Biomaterials 23 (2002) 3103–3111

The role of surface functional groups in calcium phosphate nucleation

on titanium foil: a self-assembled monolayer technique

Qing Liu

a

, Jiang Ding

b

, Francis K. Mante

c

, Stephanie L. Wunder

b

, George R. Baran

a,

*

a

Center for Bioengineering and Biomaterials, College of Engineering, Temple University, 1947 North 12th Street, Philadelphia, PA 19122, USA

b

Department of Chemistry, Temple University, Philadelphia, USA

c

School of Dental Medicine, University of Pennsylvania, Philadelphia, USA

Received 1 February 2001; received in revised form 30 November 2001; accepted 4 February 2002

Abstract

Surface functional groups play important roles in nucleating calcium phosphate deposition on surgical titanium implants. In this

study, various functional groups were introduced onto the surface of commercially pure titanium foils using a self-assembled
monolayer (SAM) technique. An organic silane, 7-oct-1-enyltrichlorosilane (OETS) was used and –OH, –PO

4

H

2

, –COOH groups

were derived from its unsaturated double bond. Ti foils were first oxidized in concentrated H

2

SO

4

/H

2

O

2

. ESCA and contact angle

measurements were used to characterize the SAM surfaces and confirm the presence of various functional groups. A fast calcium
phosphate deposition experiment was carried out by mixing Ca

2+

- and (PO

4

)

3

-containing solutions in the presence of the surface-

modified Ti samples at pH 7.4 at room temperature in order to verify the nucleating abilities of these functional groups. SEM,
Raman spectroscopy, XRD and ATR-FTIR results showed that poorly crystallized hydroxyapatite (HA) can be deposited on the
SAM surfaces with –PO

4

H

2

and –COOH functional groups, but not onto the SAM with –CHQCH

2

and –OH. –PO

4

H

2

exhibited a

stronger nucleating ability than that of –COOH. The oxidized Ti sample also showed some calcium phosphate deposition but to a
lesser extent as compared to SAM surfaces with –PO

4

H

2

and –COOH. The pre-deposited HA can rapidly induce biomimetic apatite

layer formation after immersion in 1.5 SBF for 18 h regardless of the amount of pre-deposited HA. The results suggested that
the pre-deposition of HA onto these functionalized SAM surfaces might be an effective and fast way to prepare biomimetic apatite
coatings on surgical implants. r 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Titanium; Surface modification; Hydroxyapatite; Self-assembled monolayer; Biomimetic; Coating

1. Introduction

Titanium (Ti) and its alloys have good biocompat-

ibility and mechanical properties; as a result, they are
frequently used as surgical implants in load bearing
situations, such as hip prostheses and dental implants.
However, Ti and its alloys do not bond to bone in the
early (

o6 months) post-implantation stage. Various

methods have been employed to introduce hydroxyapa-
tite (HA) or calcium phosphate (Ca/P) coatings onto
metal implant surfaces to improve and accelerate their
integration with bony tissue. Currently, most commer-
cially available HA coatings are applied onto implants
by means of plasma spraying, but these coatings have
some limitations. The process is carried out at high
temperatures (>10001C) in a line-of-sight method and

therefore cannot be applied to complex shapes and
porous implants, nor to implants where incorporation of
biologic molecules into the coating is required [1].

Biomineralization is a natural process in human

beings and animals resulting in the formation of bones
and teeth. Inspired by this process, researchers have
been interested in a biomimetic method to produce
calcium phosphate coatings. The biomimetic coating
process is defined as a method whereby a biologically
active apatite layer is formed on a substrate after
immersion in an artificially prepared supersaturated
calcium and phosphate solution such as simulated body
fluid (SBF), which is compositionally similar to human
body fluid. SBF has been frequently used to apply
biomimetic apatite coatings on various biomaterials.

A key step in the biomimetic coating process is the

initial nucleation of Ca/P on the implant surface, and
various methods may be used to achieve this goal. Li
et al. have found that sol–gel prepared titania may

*Corresponding author. Fax: [phone].

E-mail address:

[e-mail] (G.R. Baran).

