FA, the densities increased with sin-
tering temperature, reaching nearly theoretical values above
12001C. These data indicated that the final thermal treatment
temperature of the HA and FA slurry coating should be between
12001 and 13001C in order to obtain a dense and strong coating
layer.
When HA was coated on ZrO2 without the FA-intermediate
layer, high-temperature heating resulted in a severe decomposi-
tion of HA, as manifested by Figs. 5(A)–(D). There was little
decomposition of HA after heat treating at 10001C (Fig. 5(A)).
However, as the temperature increased to 11001C, a considerable
degree of HA decomposition occurred, with a large amount of b-
TCP being produced (Fig. 5(B)). Further increases in tempera-
ture up to 12001 and 13001C resulted in the formation of CaZrO3
and/or a-TCP, as shown in Figs. 5(C) and (D), respectively.
In contrast to the active decomposition of HA, when the FA
layer was present between HA and ZrO2, the decomposition of
6(A)–(D). No decomposition was detected by the XRD analysis
after heat treatment at 10001 and 11001C, as shown in Figs. 6(A)
and (B), respectively. Only a trace of b-TCP at 12001C
(Fig. 6(C)), and somewhat more at 13001C (Fig. 6(D)), was ob-
served. However, when compared with the direct coating of HA
on ZrO2, the degree of decomposition above 12001C is negli-
gible.
(3) Structure and Bonding Strength of the Coating Layer
Figures 7(A)–(C) show the microstructures of FA coating layers
on ZrO2 at different heat-treatment temperatures and coating
cycles. After heat treating at 11001C (Fig. 7(A)), the FA particles
were not fully consolidified apparently due to the insufficient
sintering temperature, as was noted in Fig. 4. However, after

25 w/o TEP/PVB
(A)
5% TEP
5% TEP + 5% PVB
Intensity [Arb. Unit.]

20
Viscosity [Pa.s]

(B)
15

(C)
10

5 (D)

0
29 31 33
0 10 20 30 40 50 2θ [°]

Loading of HA Powder [wt%]
Fig. 3. Viscosity change of hydroxyapatite (HA) slurry with respect to
the loading of HA powder: Compared with the slurry containing tri-ethyl
phosphate and/or poly vinyl butyral. Calculated at a shear rate of 1 s
1.
Fig. 5. X-ray diffraction patterns of hydroxyapatite (HA) single layer
coated on ZrO2 without the fluorapatite-intermediate layer after final
heat treatment at various temperatures: (A) 10001C, (B) 11001C,
(C) 12001C, and (D) 13001C: (J) HA, () TCP, (&) ZrO2, and ( )
CaZrO3.
August 2006 Hydroxyapatite/Fluorapatite Double Layer Coating on Zirconia 2469

(A)
Intensity [Arb. Unit.]

(B)

(C)

(D )

29 31 33
2θ [°]
Fig. 6. X-ray diffraction patterns of an HA/FA double layer coated on
ZrO2 after final heat treatment at various temperatures: (A) 10001C, (B)
11001C, (C) 12001C, and (D) 13001C: (J) HA, () TCP, and (&) ZrO2.

