Effect of Porosity and Pore Size on Microstructures

and Mechanical Properties of Poly-e-Caprolactone-

Hydroxyapatite Composites
Haiying Yu,1,2 Howard W. Matthew,1,3 Paul H. Wooley,1,2 Shang-You Yang1,2
1
Department of Biomedical Engineering, Wayne State University, Detroit, Michigan
2
Department of Orthopaedic Surgery, Wayne State University, Detroit, Michigan
3
Department of Chemical Engineering and Materials Science, Wayne State University, Detroit, Michigan

Received 23 May 2007; revised 3 November 2007; accepted 3 December 2007

Published online 11 March 2008 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.b.31054

Abstract: The influence of variant pore-size and porosity on the microstructure and the
mechanical properties of poly-e-caprolactone (PCL) and hydroxyapatite (HA) composite were
examined for an optimal scaffold in bone tissue engineering. Three various amounts of sodium
chloride (NaCl, as porogens) with two distinct particle size ranges (212–355 lm and 355–600
lm) were blended into PCL and HA mixture, followed by a leaching technique to generate
PCL-HA scaffolds with various pores and porosity. The porosities of the scaffolds were
correlated with the porogen size and concentration. The morphological properties of the
resulting scaffolds were assessed by micro-computerized tomography (lCT), scanning electron
microscopy (SEM), and energy dispersive X-ray analysis (EDX). Extensive PCL-HA pore
interconnections with thinner pore walls were present in scaffolds with higher concentration
(4:1 w/w) and larger particulate of porogen used in the fabrication process. Embedding of HA
particles in the scaffold resulted in rough surfaces on the composites. Instron actuator testing
indicated that the tensile strengths and Young’s moduli of scaffolds were influenced by both
the porosities and pore sizes of the scaffold. It was apparent that increasing the concentration
of porogen compromised the mechanical properties; and a larger porogen particle size led to
increased tensile strength but a reduction in Young’s modulus. Overall, the data indicated that
modification of the concentration and particle size of porogen altered the porous features and
mechanical strength of HA-PCL scaffolds. This provided means to manipulate the properties
of biocompatible cell-supporting scaffolds for use as bone graft substitutes. ' 2008 Wiley
Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater 86B: 541–547, 2008

Keywords: bone substitute; poly-e-caprolactone-hydroxyapatite composite; porous scaffolds;

mechanical properties

INTRODUCTION polyglycolic acid (PGA), polycaprolactone (PCL) and their

Current bone substitutes do not adequately resemble the
physiological and mechanical properties of native bone tis-
sue, which limits their use to repair bone defects caused by
disease, trauma, or surgery. One of the major goals in bone
tissue engineering is to develop an osteoconductive, porous,
and biodegradable scaffold to induce the growth of normal
bone tissue. These grafts must maintain adequate mechani-
cal strength as the bone healing, and degrade at a con-
trolled rate to provide space for the formation of new bone.
Biodegradable polymers such as polylactic acid (PLA),
Correspondence to: S.-Y. Yang (e-mail: syang@wayne.edu)
' 2008 Wiley Periodicals, Inc.
copolymers have been combined with bioceramics such as
hydroxyapatite (HA), tricalcium phosphate (TCP) to pro-
duce materials for bone repair.1–12 The hybridization of
these materials can improve the biocompatibility of each
component with respect to both the structural and biologi-
cal properties. Biodegradable polymers binding with HA or
TCP may overcome the inflexibility and brittleness of hard
ceramic materials. Simultaneously, bioceramic reinforce-
ment can improve the osteoconductivity and degradation
properties of polymers. Biodegradable polymer/bioceramic
composites are therefore promising materials for bone graft
replacement.
PCL, a linear aliphatic polyester, has been approved by
the FDA for utility in medical and drug delivery devices,
and some studies have demonstrated its biocompatibility
541
542 YU ET AL.