0142-9612/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 1 4 2 - 9 6 1 2 ( 0 2 ) 0 0 0 5 0 - 9

nucleate Ca/P because of the abundant acid hydroxyl
groups on its surface [2]. Ohtsuki et al. also showed that
basic Ti–OH groups on the surface of Ti treated by
H

2

O

2

/TaCl

5

or H

2

O

2

/SnCl

2

had the ability to induce

formation of calcium apatite [3]. Recently, some studies
have shown that Ti treated with 5 m NaOH at 601C
induced apatite nucleation in SBF [4,5]. After NaOH
treatment, a sodium titanate gel formed on the Ti
surface. Sodium ions can be released from the gel when
it is soaked in SBF, and Ti–OH groups are formed. An
increase in pH accompanies the release of sodium ions.
Wen et al. also found that titanium treated in 0.2 m
NaOH at 1401C had the ability to nucleate apatite
precipitation in a supersaturated calcium phosphate
solution apatite [6]. According to their study, a porous
TiO

2

layer formed on the surface of titanium after

NaOH treatment. They believe that this layer is
negatively charged and is responsible for the nucleation
of apatite [6].

Various functional groups can be introduced onto the

surface of Ti and its alloys to render the surface capable
of nucleating apatite. Campbell et al. introduced –SO

3

H

to the surface of Ti to initiate calcium phosphate
deposition [7,8]. Mao et al. introduced carboxyl and
hydroxyl groups on the surface of Ti using short-chain
vinyltriethoxysilane, and both functional groups showed
ability to nucleate apatite [9]. These results suggest that
surface functional groups could play a critical role in
inducing Ca/P nucleation.

The role of functional groups in inducing nucleation

was confirmed in other studies using substrate materials
other than titanium and its alloys. Tanahashi et al. used
alkanethiols to form a self-assembled monolayer (SAM)
on a gold substrate and then studied the ability of
various surface functional groups to nucleate Ca/P after
immersion in SBF [10]. They found that the most potent
nucleating group was –PO

4

H

2

, followed by –COOH.

The –CONH

2

, –OH and –NH

2

groups had weaker

nucleating ability. Sato et al. used a Langmuir–Blodgett
(LB) technique to introduce –COOH and –NH

2

groups

on the surface of glass slides or CaF

2

crystals and found

that –COOH groups could induce apatite nucleation
while –NH

2

could not [11]. While certain functional

groups are effective in nucleating Ca/P on Ti surfaces,
the subsequent deposition of significant amounts of
Ca/P requires incubation in SBF for several days or
longer

[9–12].

That

is,

the induction

period of

apatite nucleation is quite long for these functionalized
surfaces. Therefore, there is an interest in decreasing the
induction period or speeding up the nucleation process
of biomimetic apatite.

HA mineral can both provide potent nucleation sites

and serve as promoter for biomimetic apatite nucleation
because of the structural similarity of the two materials,
and increased local calcium and phosphate concentra-
tion by HA dissolution in the surrounding medium

[13,14]. Therefore, if HA mineral can be pre-deposited
on the surface in a rapid and simple way, the nucleation
and precipitation process of biomimetic apatite on this
pre-deposited HA mineral surface should be accelerated.

In this study, we have attempted to develop a fast and

easy way to pre-deposit HA minerals on Ti surface using
a self-assembly technique. To our knowledge, only a few
studies have been conducted on the use of self-assembly
techniques to produce Ca/P layers on Ti in SBF [7,9],
but we cannot find reports on inducing a pre-deposited
HA mineral on SAM surfaces before immersion in SBF.
Here, we describe a method of pre-depositing HA
minerals on a Ti surface which is coated with a
functionalized SAM. The functionalized SAM has
potent apatite nucleation groups such as –PO

4

H

2

and

–COOH. The active functional groups are coupled to
the Ti surface using a long (9 carbon chain) 7-oct-1-
enyltrichlorosilane.

2. Materials and methods

2.1. Preparation of Ti surfaces

Square Ti samples approximately 10 10 mm were

prepared from 99.7% pure titanium sheet (2 mm thick,
Aldrich). These were rinsed in distilled water and then in
acetone. After drying in air, the samples were oxidized
using Nanci et al.’s method [15]. That is, the samples
were immersed in a 1:1 (v/v) concentrated H

2

SO

4

and

30% H

2

O

2

mixture for 1 h at 251C. The samples were

rinsed three times with distilled water and two times
with acetone. After the samples were dried in air at
room temperature, the samples were further dried at
1201C for 2 h.

