Porous poly(L-lactic acid)/apatite composites created by

biomimetic process

Ruiyun Zhang, Peter X. Ma

Departments of Biologic and Materials Sciences and Biomedical Engineering, Macromolecular Science and Engineering
Center, 1011 North University Avenue, The University of Michigan, Ann Arbor, Michigan 48109-1078

Received 1 April 1998; accepted 4 November 1998

Abstract: Highly porous poly(L-lactic acid)/apatite com- tometry. These porous spherical microparticles were assem-
posites were prepared through in situ formation of carbon- blies of microflakes. They were found to be carbonated
ated apatite onto poly(L-lactic acid) foams in a simulated bonelike apatite. A series of composite foams with varying
body fluid. The highly porous polymer foams (up to 95% sizes and concentrations of the apatite particles was ob-
porosity) were prepared from polymer solution by solid– tained by varying incubation time and conditions. These
liquid phase separation and subsequent sublimation of the porous composites may be promising scaffolding materials
solvent. The foams were then immersed in the simulated for bone tissue engineering and regeneration because the
body fluid at 37°C to allow the in situ apatite formation. excellent bone-bonding properties of the apatite may pro-
After incubation in the simulated body fluid for a certain vide a good environment for osteoblast and osteoprogenitor
period of time, a large number of characteristic micropar- cells’ attachment and growth. © 1999 John Wiley & Sons,
ticles formed on the surfaces of pore walls throughout the Inc. J Biomed Mater Res, 45, 285–293, 1999.
polymer foams. The microparticles were characterized with
scanning electron microscopy, energy dispersive spectros- Key words: composite; polymer; scaffold; foam; hydroxy-
copy, Fourier transform IR spectroscopy, and X-ray diffrac- apatite

INTRODUCTION eration, and vascularization. The scaffolds should be

mechanically stable to keep their structural integrity
Bone fractures and damage are serious health prob- during the tissue regeneration. The scaffolding mate-
lems, which result in more than 1.3 million surgical rials for bone tissue engineering should also be osteo-
procedures each year in the United States.1 The cur- conductive, so that osteoblasts and osteoprogenitor
rent bone repairing or replacing materials include au- cells will attach and migrate on the scaffolds. The
tografts, allografts, and various kinds of manmade poly(␣-hydroxyl acids) such as poly(lactic acid) (PLA),
materials such as polymers, bioceramics, metallic ma- poly(glycolic acid) (PGA), and poly(lactic acid-co-
terials, and their composites. However, none of them glycolic acid) (PLGA) satisfy many of these material
is a perfect solution. Tissue engineering offers a prom- requirements and have been fabricated into scaffolds
ising new approach to bone repair.2–5 In this approach, for cell transplantation and tissue engineering.5–11
a porous scaffold is needed to guide cell attachment, Various bioactive ceramics such as sintered hy-
growth, and tissue regeneration in 3 dimensions.6–9 droxyapatite (HA)12 and glass-ceramic apatite-wollas-
There are several basic requirements for scaffolding tonite(A-W)13 were developed and clinically used as
materials. They should be biocompatible and biode- bone repairing and replacing materials. These bioac-
gradable. The degradation products should be non- tive ceramics were found to bond to bone through a
toxic and easily excreted by metabolic pathways. The layer of bonelike apatite formed on the surface of the
materials should be easy to fabricate into a desired ceramics in the body and were characterized as car-
shape and have a controlled pore architecture to allow bonate-containing HAs with small crystallites and a
for cell infiltration, attachment, growth, tissue regen- defective structure.14–17 The apatite was not observed
at the interface between nonbioactive materials and
bone.15,17 Therefore, it is believed that the bonelike
Correspondence to: Prof. Peter X. Ma; e-mail: mapx@
umich.edu apatite layer plays a key role for bone bonding to these
bioactive ceramics.
© 1999 John Wiley & Sons, Inc. CCC 0021-9304/99/040285-09 Some polymer materials such as Polyactive威 (ethyl-
286 ZHANG AND MA