heating at 12001C (Fig. 7(B)), the FA particles were consolidat-

ed to form a network while parts of the ZrO2 substrate surface
were still uncovered with the coating layer. By repeating the
coating process twice, the substrate was fully covered with the
FA coating layer (Fig. 7(C)).
Figures 8(A)–(C) show the morphologies of HA coating onto
the FA pre-coated ZrO2. After single coating and heat treatment
at 12001C, the HA layer covered the substrate only partially, as
shown in Fig. 8(A). By repeating the coating cycle once more,
the HA layer uniformly covered the whole surface of the FA-
coated ZrO2 (Fig. 8(B)). The cross-sectional morphology of the
HA/FA double layer on ZrO2 is shown in Fig. 8(C). The thick-
nesses of the HA and FA layers were B20–30 and 5–10 mm,
respectively, confirming the formation of a relatively thick coat-
ing layer, which might be useful for long-term applications.
Moreover, the two layers were bonded quite tightly and no
cracks or delamination were observed between the layers.
The bonding strength of the HA/FA double layer on ZrO2
with respect to the final heat-treatment temperature is shown in
Fig. 9. The strength was quite low (o10 MPa) at temperatures
up to 11501C, but increased sharply at 12001C (22 MPa). The Fig. 7. Scanning electron micrograph surface morphologies of fluorap-
strength remained high up to B13001C, but decreased slightly at atite coating layers on ZrO2: the coating cycle and final heat-treatment
14001C. temperature were (A) once and 11001C, (B) once and 12001C, and (C)
twice and 12001C, respectively.
(4) Cellular Response
As a preliminary biological test of the HA/FA coating on ZrO2, In the process, understanding and controlling of the slurry char-
the in vitro cell proliferation behavior was evaluated using os- acteristics, as well as of the thermal treatment behavior, were
teoblast-like MG63 cells. The morphology of the growing cells found to be essential. Generally, ceramic powders in the slurry
on the HA/FA-coated ZrO2 after culturing for 3 days is shown are known to be agglomerated due to van der Waals attraction
in Figs. 10(A) and (B). The MG63 cells were flattened and in- between particles, and the use of a polymeric dispersant is rec-
timately adhered to the rough coating surface. The membranes ommended.18–22 The addition of a small amount (5 wt%) of
spread in an elongated shape. Some spherical cells appeared to TEP as a dispersing agent considerably decreased the viscosity
undergo a cell division (arrowed). of the slurry (Fig. 1). This decrease in viscosity implies a less
The number of cells proliferated on the coating system was agglomerated state of the HA powder, and this was possible by
about the same as those on the culture dish or HA bulk ceramic, the adsorption and electrostatic stabilization mechanism in-
as shown in Fig. 11. Such a comparable proliferation rate indi- duced by the TEP. The TEP additive, due to its polarity, can
cates the favorable cell viability of the coating layers. easily adsorb onto the HA, which surely have a bunch of ad-
sorption sites, consisting of ionic groups.23,24 When adsorbed
onto the surface of HA powders, the TEP is expected to exert
IV. Discussion
electrostatic stabilization rather than steric hindrance as the
In this study, we addressed the processing conditions, mechan- TEP has a degree of polarity and short ethyl groups at the end.22
ical properties, and biological performance of a relatively thick Along with the dispersant, the addition of a PVB binder is
HA coating layer onto ZrO2 by using a powder slurry method. needed for stable thermal treatment because during the early
2470 Journal of the American Ceramic Society—Kim et al. Vol. 89, No. 8

30

25

Bonding Strength [MPa]

20

15

10

1000 1100 1200 1300 1400

Sintering Temperature [°C]

Fig. 9. Bonding strengths of an HA/FA double layer coated on ZrO2

presented as a function of final heat-treatment temperature.

lecular); the PVB would adsorb mainly by an electrostatic en-

ergy, while an ionic interaction exerts to a high degree in the
TEP adsorption. Hence, the TEP adsorption would be more ef-
ficient and strong as compared with PVB.

Fig. 8. Scanning electron micrograph morphologies of hydroxyapatite

coating layers on fluorapatite pre-coated ZrO2 after a final heat treat-
ment at 12001C: (A) surface morphology with a single coating, (B) sur-
face morphology with cycled coating twice, and (C) cross-sectional
morphology with cycled coating twice.