2.2. SAM formation on Ti

SAM formation was carried out in anhydrous

pentane (Aldrich) with 5% (v/v) 7-oct-1-enyltrichloro-
silane (OETS, United Chemical Technology). More
specifically, 20 pieces of dried Ti were immersed in 10 ml
5% OETS solution under an argon atmosphere. The
reaction vessel was then sealed and shaken for 1 h at
251C.

Then, the Ti samples were rinsed in 20 ml anhydrous

pentane three times and dried in air.

2.3. Functionalization of SAMs

2.3.1. Hydroxylation

The terminal ethylenic double bond of the SAM

formed on Ti was converted to –OH groups through
hydroboration and H

2

O

2

oxidation (Fig. 1) [16]. Tita-

nium samples with a SAM formed on the surface were
immersed in 10 ml B

2

H

6

(THF 1 m) for 30 min while

Q. Liu et al. / Biomaterials 23 (2002) 3103–3111

3104

shaking. Then, samples were transferred to 6 ml of
NaOH (0.1 m), followed by addition of 5 ml 30% H

2

O

2

.

The reaction was continued for 5 min.

2.3.2. Phosphorylation

Phosphorylation of the SAM followed the hydroxyla-

tion of the SAM on Ti (Fig. 1). The samples with a
hydroxylated SAM were placed in 0.2 m phosphorous
oxychloride (POCl

3

, Aldrich) and 0.2 m 2,4,6-collidine

(Aldrich) in dry acetonitrile under argon atmosphere for
1 h at 251C. After thorough rinsing with acetonitrile,
the samples were further washed three times with
copious amount of distilled water and dried in air.

2.3.3. Carboxylation

The terminal ethylenic double bond of the SAM

formed on Ti was converted to –COOH groups through
oxidation (Fig. 1). Samples were immersed in 5% acidic
(w/w) KMnO

4

for 5 min at 251C, thoroughly rinsed with

water, and dried in air.

2.4. Nucleation of calcium phosphate

The calcium phosphate deposition experiment was

carried out in the presence of a SAM sample using a
fast precipitation method [12]. Briefly, six pieces of
titanium samples were immersed in 20 ml of 20 mm
aqueous CaCl

2

solution buffered at pH 7.4 with Tris-

HCl buffer at 251C. Then, 20 ml of 12 mm K

2

HPO

4

solution was slowly dropped on the Ti sample/CaCl

2

mixture during 5 min under vigorous stirring. The
resulting mixture was further stirred for 30 min. The
deposition process was repeated two more times. The Ti
samples were washed in 50 ml of distilled water three
times and dried in air.

A control experiment was also performed. First, a

calcium phosphate suspension was prepared using the
same methods as mentioned above without the presence
of Ti. Then, the functionalized samples were immedi-
ately immersed in the calcium phosphate suspension.
The suspension was stirred for 2 h, then the samples
were rinsed in distilled water for three times and
subsequently dried in air.

2.5. Biomimetic coating in 1.5 SBF

1.5 SBF has ionic concentrations 1.5 times of that of

normal SBF. The composition of 1.5 SBF is as follows
(in mm): Na

+

213, K

+

7.5, Ca

2+

3.8, Mg

2+

2.3, HCO

3

6.3, Cl

223, HPO

4

2

1.5, SO

4

2

0.75. The solution was

prepared by dissolving NaCl, NaHCO

3

, KCl, K

2

HPO

4

,

MgCl

2

6H

2

O, CaCl

2

and Na

2

SO

4

in Tris-HCl buffer at

pH 7.4 (371C). Each Ti sample was immersed in 40 ml
1.5 SBF for 18 h in a polystyrene beaker. The beakers
were placed in a shaking water bath at 371C.

2.6. Characterization of Ti surfaces

2.6.1. ESCA

ESCA results were obtained using an ESCA Scientia

300 spectrometer. An Al

K

X-ray source was used and

focused on an area of 4 0.2 mm. The take off angle
was 601. Data analysis and curve fitting were done using
proprietary Scientia software.