ene oxide-butylene terephthalate copolymer) were re- reagent grade chemicals of NaCl, NaHCO 3 , KCl,
ported to have the ability to bond to bone.18 On the K 2 HPO 4 ⭈ 3H 2 O, MgCl 2 ⭈ 6H 2 O, CaCl 2 , and Na 2 SO 4 in
surface of these solid polymers, the formation of bone- deionized water. The inorganic ion concentrations (mM; Na+
like apatite was also observed after being implanted in 213, K+ 7.5, Mg2+ 2.3, Ca2+ 3.8, Cl− 223, HCO3− 27, HPO42−
1.5, SO42− 0.8) were 1.5 times those of human blood plasma.
bone.19 Highly porous poly(␣-hydroxy acids)/HA
The fluid was buffered at a pH value of 7.4 at 37°C with
composites were developed previously in our lab.5 In tris-(hydroxymethyl) aminomethane [(CH2OH)3CNH2] and
this work, highly porous poly(L-lactic acid) (PLLA) hydrochloric acid (HCl). The solution was metastable and
foams were prepared by solid–liquid phase separation does not precipitate calcium phosphate without external
of the polymer solution and subsequent sublimation stimulation.
of the solvent. The bonelike apatite was grown on the
surfaces of pore walls throughout the PLLA foams in
a simulated body fluid (SBF). The highly porous bio-
degradable polymer/apatite composites were created PLLA incubation in SBF
as a new type of composite scaffold for bone tissue
engineering. Five rectangular polymer foam specimens with dimen-
sions of 12 × 8 × 6 mm were immersed in 100 mL SBF in a
glass bottle maintained at 37°C. A series of brief evacuation–
repressurization cycles was performed to force the solution
MATERIALS AND METHODS into the pores of the foams. Cycling was continued until
there were no air bubbles emerging from the foams. The SBF
PLLA with an inherent viscosity of approximately 1.6 was was renewed every other day. After being incubated for
purchased from Boehringer Ingelheim (Ingelheim, Ger- various periods of time, the specimens were removed from
many). There were no free carboxyl end groups on the the fluid and immersed overnight in 100 mL deionized wa-
PLLA. Dioxane, chloroform, sodium chloride, calcium chlo- ter to remove the soluble inorganic ions. The solid PLLA
ride, sodium hydrogencarbonate, potassium chloride, potas- films (30 mm × 10 mm × 40 ␮m) were treated with the same
sium phosphate, magnesium chloride hexahydrate, sodium process in the SBF (five films in 100 mL SBF).
sulfate, potassium bromide (for IR spectral specimen prepa-
ration), and synthetic HA [3Ca3(PO4)2 ⭈ Ca(OH)2] were pur-
chased from Aldrich (Milwaukee, WI).
Characterizations

Polymer foams and films The porosity was determined with a liquid displacement
method reported in detail earlier.5 Ethanol was chosen as the
displacement liquid because it penetrated easily into the
The polymer foams were prepared by solid–liquid phase pores and did not induce shrinkage or swelling as a nonsol-
separation of polymer solutions and subsequent sublimation vent of the polymers. A foam sample of weight W was im-
of solvent as reported earlier.5 The polymer was dissolved in mersed in a graduated cylinder containing a known volume
dioxane to make a solution of a desired concentration (from (V1) of ethanol. The sample was kept in the ethanol for 5
2.5 to 7.5%). The polymer solution (10 mL) was transferred
min, and then a series of brief evacuation–repressurization
into a beaker (30 mL), then the beaker was rapidly trans-
cycles was conducted to force the ethanol into the pores of
ferred into a refrigerator or a freezer at a present tempera-
the foam. Cycling was continued until no air bubbles
ture to solidify the solvent and induce solid–liquid phase
emerged from the foam. The total volume of ethanol and the
separation. The solidified solution was maintained at that
ethanol-impregnated foam was then recorded as V2. The
temperature for 2 h and then immersed in liquid nitrogen to
volume difference, (V2 − V1), was the volume of the poly-
deep freeze. The frozen mixture was transferred into a
mer/apatite composite skeleton of the foam. The ethanol-
freeze-drying vessel at −5 to −10°C in an ice/salt bath and
impregnated foam was removed from the cylinder, and then
then freeze-dried at 0.5 mmHg for 7 days to completely
the residual ethanol volume was recorded as V3. The quan-
remove the solvent. Ninety-nine percent of the solvent was
tity (V1 − V3), the volume of the ethanol held in the foam,
removed in 1 day, and a constant sample weight was
was determined as the void volume of the foam; thus, the
achieved within 4 days. The foam samples were stored in a
total volume of the foam was
desiccator until incubation or characterization.
Solid PLLA films were cast from a 5% PLLA/chloroform V = (V2 − V1) + (V1 − V3) = V2 − V3.
solution on a glass plate at room temperature. Rectangular
specimens with dimensions of 30 mm × 10 mm × 40 ␮m The density of the foam, d, was expressed as
were obtained. d = W/(V2 − V3),
and the porosity of the foam, ␧, was obtained by