and applicability.13,14 PCL exhibits biodegradation in vivo; TABLE I. Weight Ratios of Components in PCL-HA
however, its degradation and resorption are slower due to a Scaffolds and Correspondent Grouping
hydrophobic character and high crystallinity.13 When com- NaCl:(PCL-HA) NaCl HA:PCL
pared with other aliphatic polyesters, PCL as a bone substi- Groups (w/w) Particles (w/w)
tute will allow prolonged mechanical strength retention over A1 4:1 212–355 lm 1/3
time during bone healing process and degrade at the rate A2 1:1 212–355 lm 1/3
coinciding with new bone formation and remodel. A3 1:4 212–355 lm 1/3
Several studies have revealed improved chemical or struc- B1 4:1 355–600 lm 1/3
tural biocompatibility using composite of HA and PCL, or B2 1:1 355–600 lm 1/3
copolymers associated with PCL, for use in bone repair.1,2,6- B3 1:4 355–600 lm 1/3
9,11,15 C 0 1/3
Because of the necessity for cellular influx and growth,
nutrient and metabolite transportation, and blood vessels
ingrowth, certain mechanical properties may need to be com-
promised to achieve the appropriate porosity and scaffold
system (GE Medical Systems, London, Canada) was
pore sizes. However, currently little information is available
employed to quantify the porosity and structure of the scaf-
on the behavior of microstructure and mechanical properties
folds. The scaffolds were scanned at 21 lm isotropic voxel
of PCL-HA scaffolds in response to variations in porosity
size, 400 projections, 400-ms exposure time, 80-KW vol-
and pore sizes as application in bone tissue engineering. In
tages, and 450-lA current. Following acquisition and
this study the porous scaffolds composed of PCL and HA
reconstruction, the image data were analyzed using GEHC
were prepared and evaluated with respect to their microstruc-
MicroView1 software with Analysis1 Version 1.0 (GE
tures and mechanical properties, with the overall aim to de-
Healthcare, London, Canada) to generate isosurfaces of the
velop PCL-HA scaffold as a bone graft substitute for clinical
region of interest (ROI) and the hard tissue fraction volume
applications such as mass bone defect repairs.
(BVF) within the porous scaffolds. The porosity (e) of the
scaffold using microCT values was calculated as e ¼ ð1
MATERIALS AND METHODS ðBVF=ROIÞÞ 3 100%, and compared among groups. In
addition, a traditional calculation based on the dimensions
Preparation of PCL-HA Scaffolds and the mass of the scaffolds was also adapted for the po-
rosity of the scaffolds.17 All specimens were standardized
PCL-HA scaffolds were prepared using a particulate leach- in size of 20 mm 3 10 mm 3 5 mm for the testing. The
ing technique described by Mikos et al.16 PCL (Mn 5 density (q) of the scaffolds was calculated using the for-
80,000, Sigma-Aldrich, USA) was dissolved in tetrahydro- mula of q ¼ w=V where w is the scaffolds weight after
furan (THF) (Sigma-Aldrich, USA) at 1:10 wt/vol at 408C leaching and dry process, and V is the volume of scaffold.
for 12 h. Particulate NaCl (Sigma-Aldrich, USA) was sepa- The porosity (e) of scaffolds was then calculated as:
rated into two particle diameter ranges of 212–355 lm and e ¼ ð1  ðq=qc ÞÞ 3 100% where qc is the density of the
355–600 lm using American Society for Testing and Mate- scaffolds from group C.
rials (ASTM)-standard brass sieves (Fisher, USA). Differ-
ent quantities of sieved NaCl were suspended with HA
particles (
40 lm particle size) into PCL solutions by son- Morphology of PCL-HA Scaffolds
ication (60 s) to form a highly viscous slurry. The mixed The PCL-HA scaffolds were sputter coated with gold and
suspension was then poured into flat-bottom glass dishes to imaged using a scanning electron microscope (SEM, Hita-
a depth of 5 mm, and dried at room temperature for 2 chi S-2400, Japan) at 15 kV. In addition, the GE MicroCT
days. After evaporation of the solvent, the scaffold was system and its analysis software (GEHC MicroView) were
washed in excess distilled water to leach out the NaCl. The employed to calculate the mean thickness of pore walls
scaffold was then dried at room temperature, and the (Pw.Th) of the samples in different groups. To evaluate the
weight of the scaffold was recorded. Table I summarized interconnectivity of the porous scaffolds, the Euler numbers
the experimental groups tested with respect to both the (EuN) were also calculated as EuN/ROI volume (mm3)18–20
scaffold sizes of NaCl particles (groups A using the 212– using the same MicroView program. Energy dispersion
355 lm NaCl particles while groups B using the 355– X-ray (EDX) analysis was also conducted in association
600 lm NaCl particles), and the concentrations of NaCl with SEM to confirm the existence of HA particles within
added to HA-PCL composite (weight ratio from 4:1, 1:1 to the composite scaffolds. Although the gold used to sputter
1:4 of NaCl to PCL-HA). HA-PCL material without the ad- the coating generated two small peaks closed to 10 KeV,
dition of any NaCl porogen was used as a control (Group C). the X-ray emission invoked from calcium (Ca) and phos-
phorus (P) were distinctive and not impeded. The atomic
Determining Porosity of PCL-HA Scaffolds
percentages of calcium and phosphorus were calculated and
The porosity of the scaffolds was determined using two compared with the theoretical atomic ratio of calcium and
separated methods. An eXplore RS80 Laboratory MicroCT phosphorus in HA (Ca10(PO4)6(OH)2).