2.6.2. Contact angle measurement

The static contact angle of water on the prepared

surfaces was measured at 251C using a contact angle
goniometer by placing 10 ml of double distilled water on
the Ti surface. The droplet was imaged using a video
camera coupled to a light microscope, and the contact
angle was determined on the screen of the monitor using
imaging software. Three measurements were made on
each sample.

2.6.3. Raman spectroscopy

Raman spectroscopy (Spex1401, SPEX Industries,

Inc.) with a double monochrometer was used to
characterize the Ca/P coatings on Ti surfaces. The
scanning rate was 2 cm

1

s

1

, and 1 s integration time

was used. The samples were illuminated with about
30 mW of 514.5 nm radiation by an argon-ion laser. All
of the spectra were recorded in a backscattering
geometry with a 80 microscope objective. The laser
beam was focused on a 1 mm spot using a confocal

1. B

2

H

6

/THF (1M)

2. H

2

O

2

/NaOH(0.1M)

POCl

3

2,4,6,-collidine

Si

Ti

Si

Si

PO

4

H

2

Si

Ti

PO

4

H

2

Si

PO

4

H

2

Si

OH

Si

Ti

OH

Si

OH

Si

H

+

KMnO

4

COOH

Si

Ti

COOH

Si

COOH

Si

Fig. 1. Methods to prepare functionalized SAM on Ti substrate.

Q. Liu et al. / Biomaterials 23 (2002) 3103–3111

3105

microscope. The depth of the field on which light can be
collected for a transparent sample is approximately
20 mm. The thickness of the surface calcium phosphate
layer is

o20 mm, and therefore, the intensity of

the signal is proportional to the thickness of the surface
calcium phosphate layer.

2.6.4. Scanning electron microscopy (SEM)

A JEOL 6300F SEM was used to examine the surface

of Ca/P coated Ti. The samples were sputtered with
carbon before observation.

2.6.5. FTIR spectroscopy

The calcium phosphate precipitate in the suspension

used for the calcium phosphate deposition experiment
was collected and dried at 501C for 48 h. The precipitate
was used to make a KBr pellet for FTIR (Mattson
Research Series I) analysis.

2.6.6. ATR-FTIR spectroscopy

The surfaces of Ca/P coated surfaces were character-

ized with an ATR-FTIR microscope (JASCO FT/IR-
300E). Scans were repeated 65 times with a resolution of
4 cm

1

.

2.6.7. XRD

X-ray diffraction analysis (Rigaku, D/Max-B, hor-

izontal diffractometer with curved crystal monochrom-
eter)

was

performed

on

the

calcium

phosphate

precipitate collected from the same suspension used for
the calcium phosphate deposition experiment.

3. Results and discussion

3.1. XPS spectra

Fig. 2 shows the XPS spectra of titanium and surface

modified titanium. All spectra showed Ti

3s

and Ti

3p

peaks in the region of 0–250 eV. However, after
formation of a SAM on the surface of Ti, Si

2s

and Si

2p

peaks were observed indicating the presence of silane.
After phosphorylation of the SAM, P

2s

and P

2p

peaks

were observed indicating that the –PO

4

H

2

groups were

successfully introduced onto the surface of Ti. The SAM
with –OH and –COOH exhibited similar spectra in this
region.

3.2. Contact angle measurement

The contact angle of water on Ti after formation of a

SAM strongly depends on the surface functional groups
present (Fig. 3). After oxidation in H

2

SO

4

/H

2

O

2

, a

significant decrease in contact angle was observed.

Fig. 2. ESCA spectra showing the appearance of Si

2s

, Si

2p

,P

2s

and P

2p

peaks after SAM formation and functionalization.

68.3

45

105

46.8

48.3

67.7

0

20

40

60

80

100

120

Ti

H2O2/H2SO4

Ti---CH2=CH2

Ti---OH

Ti---PO4H2

Ti---COOH

Sample treatment

Contact angle (degree)

Fig. 3. Contact angles of Ti with various functional groups.