SBF ␧ = (V1 − V3)/(V2 − V3).

The morphology of the incubated polymer foams and
An SBF with a modified formulation of the earlier SBF films was studied by scanning electron microscopy (SEM;
proposed by Kokubo et al.20 was prepared by dissolving S-3200N, Hitachi, Japan) at 15 kV. All SEM micrographs
POROUS POLYMER/APATITE COMPOSITES 287

shown in this article were taken from foams prepared from highly porous and had an open pore structure [Fig.
a 5% PLLA solution. The specimens were cut into halves 1(a)]. A porosity of up to 95% was achieved by this
with a razor blade, and the exposed new surfaces were ob- procedure. The porosity of the foam prepared from a
served with SEM. For microstructural observation, the speci- 5% PLLA solution was 92.7%. An anisotropic tubular
mens were coated with gold using a sputter coater (Desk-II,
morphology with an internal ladderlike structure was
Denton Vacuum Inc.). The gas pressure was less than 50
obtained. The channels were parallel to the direction
mTorr and the current was about 40 mA. The coating time
was 200 s. Energy-dispersive spectroscopy (EDS) was also of solidification (heat transfer direction). Each channel
used to obtain information on the elementary composition of had repeating partitions with uniform spacing per-
the particles grown from SBF incubation. For EDS analysis, pendicular to the solidification direction. The diameter
the specimens were not coated and the environmental mode of the channels and the spacing between repeating
was used. partitions in the channels ranging from several tens of
Fourier transform IR (FTIR) spectra were obtained with a microns to several hundred microns were controlled
Nicolet 5-DX FTIR spectrometer with a resolution of 4 cm−1. by varying the cooling rate and the concentration of
A small amount of powder was scratched from the surface the polymer solution. This characteristic morphology
of an incubated polymer foam or film, then it was milled was attributed to the crystallization of the solvent dur-
with KBr and pressed into a transparent film for IR analysis.
ing solid–liquid phase separation.5
The IR spectrum of the HA was obtained from a commercial
synthetic HA powder from Aldrich.
The PLLA foams were immersed in the SBF at 37°C
The X-ray diffraction (XRD) spectra were obtained with a to grow apatite. After 30 days, a large number of mi-
Rigaku rotating-anode X-ray diffractometer (Rotaflex) at a croparticles with a diameter up to 2 ␮m was formed
2␪ scan rate of 2.5°/min. Cu K␣ ratiation was used for the on the surfaces of the PLLA pore walls (Fig. 1). The
diffraction with a voltage of 40 kV and a current of 100 mA. particles were assembled with small flakelike pieces.
The mass increase of the foams during the apatite forma- Others observed similar morphology of carbonated
tion in the SBF was measured with an analytical balance HA formed on a silica gel surface after incubating in
accurate to 10−4 g. The PLLA foams and the PLLA/apatite SBF.16 The EDS spectrum showed that the main ele-
composite foams were dried in a fume hood for 1 week and ments of the incubated PLLA foam were carbon, oxy-
then vacuum dried at 0.5 mmHg for 24 h before measuring. gen, calcium, and phosphorus [Fig. 2(a)]. Carbon and
The percentage mass increase was normalized with the
oxygen could be from both PLLA and the particles,
foams incubated in a Tris buffer at the same pH (7.4), the
same temperature (37°C), and for the same time intervals
but calcium and phosphorus could only be from the
(control samples). Six specimens were measured for each particles. These results suggested that the particles
sample to obtain the averages and standard deviations. The formed in the PLLA foams might be similar to HA.
apatite particle diameter and number density were obtained Microparticles were also formed on the solid PLLA
from the SEM micrographs of the composite foams. The mi- films treated in SBF under the same conditions (Fig. 3).
crographs were taken from the typical internal surfaces that The particles on the PLLA films were larger than those
were oriented horizontally. The SEM micrographs were di- in the PLLA foams. The surfaces of the films were
vided into equal rectangular areas. For each sample, 10 of completely covered with the microparticles after 15
these areas that were continuous and horizontally oriented days of incubation [Fig. 3(a)]. The morphology of the
were selected to calculate the particle density per unit area formed particles was also a flakelike assembly [Fig.
(number/100 ␮m2). For each sample, 60 particles were
3(b)]. EDS analysis indicated that calcium and phos-
measured to calculate the average diameter and standard
deviation.
phorus were also the main elements in the particles
The compressive mechanical testing was conducted with [Fig. 2(b)].
an Instron 4502 mechanical tester (Instron Co., Canton, MA). Wide-angle XRD was used to characterize the po-
The specimens were circular disks with the same geometry rous composites. The two strong characteristic peaks
as a different type of composite sample used previously (∼16 of HA21 were shown from the apatite powder ob-
mm in diameter and ∼3 mm in thickness).5 The crosshead tained from the surface of PLLA films incubated in the
speed was 0.5 mm/min. The compressive modulus was de- SBF [Fig. 4(a)]. The composite foams [Fig. 4(b, c)]
fined as the initial linear modulus. Six specimens were tested showed the characteristic peaks of apatites [compare
for each sample. The averages and standard deviations were to Fig. 4(a)] and the characteristic peaks of PLLA22
graphed. [compare to Fig. 4(d)]. The amplitudes of apatite peaks
A two-tail Student’s t test (assuming equal variances) was
increased with incubation time in the SBF (in compari-
performed to determine the statistical significance (p < 0.05)
of the differences in particle size, density, mass change, and
son to PLLA peaks).
foam mechanical properties. FTIR spectroscopy was used to gain more informa-
tion on the microparticles formed in the PLLA foams
and on the PLLA films. The spectra of the formed
particles were similar to that of a commercial synthetic
RESULTS HA (Fig. 5). The characteristic absorption bands of
phosphate in HA appearing at 565, 604, 962, and 1085
The PLLA foams prepared from solid–liquid phase cm−1, which reflect the phosphate vibration mode of
separation of the PLLA/dioxane solutions were ␯4, ␯1, and ␯3, respectively,23,24 were observed for all
288 ZHANG AND MA