stage of heat treatment, the PVB sustains the initial HA coating

shape and structure, and also prevents the powder from detach-
ing off the substrate. The appropriate amount of PVB appeared
to be less than 5 wt%, a range within which the viscosity did not
increase significantly for all shear rates tested (Fig. 2). Unlike
TEP, the PVB is non-ionic and consists of long polymeric
chains; hence, higher additions of PVB should increase the vis-
cosity due to the decrease in adsorption efficiency to the HA
particles, i.e., the excess PVB chains would interact with each
other and be deleterious to the slurry property. For this reason,
the PVB is well known as a binder, normally increasing the vis-
cosity of ceramic slurries.25,26 However, at small additions, it Fig. 10. Scanning electron micrograph surface morphologies of MG63
could also reduce the viscosity, by adsorbing onto the HA sur- cells proliferated on an HA/FA double-coated layer on ZrO2 after cul-
face efficiently and rendering the HA particles to be in steric ture for 3 days: (A) at low magnification and (B) at high magnification.
repulsion. The adsorption of the PVB (non-ionic long chain) on Micrographs show the cross-sectional view at a low tilting angle. The
HA would be different from that of TEP (ionic and short mo- arrow indicates the cell undergoing division.
August 2006 Hydroxyapatite/Fluorapatite Double Layer Coating on Zirconia
culture for 3 days
8 plating cell density = 3 x 104/ml
Cell Numbers [cm–2, x 104]
6 behaved similarly with the sintering data (see Figs. 9 and 4),
4 of the FA-intermediate layer to TCP and CaZrO3. At this point,
2 strength. The epoxy might be infiltrated into the coating layer
0 failure surface revealed that failures occurred both within the
Control HA bulk HA /FA coating
Fig. 11. MG63 cell numbers proliferated on an HA/FA double-coated
layer on ZrO2 after culture for 3 days. A plastic culture dish was used as
a control. HA bulk ceramic was also tested for comparison, after full
densification and polishing down to 1 mm.
The solid loading is also important to the quality of the final
product as a higher loading generally leads to a higher packing
density and minimizes the shrinkage of the coating layer at
elevated temperatures. However, this is true only when the sus-
pension is well dispersed without agglomerates or sedimenta-
tion.21 Even at high HA loadings (40–50%), the viscosity of the
TEP-containing slurry did not increase much, in contrast to the
pure HA slurry, in which the viscosity increased drastically at
loadings above 30% (Fig. 3). Based on the rheological property
observations, a well-dispersed HA suspension containing poly-
meric precursors was prepared for further coating processes.
With viscosity-controlled slurries, the next step to obtain a
coated specimen consisted of binder burnout and final consol-
idation of the coating layer. The binder burnout was performed
at a relatively slow heating rate (21C/min) below 8001C so as not
to destroy the coating structure due to an abrupt polymer evap-
oration. After this, a relatively higher heating rate was used to
aviod HA dehydration or decomposition above 8001C. From
the powder sintering data, it was confirmed that a temperature
of at least 12001C was necessary to consolidate the HA and FA
particles (Fig. 4). In fact, the data in Fig. 4 are not directly re-
lated to the coating density as the full densification of the coat-
ing layer would be somewhat limited by the low initial packing
density and the underlying ZrO2 substrate. However, the data