Q. Liu et al. / Biomaterials 23 (2002) 3103–3111

3106

However, after formation of a SAM with OETS, the
surface became highly hydrophobic, and the contact
angle increased from 451 to 1051. After hydroxylation of
the SAM, the contact angle decreased from 1051 to 67.71
due to the introduction of –OH groups. The contact
angle decreased further after the phosphorylation or
carboxylation of the SAM.

3.3. SEM observation

Neither oxidation nor SAM formation produced

noticeable changes in the Ti surface morphology at the
magnification used. Before calcium phosphate deposi-
tion, all samples exhibited roughened surfaces similar to
that shown in Fig. 4A. After deposition of calcium
phosphate, all samples from the control experiment, as
received Ti, Ti silanated with OETS, and Ti with a
hydroxylated SAM showed no change in surface
morphology, while oxidized Ti, Ti with phosphorylated
or carboxylated SAM were covered with calcium
phosphate precipitates on the surfaces (Fig. 4B and
D–F). However, the oxidized Ti samples showed less
calcium phosphate deposition (Fig. 4B) as compared to
samples with either phosphorylated SAM or carboxy-
lated SAM (Fig. 4D and F). The phosphorylated SAM
samples showed almost complete coverage by a calcium
phosphate layer and globular calcium phosphate pre-
cipitates were visible attached onto the substrate surface
(Fig. 4E). The carboxylated SAM samples showed
incomplete coverage by calcium phosphate, and the
formed globular calcium phosphate spots were firmly
attached to the surface.

After further incubation in 1.5 SBF for 18 h, the

samples without previously deposited Ca/P did not show
any change in their surface morphologies. All samples
with a previously nucleated calcium phosphate layer
exhibited significant changes in surface morphology
(Fig. 5) indicating that a new layer of mineral precipitate
was formed on the previously deposited Ca/P layer. The
newly formed mineral layer showed many globules on
the surface. This morphology is commonly seen in other
studies with SBF. Fig. 5 gives typical SEM pictures of
such mineral layers.

3.4. Raman spectroscopy

Fig. 6 shows the Raman spectra of Ca/P coatings on

titanium samples. All Ca/P precipitates exhibited peaks
at 960 cm

1

ascribed to the –PO

4

group. Non-oxidized

Ti samples and samples with an OETS SAM and a
hydroxylated SAM did not exhibit these peaks (spectra
not shown). The spectra of all samples with a rapidly
deposited Ca/P coating had low intensity peaks at
960 cm

1

. After immersion in 1.5 SBF, the intensity of

peaks at 960 cm

1

increased significantly, indicating that

more Ca/P was present on the surfaces of these samples
after 1.5 SBF immersion.

3.5. ATR-FTIR

After fast Ca/P deposition, oxidized Ti, and Ti with

phosphorylated or carboxylated SAMs showed a –PO

4

absorption band at 1040 cm

1

, indicating that a Ca/P

mineral layer was formed on these samples (Fig. 7). The
as-received Ti, and Ti with a –CHQCH

2

or –OH

functionalized SAM did not show this absorption band.
After immersion in 1.5 SBF, all samples with pre-
deposited Ca/P showed a significant increase in peak
intensity at 1040 cm

1

, indicating that more Ca/P was

formed on the surface of the samples after immersion in
1.5 SBF.

3.6. FTIR

The FTIR spectrum of the calcium phosphate

precipitate in the solution (not the mineral precipitated
on the surface of Ti) obtained during the rapid Ca/P
deposition process exhibited H

2

O absorption bands at

3460 and 1640 cm

1

, and P–O stretching bands at 1110,

1046, 964, and 882 cm

1

(Fig. 8). All these bands are

similar to the typical IR bands of poorly crystallized
hydroxyapatite, suggesting that the precipitates are
poorly crystallized HA.

3.7. XRD

The XRD spectrum of the calcium phosphate

precipitate collected from the suspension in the calcium
phosphate deposition experiment showed that all peaks
matched the index card of hydroxyapatite but the peaks
were broad and of low intensity (Fig. 9). The result
confirmed that the calcium phosphate obtained in the
deposition experiment is poorly crystallized hydroxya-
patite.

4. Discussion

Since the introduction of Ca/P coatings onto Ti and its

alloys will greatly accelerate the bone-bonding process of
surgical prostheses, various methods have been used to
introduce Ca/P coating on Ti. In this study, we
introduced functional groups on the surface of Ti as a
means of nucleating Ca/P precipitation on Ti surfaces and
accelerating subsequent deposition of additional mineral.