Figure 1. SEM micrographs of a PLLA foam incubated in SBF for 30 days: original magnifications (a) ×100, (b) ×500, (c)
×2000, and (d) ×10,000.

three samples. The spectra of the formed particles [Fig. band of carbonate.25,26 These carbonate peaks from
5(a, b)] had a strong absorption band at 873 cm−1 cor- particles formed from SBF incubation were much
responding to the ␯2 vibration mode of carbonate. The higher than those in commercial synthetic HA. Hy-
broad peak around 1640 cm−1 was assigned to the ␯3 droxyl stretch was observed at 3570 cm−1 in the spec-
trum of commercial synthetic HA [Fig. 5(c)]. However,
no evident peak at the same wave number was ob-
served for the particles formed from SBF incubation.
The large decrease of the hydroxyl stretch band inten-
sity and the strong carbonate bands of the micropar-
ticles formed from SBF indicated the carbonate substi-
tution for OH in HA.24,26 These results suggested that
the particles in a PLLA foam or on a PLLA film from
SBF incubation were carbonated apatite, which was
similar in composition and structure to the natural
apatite in human and animal hard tissues. The peaks
at 1455 cm−1 (␦ CH3), 1759 cm−1 (␯C=O), and peaks
ranging from 2870 to 3000 cm−1 (␯C−H) in the spectrum
[Fig. 5(a)] were attributed to the PLLA, which could be
scratched with the apatite particles into the KBr film
Figure 2. EDS spectra of microparticles from (a) a PLLA prepared for IR analysis.
foam incubated in SBF for 30 days and (b) a PLLA film The variation of the particle number and size in the
incubated in SBF for 15 days. PLLA foams was achieved by varying the incubation
POROUS POLYMER/APATITE COMPOSITES 289