Journal of Biomedical Materials Research Part B: Applied Biomaterials

 POLY-e-CAPROLACTONE-HYDROXYAPATITE COMPOSITES
Mechanical Testing
To examine the mechanical properties of the porous scaf-
folds, an Instron (Model 8841, Universal Materials Testing
Machine, USA) was employed to compare the tensile pro-
perties among samples with a 250 N static load cell at a
crosshead speed of 2 mm/min. The scaffolds were cut into
strips of 20 mm length, 10 mm width, and 5 mm thickness
for the test. After the stress–strain curves were obtained,
the Young’s moduli were calculated from the slope at the
initial linear stage of stress–strain curve (strain range, \3%
strain), and tensile strength was determined from the maxi-
mum load.
Statistical Analysis
Five specimens per group were measured during porosity
calculation, morphological and mechanical experiment. For
microCT assessment, at least 30 measurements per sample
were averaged to avoid bias. Data were expressed as means
6 standard deviations. Statistical analysis was performed
using SPSS version 10.0 (SPSS, Chicago, IL). Data com-
parison between groups A and B at identical porogen con-
centration were analyzed using Student t test, while the
comparison within groups A or B with different concentra-
tion of porogen were performed using one-way ANOVA
with the LSD post hoc multiple comparisons. In all statisti-
cal evaluations, p \ 0.05 was considered as statistically
significant.
RESULTS
Porosities
The porosity of the PCL-HA scaffolds was highly depend-
ant upon the ratio and size of NaCl to total weight of com-
posite materials (Figure 1). Scaffold porosity calculation
based on MicroCT hard tissue fraction volume over total
volume (ROI) indicated that higher concentrations of NaCl
resulted in higher porosities in each group (p \ 0.001), and
porosities up to 75–85% were observed in groups fabri-
cated with 4:1 porogen over scaffold composite (groups A1
and B1). Within a constant NaCl concentration, samples
with larger NaCl particles exhibited significantly higher
porosities than samples employing the smaller particles
(groups B1 vs. A1, B2 vs. A2, and A3 vs. B3, p \ 0.01).
The porosity calculation based on the dimensions and the
mass of the scaffolds resembled the findings (data not
shown), illustrating clear positive correlation of porosity
with the concentration and sizes of porogen.
Morphology
The PCL-HA scaffolds exhibited multiporous morphology
due to the leaching of NaCl particles (Figure 2). The pore
size corresponded well with the size of porogen particles
used. As the NaCl concentration increased, a corresponding
Journal of Biomedical Materials Research Part B: Applied Biomaterials
543
Figure 1. The porosities of the PCL-HA composites fabricated with
two distinct sized porogen particles at various concentrations. Data
collection and analysis were based on GE MicroCT system (*p \
0.01).
increase in the number of pores and the interconnections
between pores was observed. However, clear interconnec-
tions between pores in the PCL-HA composite could only
be detected using the highest concentration of NaCl [Figure
2(A,D), groups A1 and B1], where the larger NaCl par-
ticles (group B1) resulted in an increased number of pores
and wider interconnection openings when compared with
the smaller NaCl particles (group A1). Lower concentra-
tions of NaCl resulted in more isolated pores and sporadic
interconnections [Figure 2(B,C,E,F)]. Control composites
lacking NaCl resulted in solid PCL-HA composites without
pores and interconnections [Figure 2(G)]. With the
microCT isosurface reconstruction technique, 3D PCL-HA
scaffolds with 4:1 porogen ratio, and porogen size 212–355
lm [Figure 2(H), group A1] and 355–600 lm [Figure 2(I),
group B1] were exemplified. Extensive interconnections in
the two samples were visualized.
MicroCT analysis also quantified the mean thickness of
the pore walls within the scaffolds. As gradient increase in
concentration of porogen, the mean thickness of pore walls
decreased significantly (p \ 0.001) (Table II, the compari-
son within groups A or B). With identical porogen concen-
tration, the larger particulate size (355–600 lm, groups B)
generated significantly thinner pore walls in PCL-HA scaf-
folds than the smaller ones (212–355 lm, groups A) did
(p \ 0.001) (Table II, the comparison across groups A and
B), except for comparison between group B1 and group
A1. The interconnectivity of the pores within the porous
scaffolds (denoted as the EuN/ROI volumes) was positively
correlated with the porogen ratios (p \ 0.001, Table II).
Decreasing the ratio of porogen resulted in the correlating
change of EuN/ROI volumes from positive to negative,
indicating decreased pore connectivity and increased
connected solid domain. With same porogen ratio, the
influence of NaCl sizes on EuN/ROI volume was not sig-
nificant except for the pair of groups A2 and B2 (p \ 0.05,
Table II).
544 YU ET AL.