When the as-received Ti was oxidized in H

2

SO

4

/H

2

O

2

,

the contact angle of water was significantly decreased.
This decrease could have several possible causes: first,
the oxidation process increases the thickness of the oxide
layer [15], and second, the number of surface hydroxyl
groups could increase due to the use of H

2

O

2

[17,18]. An

Q. Liu et al. / Biomaterials 23 (2002) 3103–3111

3107

additional cause for the decrease in contact angle is that
the oxidation process increases surface roughness (in the
sub-micron range) [15].

The formation of a SAM on the surface of Ti was

confirmed by ESCA studies (Fig. 2) and contact angle
measurement (Fig. 3). The presence of OETS on the Ti

Fig. 4. SEM pictures of titanium surfaces before and after calcium phosphate deposition. (A) Oxidized Ti surfaces before calcium phosphate
deposition. (B) Oxidized Ti surface after calcium phosphate deposition. (C) Ti surface with hydroxylated SAM after calcium phosphate deposition.
No calcium phosphate was observed. (D) and (E) Ti surface with phosphorylated SAM after calcium phosphate deposition. Calcium phosphate was
observed on the surface. (F) Ti surface with carboxylated SAM after calcium phosphate deposition. Calcium phosphate was observed on the surface
(scale bar=20 mm).

Q. Liu et al. / Biomaterials 23 (2002) 3103–3111

3108

surface was confirmed by the Si

2s

and Si

2p

peaks on

ESCA spectra. After phosphorylation of the SAM, P

2s

and P

2p

peaks from the –PO

4

H

2

functional groups of the

SAM were observed. The increase in contact angle after
OETS SAM formation (from

B451 to 1051) indicated

that the surface of the Ti was covered by the SAM layer.
The structure of OETS shows that it has a hydrophobic
chemical structure after it is immobilized on the surface
of Ti (Fig. 1). The introduction of hydroxyl end groups
to the SAM after hydroxylation decreases the contact
angle by increasing the hydrophilicity of the SAM layer,
though the hydrophilicity of oxidized Ti is still greater.
After the phosphorylation treatment, the contact angle
decreased further and indicated that the –PO

4

H

2

groups

were introduced to the SAM. Since the –PO

4

H

2

group is

highly hydrophilic, the contact angle of the phosphory-
lated SAM surface decreased. The carboxylation process
also leads to a decrease in the contact angle from 1051
to 48.31, indicating that the carboxylation process

was successful. The carboxyl group also is highly
hydrophilic.

After rapid Ca/P deposition, SEM observation

showed that only oxidized Ti, phosphorylated, and
carboxylated SAM samples have Ca/P deposits on the
surface (Fig. 4B and D–F). Raman spectroscopy (Fig. 6)
as well as ATR-FTIR (Fig. 7) confirmed the presence of
the Ca/P layer on the surface of samples. This Ca/P
deposition process was previously used by Kato et al.
[12] to introduce HA coatings onto surface-modified
poly(ethylene terephthalate) (PET) films. However,
SEM observation did reveal that a difference in calcium
phosphate coverage. Phosphoralated SAM showed
the highest Ca/P coverage while the oxidized Ti sample
showed the least. These results suggested that the
nucleation ability of the surface functional groups is
different.

For nucleation to occur, an activation energy barrier

must be exceeded. This activation energy can be

Fig. 5. SEM pictures showing that after immersion in 1.5 SBF for 18 h, biomimetic mineral layers were formed on SAM surfaces with pre-deposited
calcium phosphate. (A) On oxidized Ti surface with pre-deposited Ca/P; (B) on phosphorylated SAM with pre-deposited Ca/P; (C) on carboxylated
SAM with pre-deposited Ca/P (scale bar=20 mm).