Figure 3. SEM micrographs of a PLLA film incubated in SBF for 15 days: original magnifications (a) ×2000 and (b) ×10,000.

time in the SBF (Fig. 6). Almost no apatite micropar- 9) than those in the PLLA foams without water treat-
ticles were observed on the surfaces of the PLLA pore ment [Fig. 6(d)] for the same SBF incubation time (189
walls after 3 days of incubation. Scattered and small ± 32 vs. 25.3 ± 6.8 particles/100 ␮m2, p = 1.4 × 10−13).
microparticles were observed after 6 days of incuba- However, the average diameter of the particles
tion. After 15 days of incubation, a large number of formed on the water treated foams was significantly
apatite microparticles with relatively bigger particle smaller (0.273 ± 0.084 vs. 0.837 ± 0.164 ␮m, p = 3.8 ×
size was observed. EDS analysis also demonstrated 10−15).
that the calcium and phosphate contents increased The compressive modulus increased with the incu-
with incubation time (Fig. 7). As a consequence of par- bation time in SBF (Fig. 10). After 30 days or longer,
ticle growth from the nuclei formed at different times, the compressive modulus became significantly higher
there was a wide size distribution. The average par- than that of the initial foam (p < 0.05). In contrast, the
ticle diameter, density (number of particles per unit compressive modulus of the foams incubated in a Tris
surface area), and total apatite mass increased with buffer at the same pH (7.4) and temperature (37°C) did
incubation time (Fig. 8). not change significantly within 60 days of incubation
Some PLLA foam samples were immersed in dis- (Fig. 10).
tilled water at 37°C for 15 days before incubation in
SBF for 15 days. There were significantly more apatite
particles formed in the water treated PLLA foams (Fig.

Figure 4. Wide-angle X-ray diffraction spectra of (a) pow-

der of apatite obtained from the surface of PLLA films in-
cubated in SBF for 30 days, (b) PLLA foam incubated in SBF Figure 5. FTIR spectra of a commercial HA; the apatite
for 30 days, (c) PLLA foam incubated in SBF for 15 days, and particles formed from SBF (a) in a PLLA foam, (b) on a PLLA
(d) PLLA foam. film, or (c) a commercial HA.
290 ZHANG AND MA

Figure 6. SEM micrographs of a PLLA foam incubated in SBF for (original magnification ×2000) (a) 3 days, (b) 6 days, (c)
10 days, or (d) 15 days.

DISCUSSION the solid–liquid phase separation of PLLA/dioxane

solution and subsequent sublimation of the dioxane
A scaffold for tissue engineering should have a high have been used to obtain highly porous PLLA foams.
porosity and an appropriate pore size. Various tech- The foams have an open pore structure. Varying the
niques, such as particulate leaching,6,11 liquid–liquid concentration of the polymer solution and phase sepa-
phase separation,27,28 and others,7,29 have been devel- ration temperature can control the pore size ranging
oped to fabricate highly porous scaffolds. In this work, from several tens of microns to several hundred mi-

Figure 7. EDS spectra of a

PLLA foam incubated in SBF for
(a) 3 days, (b) 6 days, or (c) 15
days.
POROUS POLYMER/APATITE COMPOSITES 291

