ARTICLE IN PRESS
Biomaterials 25 (2004) 5575–5581
Biomimetic growth of apatite on hydrogen-implanted silicon
Xuanyong Liua,b, Ricky K.Y. Fua, Ray W.Y. Poona, Peng Chena, Paul K. Chua,*,
Chuanxian Dingb
a
Department of Physics and Materials Science, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong
b
Shanghai Institute of Ceramics, Chinese Academy of Sciences, 1295 Dingxi Road, Shanghai 200050, China
Received 30 September 2003; accepted 27 December 2003
Abstract
Hydrogen in silicon has been widely applied in semiconductor fields. In this paper, the application of hydrogen-implanted silicon
wafer in biomedical fields was explored by investigating its bioactivity. Hydrogen implanted silicon wafers were prepared using
plasma immersion ion implantation. The surface structures of the 1.4 1017 cm 2 hydrogen-implanted silicon wafers were
investigated using atomic force microscopy and transmission electron microscopy (TEM). The hydrogen depth profiles were
acquired by SIMS and the crystal quality of the as-implanted silicon was studied by channeling Rutherford backscattering
spectrometry (RBS). The bioactivity of the implanted silicon was evaluated using the biomimetic growth of apatite on its surface
after it was soaked in simulated body fluid for a period of time. The TEM, SIMS and RBS results indicate the formation of an
amorphous hydrogenated silicon (a-Si:Hx) layer has been formed on the surface of the hydrogen-implanted silicon wafer. After
immersion in SBF for 14 days, bone-like apatite is observed to nucleate and grow on the surface. With longer soaking time, more
apatite appeared on the surface of the hydrogen implanted silicon but our control experiments did not reveal any apatite formation
on the surface of the un-implanted silicon wafer, hydrogenated crystalline silicon wafer (with hydrogen, but no amorphous surface),
or argon-implanted silicon wafer (amorphous surface but without hydrogen). Our results indicated that the bioactivity of silicon
wafer can be improved after hydrogen implantation and the formation of the amorphous hydrogenated silicon (a-Si:Hx) surface also
plays a synergistic role to improve the bioactivity.
r 2004 Elsevier Ltd. All rights reserved.
Keywords: Hydrogen-implanted silicon; Plasma; Bioactivity; Apatite
1. Introduction
Since the discovery of SiO2-based bioglass by Hench
et al. in 1969 [1], many attempts have been made to
explore the possibility of some silicon-containing
materials applied in biological fields, such as wollastonite, silica gel and porous silicon [2–6]. Silicon by itself
was generally considered to be nonessential in biochemistry, except in certain primitive organisms like diatoms
[7]. However, in the past 30 years, silicon has gradually
been recognized as an essential trace element in the
normal metabolism of higher animals. Carlisle has
shown that silicon is required in bone, cartilage and
connective tissue formation as well as several other
important metabolic processes [8]. Some researchers
*Corresponding author. Tel.: +852-278-877-24; fax: +852-278-89549/878-30.
E-mail address: [email protected] (P.K. Chu).
0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2004.01.015
have confirmed that silicon is involved in an early stage
of bone formation. Hidebrant et al. [9,10] have used
modern genetic engineering techniques to demonstrate
that certain genes are activated by hydrated silicon.
Hydrated soluble silicon will enhance the proliferation
of bone cells (osteoblasts) and active cellular production
of transforming growth factors [10,11]. Xynos et al.
[12,13] have also shown that the critical concentration of
ionic products dissolved from the bioactive glass
composed of soluble silicon and calcium ions can
enhance osteogenesis through direct control over genes
that regulate cell cycle induction and progression.
Therefore, an increasing number of silicon-containing
materials are being studied for applications in biomedical materials and devices.
Hydrogen in silicon has been widely investigated in the
semiconductor area. It has been discovered that hydrogen in silicon greatly changes the electrical property
of the resultant electronic devices by passivating
ARTICLE IN PRESS
X. Liu et al. / Biomaterials 25 (2004) 5575–5581
5576
0.5 mTorr. The instrumental parameters are listed in
Table 1. Previous results have shown that under these
conditions in our instrument, H+
3 is the dominated ion
species in the plasma [24].