would give important information at least on the sintering phe-
nomenon occurring locally. Practically, when heat treated above
12001C, a considerable necking and sintering of the powders
were observed to occur on the porous coating layer.
When there was no FA layer between the HA layer and
substrate, heat treatment above 12001C caused a severe reaction
between HA and ZrO2 to form TCP (a- and/or b -) and CaZrO3
(Fig. 5). The TCP formed at the interface is undesirable due to
its possible bioresorption prior to HA dissolution and subse-
quent detachment of the coating layer from the substrate. To
overcome this problem, an FA thin layer was inserted into the
HA/ZrO2 interface and was observed to suppress the HA de-
composition effectively (Fig. 6). The higher chemical stability of
FA with respect to ZrO2 was responsible for the prevention of a
decomposition reaction.
To form the double-layered HA/FA coating uniformly, each
coating step (for both HA and FA) was performed repeatedly.
The single coating of each layer resulted in an uneven coverage,
while the whole surface was covered uniformly after repeated
coating (Figs. 7 and 8). Within the HA/FA double coating layer,
there were no cracks or delamination, presumably due not only
to the relatively micro-porous structure that alleviates thermal
mismatch but also due to the composition and crystallographic
2471
similarity between HA and FA (Fig. 8(C)). In practice, the
measured bonding strength of the HA/FA coating to ZrO2
substrate was as high as B22–25 MPa after heat treatment
above 12001C. Such values are comparable to those obtained by
plasma-sprayed coating, confirming the possible application
to load-bearing parts.7 As expected, the bonding strength data
illustrating that the consolidification of the coating layer is
crucial to obtain a strong coating layer. The slight decrease in
strength at 14001C was deemed to be due to the degradation
a question might be raised on the ingress of the epoxy through
the porous coating layer, and its influence on the bonding
to some extent, but it should not ingress so deep into the coating
layer as to affect the final strength. Such problems are usually
raised in thin and porous coatings. The SEM observation of the
coating layer and at the coating/substrate interface, and there
was no epoxy residue detected on the failure surface.27 Such a
fact confirmed that the bonding strength was dominated by both
the interface interlocking and the coating layer integrity. Hence,
if there were epoxy ingress deep into the coating layer or even to
the interface, and if the epoxy would affect the coating integrity
or interface strength, the failure would occur mostly at the in-
terface or within the coating accompanied by an epoxy residue.
Therefore, there should be little, if any, effect of epoxy on the
final coating strength.
The biological properties of the HA/FA coating system were
investigated by the proliferation behavior of MG63 osteoblast-
like cells. The cells grew, spread, and proliferated actively on the
specimen, confirming that the HA/FA-coated ZrO2 system has
no cytotoxic effect on the osteoblastic cells and retains a rea-
sonable cytocompatibility. Moreover, similar proliferation of
coating sample to the Thermanox control confirmed its reason-
able cell viability.14,28–30 In many reports, the physical status of
the surface, such as roughness, morphology, and porosity, is
known to affect the cell attachment, proliferation, and differen-
tiation.31,32 However, in this study, the relatively rough HA