body on surfaces of various bioactive materials.13–19,30

The bonelike apatite makes the bioactive material fa-
vorable to bond to bone chemically. The mechanism of
the bonelike apatite formation on CaO−SiO2 based ce-
ramic surfaces was proposed as follows31,32: the cal-
cium ions dissolved from the ceramics increased the
ionic activity product of the apatite in the surrounding
body fluid, which was already supersaturated with
respect to apatite; and the hydrated silicon surfaces
provided favorable sites for apatite nucleation. As a
result, a large number of apatite nuclei formed on ce-
ramic surfaces, grew spontaneously, and consumed
the calcium and phosphate ions from the surrounding
fluid. Based on this mechanism, Kokubo et al. used
CaO−SiO2 glass particles as a nucleation-inducing
agent. With the induced nucleation and subsequent
incubation in SBF, apatite was grown on several poly-
mer surfaces, such as polyethylene, poly(ethylene ter-
ephthalate), poly(methyl methacrylate), Nylon-6, and
poly(ether sulfone).33,34
To create better scaffolds for bone tissue engineer-
ing, porous poly(␣-hydroxyl acids)/HA composites
were developed by phase separation of poly(␣-
hydroxyl acids)/HA/solvent mixtures and subse-
quent solvent sublimation in our laboratory.5 To our
knowledge, this work is the first successful growth of
apatite on the inner pore surfaces of a porous material.
The apatite particles formed in the PLLA foams incu-
bated in SBF is similar to the bonelike apatite based on
SEM, EDS, IR, and XRD analyses. For the apatite for-
mation, the environment that can enhance local cal-
cium and phosphate ion concentrations may be desir-
able. In the SBF, PLLA could be partially hydrolyzed
to generate reasonable numbers of COOH and OH
groups. The weak acidic group COOH may dissociate
partially in the aqueous solution of SBF (pH 7.4) at
37°C, leaving the polymer surface negatively charged.
The calcium and phosphate ions may be accumulated
by COO−, COOH, and OH groups through electro-
static force and hydrogen bonding. The local increase
of supersaturation of calcium and phosphate ions
makes the nucleation of apatite possible. The hydro-
lysis of the PLLA is a slow process and the degree of
the hydrolysis is time dependent. With increasing in-
cubation time, the COOH and OH groups increase,
which leads to the increased apatite nucleation rate.
Figure 8. The average apatite particle (a) diameter, (b) den-
In order to test whether the polymer hydrolysis is
sity, and (c) total mass percentage increase over PLLA foam
against incubation time in SBF. important in the apatite nucleation, PLLA foam
samples were immersed in distilled water at 37°C for
15 days before incubation in SBF. The number of apa-
crons.5 A 5% PLLA/dioxane solution has been used in tite particles formed in the water treated PLLA foams
this study to prepare PLLA foams for in situ apatite was significantly larger (189 vs. 29.3/100 ␮m2) while
formation because this foam has been well character- the average particle diameter was significantly smaller
ized previously to ensure consistency.5 However, the (0.273 vs. 0.837 ␮m) than those in the PLA foams with-
polymer concentration (5%) is not a limiting factor to out water treatment for the same SBF incubation time.
this new technology. The hydrated groups such as COOH and OH from the
Bonelike apatite has been reported to form in the PLLA hydrolysis may contribute to the higher apatite
292 ZHANG AND MA

Figure 9. SEM micrographs of a PLLA foam immersed in H2O (at 37°C) for 15 days and then incubated in SBF for 15 days:
original magnifications (a) ×2000 and (b) ×10,000.

nucleation rate in the water treated PLLA foams. The mechanical properties of the foams. The compressive
higher number of initial nucleation sites may compete modulus of the foams incubated in a Tris buffer at the
with each other for the calcium and phosphate ions, same conditions within 60 days had no significant
thus leading to the smaller average size. This result change compared to the initial foam. It seems that the
indicates that the hydrolysis of PLLA may have molecular weight decrease within this time frame was
played an important role during the apatite formation not large enough to cause mechanical property dete-
in the PLLA foams, which corroborates our hypoth- rioration. This was presumably due to the slow hy-
esis. drolysis rate of the hydrophobic PLLA.
The solid PLLA films were completely covered with This work demonstrated how the biomimetic pro-
apatite particles in a relatively short incubation time. cess of in situ apatite formation from an SBF can be
Much larger apatite particles were formed on the used to fabricate porous biodegradable polymer/
PLLA films than in PLLA foams with the same incu- apatite composites. Although the PLLA foams were
bation time, which was presumably due to a much created with a solid–liquid phase separation technique
smaller surface area on the solid films, leading to a in this work, the method could be utilized to grow
higher degree of supersaturation with respect to the apatite for other porous polymeric or nonpolymeric
small number of nucleation sites and a fast apatite materials as long as they have an open pore structure
particle size growth. to allow SBF penetration.
The compressive modulus of the PLLA/apatite
foams was higher than that of the initial PLLA foam.
This indicated that the apatite particles enhanced the
CONCLUSIONS