Q. Liu et al. / Biomaterials 23 (2002) 3103–3111

3109

decreased by increasing the degree of super-saturation
or by decreasing the interfacial energy. The surface
functional groups play important roles in determining

the solid/ion cluster interfacial energies and therefore,
different functional groups can have different nucleation
abilities for calcium phosphate. –PO

4

H

2

and –COOH

are negatively charged at pH 7.4, and it is likely that
they will attract more Ca

2+

ions from solution so that

the interfacial energies of Ti surface/ion clusters are
effectively decreased. In fact, several studies have found
that the Ca

2+

ion is the first to be adsorbed to the Ti

surface when calcium phosphate formed [10,19]. In this
study, –PO

4

H

2

showed that it has an even stronger

capability of nucleating calcium phosphate deposition
than –COOH. It is interesting to see that although the
as-received Ti, oxidized Ti, and Ti with a hydroxylated
SAM all have surface hydroxyl groups, they showed
different abilities in nucleating Ca/P deposition. Since
the oxidized Ti sample in principle had the highest
hydroxyl group density, it is reasonable to assume that
the hydroxyl surface density is also important in
nucleation of calcium phosphate deposition.

It has been reported that the rapidly deposited Ca/P

layer was poorly crystallized hydroxyapatite [12]. Here,
FTIR and XRD analysis of excess precipitate formed

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

350

550

750

950

1150

Rama n Shift (cm-1)

Intensity (cps)

a

b

c

d

e

Fig. 6. Raman spectra of Ca/P deposition on Ti samples with various
surface functional groups. (a) Oxidized Ti samples; (b) Ti samples with
phosphorylated SAM; (c) Ti samples with carboxylated SAM; (d) after
sample in (b) was immersed in 1.5 SBF for 18 h; (e) Ca/P precipitates
collected from the suspension in the Ca/P deposition experiment.

82

87

92

97

102

600

800

1000

1200

Wavenumber (cm-1)

Transmittance (%)

A

B

C

D

Fig. 7. ATR-FTIR spectra of the surface deposited Ca/P coating
showing P–O absorption bands at 1040 cm

1

. (A) Ca/P deposited on

oxidized Ti sample; (B) Ca/P deposited on Ti with phosphorylated
SAM; (C) Ca/P deposited on Ti with carboxylated SAM; (D)
biomimetic apatite coating on oxidized Ti with pre-deposited Ca/P
on phosphorylated SAM.

0

20

40

60

80

100

600

1100

1600

2100

2600

3100

3600

Wavenumber (cm-1)

Transmittance (%)

Fig. 8. FTIR spectrum from the calcium phosphate precipitate
collected from the suspension in calcium phosphate deposition
experiment.

0

200

400

600

800

1000

0

20

40

60

Degrees 2

θ

Intensity

Fig. 9. XRD spectrum of the Ca/P precipitate obtained from the
suspension in Ca/P deposition experiment.

Q. Liu et al. / Biomaterials 23 (2002) 3103–3111

3110

during Ca/P deposition also showed a poorly crystal-
lized hydroxyapatite to be present (Figs. 8 and 9).
Therefore, we believe that the rapidly deposited Ca/P
layer on the surface was composed of poorly crystallized
hydroxyapatite.

The incubation of Ti samples without pre-deposited

Ca/P on the surface in 1.5 SBF showed no Ca/P layer
formation after a 3-day incubation period at 371C. It
has been shown that surface functional groups intro-
duced by a SAM or other methods could induce Ca/P
precipitation only after an extended period of time
(>1 week) in SBF or 1.5 SBF [9–12]. In this study, the
samples with pre-deposited Ca/P coating showed rapid
(less than a day) new biomimetic apatite formation in
1.5 SBF. As seen in Fig. 5, a continuous layer of apatite
was quickly formed on the Ti samples with pre-
deposited Ca/P layer. Raman and ATR-FTIR spectra
indicated that the intensity of the –PO

4

bands was

greatly increased on these samples incubated in 1.5 SBF
(Figs. 6 and 7). These results indicated that the pre-
deposited Ca/P acts as an effective nucleation surface to
induce formation of a biomimetic apatite coating on the
Ti surface.

Acknowledgements

The authors of this research acknowledge the alloca-

tion of time and services in the SCIENTA ESCA
laboratory of Lehigh University. The professional and
technical assistance of Dr. Alfred C. Miller is greatly
appreciated.

This study was supported by USPHS grant DE 12345.