coating surface appeared to have no significant influence on the
cell response at least at a proliferation level, as confirmed by the
similar cell proliferation rate with the smooth HA bulk material.
Along with the proliferation, the cells should further undergo a
differentiation step for bone formation. Previously, the authors
had observed that in a porous scaffold system, the alkaline
phosphate (ALP) activity was expressed to a higher degree
on the HA/FA-coated ZrO2 when compared with a pure
ZrO2 scaffold, and suggested its improved biocompatibility.
Although there would be morphological differences between
the scaffold and disk on which the cells grow, the results would
be applicable to this disk system.
V. Conclusions
A double layer, composed of outer HA and inner FA layers, was
coated on ZrO2 by a powder slurry method. Slurry and coating
parameters were controlled to obtain an optimal coating con-
dition. Appropriate additions of TEP dispersant and PVB bind-
er effectively reduced the viscosity of the slurry containing the
HA powder up to 50 wt%. Through repeated coatings and heat
treatment above 12001C, a uniform HA/FA double layer was
deposited on ZrO2. The FA-intermediate layer completely pre-
vented the reaction between HA and ZrO2 up to 13001C. The
bonding strength of the double layer was approximately 25 MPa
after heat treatment above 12001C. The MG63 cells cultured on
the coating system spread, grew, and proliferated favorably,
demonstrating a comparable cell viability of the specimen com-
pared with HA bulk ceramic.
References
1
M. Jarcho, ‘‘Calcium Phosphate Ceramics as Hard Tissue Prosthetics,’’ Clin.
Orthop., 157, 259–78 (1981).
2472 Journal of the American Ceramic Society—Kim et al.
2 18
L. L. Hench and E. C. Ethridge, Biomaterials: An Interfacial Approach. Aca-
demic Press, New York, 1982.
3
W. Suchanek and M. Yoshimura, ‘‘Processing and Properties of Hydroxyap-
atite-Based Biomaterials for Use as Hard Tissue Replacement Implant,’’ J. Mater.
Res., 13, 94–117 (1998).
4
Y. M. Kong, S. Kim, and H. E. Kim, ‘‘Reinforcement of Hydroxyapatite Bio-
ceramics by Addition of ZrO2 Coated with Al2O3,’’ J. Am. Ceram. Soc., 82, 2963–8
(1999).
5 21
J. Li, H. Liao, and L. Hermansson, ‘‘Sintering of Partially-Stabilized Zirconia
and Partially-Stabilized Zirconia–Hydroxyapatite Composites by Hot Isostatic
Pressing and Pressureless Sintering,’’ Biomaterials, 17, 1787–90 (1996).
6 22
R. G. T. Geesink, K. de Groot, and C. P. A. T. Klein, ‘‘Bonding of Bone to
Apatite-Coated Implants,’’ J. Bone Joint Surg., 70B, 17–22 (1988).
7 23
K. de Groot, R. G. T. Geesink, C. P. A. T. Klein, and P. Serekian, ‘‘Plasma-
Sprayed Coatings of Hydroxyapatite,’’ J. Biomed. Mater. Res., 21, 1375–81 (1987).
8
D. M. Liu, H. M. Chou, and J. D. Wu, ‘‘Plasma-Sprayed Hydroxyapatite
Coatings: Effect of Different Calcium Phosphate Ceramics,’’ J. Mater. Sci.: Ma-
ter. Med., 5, 147–53 (1994).
9
W. Weng and J. L. Baptista, ‘‘Sol–Gel Derived Porous Hydroxyapatite Coat-
ings,’’ J. Mater. Sci.: Mater. Med., 9, 159–63 (1998).
10
T. Brendel, A. Engel, and C. Russel, ‘‘Hydroxyapatite Coatings by a Poly-
meric Route,’’ J. Mater. Sci.: Mater. Med., 3, 175–9 (1992).
11
J. Lee and H. Aoki, ‘‘Hydroxyapatite Coating on Ti Plate by a Dipping
Method,’’ Biomed. Mater. Eng., 5, 49–58 (1995).
12
J. M. Wu and T. S. Yeh, ‘‘Sintering of Hydroxyapatite–Zirconia Composite
Materials,’’ J. Mater. Sci., 23, 3771–7 (1988).
13
C. P. A. T. Klein, A. A. Driessen, A. de Groot, and A. Van den Hooff, ‘‘Bi-