Highly porous PLLA foams with a ladderlike po-

rous structure were prepared from phase separation
of the polymer solutions and subsequent solvent sub-
limation. New porous PLLA/apatite composites were
created by incubating the polymer foams in the SBF. A
large number of apatite microparticles was formed in
situ on the surfaces of pore walls throughout the PLLA
foams. Based on SEM, EDS, XRD, and FTIR analyses,
the apatite particles were similar to the apatite in natu-
ral bone. These apatite particles were characteristic po-
rous spheres assembled with small flakes. The apatite
particles were formed through a nucleation phase and
a growth phase. The particle number and size were
determined by several factors such as incubation time,
Figure 10. Compressive modulus against incubation time ionic concentration of the SBF, water treatment, and
at 37°C in (⽧) the SBF and (䊏) a Tris buffer at pH 7.4. the polymer surface area. The compressive modulus
POROUS POLYMER/APATITE COMPOSITES 293

increased with incubation time in the SBF. The excel- 17. Kitsugi T, Yamamuro T, Nakamura T, Kokubo T. The bonding
lent bone-bonding property of the bonelike apatite of glass ceramics to bone. Int Orthop (SICOT) 1989;13:199–206.
may provide a good environment for the attachment 18. van Blitterswijk CA, Bakker D, Leenders H, Blink Jvd, Hessel-
ing SC, Bovell YP, Radder AM, Sakkers RJ, Gaillard ML,
and growth of osteoblasts and osteoprogenitor cells. Heinze PH, Beumer GJ. Interfacial reactions leading to bone
bonding with PEO/PBT copolymers (Polyactive). In: Du-
cheyne P, Kokubo T, van Blitterswijk CA, editors. Bone-
References bonding biomaterials. Leiderdrop: Reed Healthcare Commu-
nication; 1992. p 13–30.
1. Langer R, Vacanti J. Tissue engineering. Science 1993;260
19. Li P, Bakker D, van Blitterswijk CA. The bone-bonding poly-
(5110):920–926.
mer Polyactive 80/20 induces hydroxycarbonate apatite for-
2. Vacanti CA, Vacanti JP. Bone and cartilage reconstruction. In:
mation in vitro. J Biomed Mater Res 1997;34:79–86.
Lanza RP, Langer R, Chick WL, editors. Principles of tissue
engineering. Austin, TX: R. G. Landes Company; 1997. p 619– 20. Kokubo T, Kushitani H, Sakka S, Kitsugi T, Yamamuro T. So-
631. lutions able to reproduce in vivo surface-structure changes in
3. Crane G, Ishaug S, Mikos A. Bone tissue engineering. Nat Med bioactive glass-ceramic A-W. J Biomed Mater Res 1990;24:721–
1995;1:1322–1324. 734.
4. Vacanti C, Vacanti J. Bone and cartilage reconstruction with 21. Aoki H. Science and medical applications of hydroxyapatite.
tissue engineering approaches. Otolaryngol Clin North Am Tokyo: Jaas; 1991. p vii, xv, 214.
1994;27(1):263–276. 22. Brizzolara D, Cantow HJ, Diederichs K, Keller E, Domb AJ.
5. Zhang R, Ma PX. Poly(␣-hydroxy acids)/hydroxyapatite po- Mechanism of the stereocomplex formation between enantio-
rous composites for bone tissue engineering. I. Preparation and meric poly(lactide)s. Macromolecules 1996;29:191–197.
morphology. J Biomed Mater Res 1999;44:446–455. 23. Fowler BO. Infrared study of apatite. Inorg Chem 1974;13:194–
6. Ma PX, Langer R. Fabrication of biodegradable polymer foams 214.
for cell transplantation and tissue engineering. In: Yarmush M, 24. Rehman I, Bonfield W. Characterization of hydroxyapatite and
Morgan J, editors. Tissue engineering methods and protocols. carbonated apatite by photo acoustic FTIR spectroscopy. J
Totowa, NJ: Humana Press Inc.; 1998. Mater Sci Mater Med 1997;8:1–4.
7. Ma PX, Langer R. Degradation, structure and properties of