The surface structure and roughness of hydrogenimplanted silicon was investigated using atomic force
microscopy (AFM) and cross-sectional transmission
electron microscopy (XTEM) was performed to identify
the nature of defects produced by ion implantation. The
crystalline quality of the as-implanted samples was
assessed using channeling Rutherford Backscattering
Spectrometry (c-RBS) using an incident beam of 2 MeV
He2+ at a backscattering angle of 170 . Secondary ion
mass spectrometry (SIMS) measurements were carried
out using a Cameca IMS-4F with a Cs+ primary beam
and the intensities of 28Si2+ and H+ ions sputtered from
the center of a 150-mm-diameter crater were monitored.
After ultrasonically washed in acetone and rinsed in
deionized water, the silicon wafers prior to and after
hydrogen-implanted were soaked in a simulated body
fluid (SBF) for 14 and 28 days. The SBF solution was
not replenished during the soaking procedure. The SBF
solution was buffered at pH 7.4 with trimethanol
aminomethane-HCl. The ionic concentrations in the
solution are nearly equal to those in human body blood
plasma, as shown in Table 2 [25].
The surface views of silicon wafers prior to and after
soaking in the simulated body fluid were observed using
cold field emission scanning electron microscopy (SEM).
The structure and phase compositions of the surfaces
were analyzed by thin film X-ray diffraction (TF-XRD)
at a glancing incident angle of 2 , X-ray fluorescence
(XRF) as well as Fourier transform infrared spectroscopy (FTIR).
shallow-level and deep-level defects [14]. It has been
further reported that high dose implantation of hydrogen and subsequent annealing induce splitting of Si,
which is utilized in the fabrication of silicon on insulator
(SOI) materials [15]. As a result, a large amount of
literature reporting the structure and properties of
hydrogen in silicon can be found [16–20], but very few
research groups have addressed the role of hydrogen in
silicon in biological applications. Dahmen and his
colleagues [21] have shown that surface functionalization
of amorphous hydrogenated silicon (a-Si:H) and amorphous silicon suboxide films (a-SiOx:H) by hydrosilylation reaction have been proved to be largely
biocompatible. Based on our literature search, the
application of hydrogenated silicon to improve the
bioactivity or bone conductivity of silicon has pretty
much been unexplored, but the materials could have
exciting implications in biomedical engineering and
biosensor technologies because silicon-based devices
often suffer from problems associated with interfacing
to the biological environment. The concept of utilizing Si
itself as a bioactive miniaturizable packaging material
might provide a solution to some of these problems.
In this work, hydrogen was implanted into silicon
wafer using plasma immersion ion implantation (PIII).
The bioactivity or bone conductivity of hydrogenimplanted silicon was investigated using a simulated
body fluid (SBF) soaking test.
2. Experimental methods
Single crystal /1 0 0S silicon wafers measuring
100 mm polished on one side were used in the experiments. The root mean square (RMS) roughness of the
polished surface is about 0.215 nm. Hydrogen was
implanted into the polished side using PIII in the
plasma laboratory of the City University of Hong Kong
[22,23]. The background pressure was pumped to
0.6 mTorr and high-purity hydrogen gas was bled into
the vacuum chamber to establish a working pressure of
3. Results
The AFM images of the hydrogen-implanted silicon
wafer displayed in Fig. 1 indicate that the surface of the
hydrogen-implanted silicon wafer is quite smooth with a
Table 1
Implantation parameters
Implantation voltage (kV)
Pulse frequency (Hz)
Pulse duration (ms)
30
50
500
RF Discharge power (W)
Implantation time (minutes)
Implantation dose (cm 2)
1400
20
B1.4 1017
Table 2
Ion concentration of SBF in comparison with human blood plasma
Concentration (mM)
SBF
Blood plasma
Na+
K+
Ca2+
Mg2+
HCO3
Cl
HPO24
SO24
142.0
142.0
5.0
5.0
2.5
2.5
1.5
1.5
4.2
27.0
148.5
103.0
1.0
1.0
0.5
0.5
ARTICLE IN PRESS
X. Liu et al. / Biomaterials 25 (2004) 5575–5581
calibration and instrumentation issues) followed by a
dislocation zone located close to the projected range of
H+
3 . Hence, the implanted sample consists of a highly
hydrogen-doped surface with high crystalline disorder.