odegradation Behavior of Various Calcium Phosphate Materials in Bone Tissue,’’
J. Biomed. Mater. Res., 17, 769–76 (1983).
14
H.-W. Kim, Y.-M. Kong, Y.-H. Koh, H.-E. Kim, H.-M. Kim, and J. Ko,
‘‘Pressureless Sintering, Mechanical and Biological Properties of Fluor-Hydroxy-
apatite Composites with Zirconia,’’ J. Am. Ceram. Soc., 86 [12] 2019–26
(2004).
15
E. C. Moreno, M. Kresak, and R. T. Zahradnik, ‘‘Fluoridated Hydroxyap-
atite Solubility and Caries Formation,’’ Nature, 247, 64–5 (1974).
16
G. S. Ingram and P. F. Nash, ‘‘Mechanism for the Anticaries Action of Flu-
oride,’’ Caries Res., 14, 298–303 (1980).
17
J. M. Choi, Y. M. Kong, S. Kim, H. E. Kim, C. S. Hwang, and I. S. Lee,
‘‘Formation and Characterization of Hydroxyapatite Coating Layer on
Ti-Based Metal Implant by Electron Beam Deposition,’’ J. Mater. Res., 14,
2980–5 (1999).
Vol. 89, No. 8
T. F. Tadros, ‘‘Correlation of Viscoelastic Properties of Stable and Floccu-
lated Suspension with their Interparticle Interactions,’’ Adv. Colloid. Interface Sci.,
68, 97–200 (1996).
19
S. Gautier, E. Champion, D. Bernache-Assollant, and T. Chartier, ‘‘Rheo-
logical Characteristics of Alumina Platelet-Hydroxyapatite Composite Suspen-
sions,’’ J. Eur. Ceram. Soc., 19, 469–77 (1999).
20
D. M. Liu, ‘‘Preparation and Characterization of Porous Hydroxyapatite
Bioceramics Via a Slip-Casting Route,’’ Ceram. Int., 24, 441–6 (1998).
P. Misubbana and S. G. Malghan, ‘‘Shear Yield Stress of Flocculated Alu-
mina-Zirconia Mixed Suspension: Effect of Solid Loading, Composition and Par-
ticle Size Distribution,’’ Ceram. Eng. Sci., 53, 3073–9 (1999).
D. H. Napper, Polymeric Stabilization of Colloidal Dispersions, pp. 4–13. Ac-
ademic Press, London, 1983.
J. Ma, Z. Xie, H. Miao, L. Zhou, Y. Huang, and Y. Cheng, ‘‘Elimination of
Surface Spallation of Alumina Green Bodies Prepared by Arylamid-Based Gel
Casting Via Poly(vinlypyrrolidone),’’ J. Am. Ceram. Soc., 86 [2] 266–72 (2003).
24
Q. Q. Hoang, F. Sicheri, A. J. Howard, and D. S. C. Yang, ‘‘Bone Recognition
Mechanism of Porcine Osteocalcin From Crystal Structure,’’ Nature, 425, 977–80
(2003).
25
H. A. Goldberg, K. J. Warner, and G. K. Hunter, ‘‘Binding of Bone Sialo-
protein, Osteopontin and Synthetic Polypeptides to Hydroxyapatite,’’ Connect.
Tissue Res., 42 [1] 25–37 (2001).
26
Z. Xie, C. Ma, Y. Huang, and J. Xiang, ‘‘Gel Tape Casting Ceramic Sheets,’’
Am. Ceram. Soc. Bull., 81 [10] 33–7 (2002).
27
H.-W. Kim, S.-Y. Lee, C.-J. Bae, Y.-J. Noh, H.-E. Kim, H.-M. Kim, and J.
Ko, ‘‘Porous ZrO2 Bone Scaffold Coated with Hydroxyapatite with Fluorapatite
Intermediate Layer,’’ Biomaterials, 24, 3277–84 (2003).
28
H.-W. Kim, Y.-J. Noh, Y.-H. Koh, H.-E. Kim, and H.-M. Kim, ‘‘Effect of
CaF2 on Densification and Properties of Hydroxyapatite–Zirconia Composites for
Biomedical Applications,’’ Biomaterials, 23, 4113–21 (2002).
29
H.-W. Kim, Y.-H. Koh, S.-B. Seo, and H.-E. Kim, ‘‘Properties of Fluoridated
Hdroxyapatite–Alumina Biological Composites with Addition of CaF2,’’ Mater.
Sci. Eng: C, 23, 515–21 (2003).
30
H.-W. Kim, Y.-J. Noh, Y.-H. Koh, and H.-E. Kim, ‘‘Enhanced Performance
of Fluorine Substituted Hydroxyapatite Composites for Hard Tissue Engineer-
ing,’’ J. Mater. Sci.: Mater. Med., 14, 899–904 (2003).
31
K. Anselme, ‘‘Osteoblast Adhesion on Biomaterials: Review,’’ Biomaterials,
21, 667–81 (2000).
32
A. Okumura, M. Goto, T. Goto, M. Yoshinara, S. Masuko, T. Katsuki, and
T. Tanaka, ‘‘Substrate Affects the Initial Attachment and Subsequent Behavior of
Human Osteoblastic Cells (Saos-2),’’ Biomaterials, 22, 2263–71 (2001). &