References

[1] de Groot K, Wen HB, Liu Y, Layrolle P, Barrere F. Biomimetic

coatings on orthopedic implants: a review. In: Li P et al., editors.
Mineralization in natural and synthetic biomaterials. Warrendale,
PA: Materials Research Society, 2000. p. 109–16.

[2] Li PJ, de Groot K. A calcium phosphate formation within sol–

gel-prepared titania in vitro and in vivo. J Biomed Mater Res
1993;27:1495–500.

[3] Ohtsuki C, Iida H, Hayakawa S, Osaka A. Bioactivity of titanium

treated with hydrogen peroxide solutions containing metal
chlorides. J Biomed Mater Res 1997;35:39–47.

[4] Kokubo T, Miyaji F, Kim H-M, Nakamura T. Spontaneous

apatite formation on chemically treated titanium metals. J Am
Ceram Soc 1996;1996:1127–9.

[5] Kim H-M, Miyaji F, Kokubo T, Nishiguchi S, Nakamura T.

Graded surface structure of bioactive titanium prepared by
chemical treatment. J Biomed Mater Res 1999;45:100–7.

[6] Wen HB, de Wijn JR, Liu Q, de Groot K, Cui FZ. A simple

method to prepare calcium phosphate coatings on Ti6Al4V.
J Mater Sci: Mater Med 1997;8:765–70.

[7] Campbell AA, Fryxell GE, Linehan JC, Graff GL. Surface-

induced mineralization: a new method for producing calcium
phosphate coatings. J Biomed Mater Res 1996;32:111–8.

[8] Campbell AA, Song L, Li XS, et al. Development, characteriza-

tion, and anti-microbial efficacy of hydroxyapatite-chlorhexidine
coatings produced by surface-induced mineralization. J Biomed
Mater Res 2000;53:400–7.

[9] Mao C, Li H, Cui F, Feng Q, Wang H, Ma C. Oriented growth of

hydroxyapatite on (0 0 1) textured titanium with functionalized
self-assembled silane monolayer as template. J Mater Chem
1998;8:2795–801.

[10] Tanahashi M, Matsuda T. Surface functional group dependence

on apatite formation on self-assembled monolayers in a simulated
body fluid. J Biomed Mater Res 1997;34:305–15.

[11] Sato K, Kumagai Y, Tanaka J. Apatite formation on organic

monolayers in simulated body environment. J Biomed Mater Res
2000;50:16–20.

[12] Kato K, Eika Y, Ikada Y. Deposition of a hydroxyapatite thin

layer onto a polymer surface carrying grafted phosphate polymer
chains. J Biomed Mater Res 1996;32:687–91.

[13] Weng J, Liu Q, Wolke JGC, Zhang XD, de Groot K. Formation

and characteristics of the apatite layer on plasma-sprayed
hydroxyapatite coatings in simulated body fluid. Biomaterials
1997;18:1027–35.

[14] Weng J, Liu Q, Wolke JGC, Zhang XD, de Groot K. The role of

amorphous phase in nucleating bone-like apatite on plasma-
sprayed hydroxyapatite coatings in simulated body fluid. J Mater
Sci Lett 1997;16:335–7.

[15] Nanci A, Wuest JD, Peru L, et al. Chemical modification of

titanium surfaces for covalent attachment of biological molecules.
J Biomed Mater Res 1997;40:324–35.

[16] Brown AC. Organic synthesis via boranes. New York: Wiley,

1975.

[17] Pan J, Thierry D, Leygraf C. Hydrogen peroxide toward

enhanced oxide growth on titanium in PBS solution: blue
coloration and clinical relevance. J Biomed Mater Res 1996;
30:393–402.

[18] Pan J, Liao H, Leygraf C, Thierry D, Li J. Variation of oxide

films on titanium induced by osteoblast-like cell culture and the
influence of an H

2

O

2

pretreatment. J Biomed Mater Res

1998;40:244–56.

[19] Yang BC, Weng J, Li XD, Zhang XD. The order of calcium

and phosphate ion deposition on chemically treated titanium
surfaces soaked in aqueous solution. J Biomed Mater Res 1999;
47:213–9.

Q. Liu et al. / Biomaterials 23 (2002) 3103–3111

3111