Figs. 4 and 5 depict the surface views of the unimplanted and hydrogen-implanted samples after soaking in a simulated body fluid for 14 and 28 days.
After 14 days immersion in the simulated body fluid,
the surface of un-implanted silicon wafer remains
smooth similar to that of a silicon wafer before
immersion (Fig. 4a), while some single and clustered
ball-like particles are observed on the surface of the
hydrogen-implanted silicon surface (Fig. 5a). The surface of silicon wafer is, however, not covered completely. The higher magnification micrograph indicates that
they have a coral-like structure composed of many
1022
5
SIMS
RBS
Fig. 1. AFM view of the surface of the hydrogen-implanted silicon
wafer.
Concentration (cm-3)
4
3
1021
2
1
1020
0
Atomic displacement density (x1022cm-3)
RMS roughness of about 0.235 nm which is closed to the
roughness of the silicon wafer before hydrogen implantation. The cross-section TEM micrograph of the
hydrogen-implanted silicon wafer in Fig. 2 reveals the
formation of defects. There exists a top amorphous zone
(about 60 nm in thickness) and a dense dislocation zone
(about 150 nm). The dense dislocation zone is located
around the projected range of H3+. Fig. 3 plots the
atomic dislocation density versus depth derived from
c-RBS and the hydrogen elemental depth profile
acquired by SIMS. Hydrogen is mainly found on the
near surface to about 170 nm (B8 at%), which is the
projected range of H+
3 obtained from the SRIM code. A
total atomic displacement zone (amorphous silicon a-Si)
is extended from the top surface to the depth of about
50 nm (slightly different from the TEM results due to
5577
0
100
200
300
400
Depth (nm)
Fig. 3. Depth profile of defects and hydrogen ion concentration in
hydrogen-implanted silicon wafer obtained from RBS and SIMS. (Left
ordinate: the hydrogen ion concentration; Right ordinate: Silicon
atomic displacement density.)
Fig. 2. Cross-section views of hydrogen-implanted silicon wafer
obtained from TEM analyzing.
Fig. 4. Surface views of silicon wafer soaked in SBF for 14 days (a)
and 28 days (b).
ARTICLE IN PRESS
5578
X. Liu et al. / Biomaterials 25 (2004) 5575–5581
Fig. 5. Surface views of hydrogen-implanted silicon wafer soaked in
SBF for: (a) 14 days, (b) 28 days and (c) the higher magnification of the
ball-like particles in (a).
sheet-like crystallites (Fig. 5c). After an immersion time
of 28 days, the number of these ball-like particles
increases, and the surface is totally covered by the newly
formed layer (Fig. 5b). In contrast, no new substance
can be found on the surface of the un-implanted silicon
wafer even after soaking in SBF for 28 days (Fig. 4b).
Fig. 6 shows the XRF spectra of the silicon wafers
soaked in simulated body fluid for 14 and 28 days.
Calcium and phosphorus cannot be detected on the
surface of the un-implanted silicon wafer after soaking
for 14 and 28 days (Fig. 6a and b) indicating that no
Ca–P layer has formed on the surface. On the other
hand, the XRF spectra acquired from the hydrogenimplanted silicon wafer soaked in SBF for 14 days show
the existence of calcium and phosphorus on the surface
(Fig. 7a). After immersion in SBF for 28 days, more
calcium and phosphorus are detected indicating the
formation of a denser and thicker Ca–P layer (Fig. 7b).
The atomic ratios of Ca to P calculated from the XRF
Fig. 6. XRF spectra of silicon wafer soaked in SBF for: (a) 14 days and (b) 28 days.
Fig. 7. XRF spectra of hydrogen-implanted silicon wafer soaked in SBF for: (a) 14 days and (b) 28 days.
ARTICLE IN PRESS
X. Liu et al. / Biomaterials 25 (2004) 5575–5581
spectra of the hydrogen-implanted silicon soaked in SBF
for 14 and 28 days are about 1.33 and 1.58, respectively.