25. El Feki H, Rey C, Vignoles M. Carbonate ions in apatite, infra-
fibrous nonwoven poly(glycolic acid) scaffolds for tissue engi-
red investigation in ␯r CO3 domain. Calcif Tissue Int 1991;49:
neering. In: Mikos AG, Leong KW, Yaszemski MJ, Tamada JA,
269–274.
Radomsky ML, editors. Polymers in medicine and pharmacy.
Pittsburgh, PA: MRS; 1995. p 99–104. 26. Elliott JC, Holcomb DW, Young RA. Infrared determination of
8. Ma PX, Schloo B, Mooney D, Langer R. Development of bio- the degree of substitution of hydroxyl by carbonate ions in
mechanical properties and morphogenesis of in vitro tissue en- human dental enamel. Calcif Tissue Int 1985;37:372–375.
gineered cartilage. J Biomed Mater Res 1995;29:1587–1595. 27. Schugens C, Maquet V, Grandfils C, Jerome R, Teyssie P. Poly-
9. Ma PX, Shin’oka T, Zhou T, Shum–Tim D, Lien J, Vacanti JP, lactide macroporous biodegradable implants for cell transplan-
Mayer J, Langer R. Biodegradable woven/nonwoven compos- tation. II. Preparation of polylactide foams by liquid–liquid
ite scaffolds for pulmonary artery engineering in an juvenile phase separation. J Biomed Mater Res 1996;30:449–461.
lamb model. Trans Soc Biomater 1997;20:295. 28. Lo H, Ponticiello MS, Leong KW. Fabrication of controlled
10. Ma PX, Langer R. Morphology and mechanical function of release biodegradable foams by phase separation. Tissue Eng
long-term in vitro engineered cartilage. J Biomed Mater Res 1995;1:15–28.
1999;44:217–221. 29. Wintermantel E, Mayer J, Blum J, Eckert KL, Luscher P,
11. Mikos AG, Thorsen AJ, Czerwonka LA, Bao Y, Langer R, Win- Mathey M. Tissue engineering scaffolds using superstructures.
slow DN, Vacanti JP. Preparation and characterization of Biomaterials 1996;17:83–91.
poly(L-lactic acid) foams. Polymer 1994;35:1068–1077. 30. Jarcho M, Kay JF, Gumaer KI, Doremus RH, Drobeck HP. Tis-
12. Jarcho M. Retrospective analysis of hydroxyapatite develop- sue, cellular and subcellular events at a bone–ceramic hydrox-
ment for oral implant applications. Dent Clin North Am 1992; ylapatite interface. J Bioeng 1977;1:79–92.
36:19–26.
31. Ohtsuki C, Kokubo T, Yamamuro T. Mechanism of apatite
13. Kokubo T, Shigematsu M, Nagashima Y, Tashiro M, Yama-
formation on CaO−SiO2−P2O5 glasses in a simulated body
muro T, Higashi S. Apatite- and wollastonite-containing glass-
fluid. J Non-Cryst Solids 1992;143:84–92.
ceramics for prosthetic application. Bull Inst Chem Res Kyoto
Univ 1982;60:260–268. 32. Li P, Ohtsuki C, Kokubo T, Nakanishi K, Soga N, Nakamura T,
14. Hench LL. Bioactive ceramics. Ann New York Acad Sci 1988; Yamamuro T. Apatite formation induced by silica gel in a
523:54–71. simulated body fluid. J Am Ceram Soc 1992;75:2094–2097.
15. Kitsugi T, Yamamuro T, Nakamura T, Kokubo T. Bone bond- 33. Kokubo T, Hata K, Nakamura T, Yamamuro T. Apatite forma-
ing behavior of MgO−CaO−SiO2−P2O5−CaF2 glass (mother tion on ceramics, metals and polymers induced by a
glass of AW-glass-ceramics). Biomed Mater Res 1989;23:631– CaO−SiO2-based glass in a simulated body fluid. In: Bonfield
648. W, Hastings GW, Tunner KE, editors. Bioceramics. Guiford,
16. Li P, Nakanishi K, Kokubo T, de Groot K. Induction and mor- UK: Butterworth–Heinemann; 1991. p 113–120.
phology of hydroxyapatite, precipitated from metastable simu- 34. Tanahashi M, Yao T, Kokubo T, Minoda M, Miyamoto T, Na-
lated body fluids on sol-gel prepared silica. Biomaterials 1993; kamura T, Yamamuro T. Apatite coating on organic polymers
14:963–968. by a biomimetic process. J Am Ceram Soc 1994;77:2805–2808.