Our data show that the atomic ratio of Ca to P in the
Ca–P layer increases gradually to that of hydroxyapatite
(1.67) with longer immersion time in SBF. This indicates
that the Ca–P layer formed on the surface of the
hydrogen-implanted silicon can crystallize to form
hydroxyapatite with increasing immersion time in SBF
based on the supporting XRD results that will be
discussed as follows.
The XRD patterns obtained from the un-implanted
and hydrogen-implanted silicon wafers before and after
immersion in SBF for 14 and 28 days are shown in
Fig. 8. Only crystalline silicon peaks are observed while
no crystalline apatite peaks appear in the XRD patterns
Si
(a)
Si
Si
14d
Before immersion
20
30
40
50
60
2-theta (degree)
(b)
Transmittance (orb. units)
Si
A-- Apatite
28d
of the un-implanted silicon wafer soaked in SBF for 14
days and 28 days (Fig. 8a). Compared with the XRD
pattern of the hydrogen-implanted silicon wafer before
immersion, the XRD pattern of the hydrogen-implanted
silicon wafer soaked in SBF for 14 days reveals obvious
changes. One silicon peaks disappears as shown in
Fig. 8b, but the newly formed Ca–P layer cannot yet be
identified unambiguously. The reason for this is believed
to be that the Ca–P layer is not yet crystallized to form
apatite and the amorphous structure still exists. After an
immersion time of 28 days, the peaks of crystalline
apatite can be easily identified in the XRD spectra
indicating the formation of a new surface layer
composed of crystalline apatite. The broadening of
the peaks suggests that the apatite particles formed on
the hydrogen implanted silicon wafer are superfine or
have low crystallinity.
The FTIR spectra acquired from the hydrogenimplanted silicon wafer before and after immersion in
SBF for 14 and 28 days are displayed in Fig. 9. Broad
OH absorption bands from 3700 to 2500 cm 1 and a
weak water absorption band around 1650 cm 1 can be
seen in these spectra. Bands between 1400 and
1550 cm 1 are due to the carbonate IR absorption n3.
The peak around 870 cm 1 arises from the both the
carbonate and HPO24 ions [26]. The broad bands
around 1100 cm 1 is mainly attributed to the phosphate
[26]. According to Canham and co-workers [6], a sharp
P–O bending mode doublet around 600 cm 1 is indicative of a crystalline phase of hydroxyapatite being
present. In Fig. 9, the double peak around 600 cm 1 in
the FTIR spectra of the hydrogen-implanted silicon
wafer soaked in SBF for 28 days is clearer than that in
the hydrogen-implanted silicon wafer soaked in SBF for
14 days.
A
A
A
14d
Before immersion
14d
28d
Si
Before immersion
H-O
P-O
Si
20
P-O
2CO 3
2HPO 4
H 2O
CO 32-
30
40
50
60
2-theta (degree)
Fig. 8. XRD patterns of (a) non-implanted silicon wafer and (b)
hydrogen-implanted silicon wafer soaked in SBF for 14 and 28 days.
4000
3500
3000
2500
2000
1500
1000
Si-O-Si
28d
5579
500
Wavenumber (cm-1)
Fig. 9. FTIR spectra of hydrogen-implanted silicon wafers before and
after immersion in SBF for 14 and 28 days.
ARTICLE IN PRESS
5580
X. Liu et al. / Biomaterials 25 (2004) 5575–5581
Our results obtained from the FTIR and TF-XRD
showed that carbonate-containing hydroxyapatite
(bone-like apatite) forms on the surface of the hydrogen-implanted silicon wafer soaked in SBF, indicating
that the hydrogen-implanted silicon wafer has good
bioactivity.
4. Discussion
The results obtained from the SBF soaking test
conform that apatite cannot form on the surface of
virgin silicon wafers even after soaking in SBF for 28
days, but on the other hand, bone-like apatite can form
on the surface of hydrogen-implanted silicon wafer after
soaking in SBF for a certain time, indicating the
bioactivity of silicon wafer can be improved by
implanting hydrogen.
It is well known that the surface plays an important
role in the response of the biological environment of the
artificial biomedical device. Therefore, it is logical to
believe that the improvement of the bioactivity of the
implanted silicon wafer results from the modified
surface by hydrogen implantation. The results from
TEM, RBS and SIMS reveal the presence of an
amorphous hydrogenated silicon layer (a-Si:Hx) after
hydrogen PIII.
In order to investigate clearly the formation mechanism of bone-like apatite on the surface of the hydrogenimplanted silicon wafer, two comparative experiments
were also conducted. One is to investigate the bioactivity
of hydrogenated silicon wafer with no surface damage
and the other one is to evaluate the bioactivity of argonimplanted silicon wafer which possesses an amorphous
surface but no hydrogen. In our experiments, after the
hydrogenated silicon wafer and argon-implanted wafer
are soaked in SBF for 28 days, no apatite particles can
found on either surface, indicating poor bioactivity on
both samples. Our results suggest that only the
formation of an amorphous hydrogenated silicon (aSi:Hx) surface can improve the bioactivity of silicon
wafer and results in the formation of bone-like apatite
on its surface after treatment in SBF. Experimental
evidence indicates that the formation of apatite requires
that the surface be both amorphous and be hydrogenated. The detailed mechanism as well as further work
are being pursued in our laboratory and will be
promulgated in due course.
5. Conclusion
Hydrogen-implanted silicon wafers were prepared
using plasma immersion ion implantation. The results
obtained from TEM, SIMS and RBS indicated that an
amorphous silicon with hydrogen (a-Si:Hx) layer formed
on the surface of hydrogen-implanted silicon wafer.
After immersion in SBF for 14 days, bone-like apatite
was observed to nucleate and grow on the surface. With
longer soaking time, more apatite appeared on the
surface of the hydrogen-implanted silicon but our
control experiments did not reveal any apatite formation on the surface of the un-implanted silicon wafer,
hydrogenated crystalline silicon wafer (with hydrogen,
but no amorphous surface), or argon-implanted silicon
wafer (amorphous surface but without hydrogen). The
results obtained from this work indicated that the
bioactivity of silicon wafer can be improved after
hydrogen implantation. The formation of the amorphous hydrogenated silicon (a-Si:Hx) surface on the
hydrogen-implanted silicon wafer is the key to improves
the bioactivity of silicon wafer and results in the
formation of bone-like apatite on its surface after it
was soaked in SBF for a period.
Acknowledgements
This work was supported by City University of Hong
Kong Strategic Research Grant #7001447, Shanghai
Science and Technology R&D Fund under grant
02QE14052 and 03JC14074, National Basic Research
Fund under Grant G1999064701.
References
[1] Hench LL, Anderson O. Bioactive glass. In: Hench LL, Wilson J,
editors. An introduction to bioceramics. USA: World Scientific;
1993. p. 41–62.
[2] De Aza PN, Guitian F, De Aza S. Bioactivity of wollastonite
ceramics: in vitro evaluation. Sripta Metall Mater 1994;31:1001–5.
[3] Liu X, Ding C. Reactivity of plasma sprayed wollastonite in
simulated body fluid. J Biomed Mater Res 2002;59(2):259–64.
[4] Liu X, Ding C. Apatite formed on the surface of plasma sprayed
wollastonite coating immersed in simulated body fluid. Biomaterials 2001;22(14):2007–12.
[5] Cho S, Nakanishi K, Kokubo T, Soga N. Dependence of apatite
formation on silica gel on its structure: effect of heat treatment.
J Am Ceram Soc 1995;78:1769–74.
[6] Canham LT, Reeves CL. Apatite nucleation on low porosity
silicon in acellular simulated body fluid. In: Cotell CM, Meyer
AE, Gorbatkin SM, Grobe III GL, editors. Thin films and
surfaces for bioactivity and biomedical applications: symposium
held November 28–29, 1995. Boston, MA, USA, Pittsburgh:
Materials Research Society; 1996. p. 189–94.
.
[7] Karlsson KH, Froberg
K, Ringbom T. A structural approach to
bone adhering of bioactive glasses. J Non-Cryst Solids 1989;112:
69–72.
[8] Carlisle EM. Silicon as an essential trace element in animal
nutrition. In: Evered D, O’Connor M et al., editors. Silicon
biochemistry. Chichester, New York, Sydney, Toronto, Singapore: Wiley; 1986. p. 123–39.
[9] Hidebrand M, Higgins DR, Busser K, Volcani BE. Silicon
responsive c-DNA clone isolated from the marine diatom
cylindrotheca fusiformis. Gene 1993;132:213.
[10] Hench LL. Life and death: the ultimate phase transformation.
Thermochim Acta 1996;280/281:1–13.
ARTICLE IN PRESS
X. Liu et al. / Biomaterials 25 (2004) 5575–5581
[11] Keeting PE, Wiegand KE, Spelssberg TC, Riggs BL. A novel
silicon-containing osteotropic agent, zeolite A, induces proliferation and differentiation of normal human osteoblast-like cells.
American Society of Bone and Mineral Research Annual Meeting, Atlanta, G.A. USA, Abst. #184.
[12] Xynos ID, Hukkanen MVJ, Batten JJ, Buttery ID, Hench LL,
Polak JM. Bioglasss 45S5 stimulates osteoblast turnover and
enhance bone formation in vitro: implications and applications
for bone tissue engineering. Calcif Tiss Int 2000;67:321–9.
[13] Xynos ID, Edgar AJ, Buttery LD, Hench LL, Polak JM. Ionic
products of bioactive glass dissolution increase proliferation of
human osteoblasts and induce insulin-like growth factor II
mRNA expression and protein synthesis. Biochem Biophys Res
Commun 2000;276:461–5.
[14] Corbett JW, Deak P, Desnica UV, Pearton SJ. Hydrogen
passivation of damage centers in semiconductors. In: Pankove
JI, Johnson NM, editors. Hydrogen in semiconductors. San
Diego: Academic Press, Inc.; 1991. p. 49–64.
.
[15] Tong QY, Gosele
UM. Wafer bonding and layer splitting for
microsytems. Adv Mater 1999;11(17):1409–25.
[16] Piatkowska A, Gawlik G, Jagielski J. AFM studies of hydrogen
implanted silicon. Appl Surf Sci 1999;141:333–8.
[17] Grisolia J, Cristiano F, Ben Assayag G, Claverie A. Kinetic
aspects of the growth of platelets and voids in H implanted Si.
Nucl Instrum and Methods B 2001;178:160–4.
[18] Reboredo FA, Ferconi M, Pantelides ST. Theory of the
nucleation, growth and structure of hydrogen induced extended
defects in silicon. Phys Rev Lett 1999;82(24):4870–3.
5581
[19] Zhang SB, Jackson WB. Formation of extended hydrogen
complexes in silicon. Phys Rev B 1991;43(14):12142–5.
[20] Iwata H, Takagi M, Tokuda Y, Imura T. Analysis of platelet
distribution in H ion-implanted silicon. J Crystal Growth 2000;
210:94–7.
[21] Dahmen C, Janotta A, Dimova-Malinovska D, Marx S, Jeschke
B, Nies B, Kessler H, Stutzmann M. Surface functionlization of
amorphous silicon and silicon suboxides for biological applications. Thin Solid Films 2003;427:201–7.
[22] Chu PK, Qin S, Chan C, Cheung NW, Larson LA.
Plasma immersion ion implantation—A fledgling technique for
semiconductor processing. Mater Sci Eng: Rep 1996;R17:
207–80.
[23] Chu PK, Tang BY, Cheng YC, Ko PK. Principle and
characteristics of a new generation plasma immersion ion
implanter. Rev Sci Instrum 1997;68:1866–74.
[24] Fan ZN, Zeng XC, Kwok DTK, Chu PK. Surface hydrogen
incorporation and profile broadening caused by sheath expansion
in hydrogen plasma immersion ion implantation. IEEE Trans
Plasma Sci 2000;28:371–5.
[25] Kokubo T, Kushitani H, Sakka S, Kitsugi T, Yamamuro T.
Solutions able to reproduce in vivo surface-structure changes
in bioactive glass–ceramic A–W. J Biomed Mater Res 1990;24:
721–34.
[26] Weng J, Liu Q, Wolke JGC, Zhang X, de Groot K. Formation
and characteristics of the apatite layer on plasma-sprayed
hydroxyapatite coatings in simulated body fluid. Biomaterials
1997;18:1027–35.