Figure 2. The morphological characteristics of PCL-HA scaffolds examined by Scanning Electron

Microscopy. Panels A–C were fabricated with 212–355 lm porogens particles and Panels D–F with
355–600 lm particles. The inserts represent high magnification (10003) of the main images (magni-
fication 603). Panels A and D are the scaffolds with highest NaCl concentration (4:1, w/w); B and E
example the composites with middle concentration of the porogen (1:1, w/w); and Panels C and F
illustrate the ones with lowest concentration of porogen (0.25:1, w/w). Panel G shows a PCL-HA
composite without NaCl process during the fabrication. Panel H and I is examples of 3D MicroCT
isosurface image of the PCL-HA scaffold with 4:1 porogen ratio, and porogen size 212–355 lm
(panel H, group A1) or 355–600 lm (panel I, group B1), respectively. MicroView software was used
on the CT images to calculate the porosity and pore-wall thickness of the scaffolds (see Materials
and Methods for details).

At high magnification of SEM, partially embedded HA
particles protruded out of the surfaces of pore walls and net-
works within PCL-HA scaffolds, resulted in the rough
appearance on the pore walls. The mean size of lumpy HA
particles were less than 20 lm, in comparison with the 40-lm
dimensions when examined before combination process. The
groups of PCL-HA composite fabricated with higher concen-
tration of the porogen revealed increased surface roughness,
defined as more HA particles exposed on the surface [Figure
2(A,D), inserts]. In contrast, smoother surfaces appeared pro-
portionate to the decrease of the NaCl ratio [Figure 2(C,F),
inserts]. The protruded HA components were confirmed by
EDX spectrum (Figure 3). The calcium and phosphorus peaks
were prominent, and quantified as atom percentages of Ca
and P (Ca/P 5 68.50/38.51) (Figure 3, inset). The similarity
of the percentages to the theoretical fraction (Ca/P 5 62.5/
37.5 or 10/6) in HA (Ca10(PO4)6(OH)2) suggested that the
intact HA particles incorporated into the composites.

TABLE II. Pore-Wall Thickness (Pw.Th) and Euler Number (EuN) of the Composite Scaffolds Quantified Using a GE MicroCT System

Group A Group B
NaCl Ratio Pw.Th EuN/ROI Pw.Th EuN/ROI
(w/w) (mm) Volume (mm3) (mm) Volume (mm3) p (A vs. B)
1:4 1.146 6 0.750* 29.756 6 7.965 0.690 6 0.274* 210.076 6 8.645 *p \ 0.001
1:1 0.187 6 0.049* 5.089 6 2.887y 0.164 6 0.031* 11.065 6 6.423y *yp \ 0.05
4:1 0.100 6 0.018* 34.856 6 9.972 0.108 6 0.017* 35.945 6 13.587 *p \ 0.05
p (in group) p \ 0.001 p \ 0.01 p \ 0.001 p \ 0.01
*Pw.Th comparisons between Groups A and B.
y
EuN/ROI comparisons between Groups A and B.

Journal of Biomedical Materials Research Part B: Applied Biomaterials

 POLY-e-CAPROLACTONE-HYDROXYAPATITE COMPOSITES
Figure 3. EDX spectrum showed the elements existing on the PCL-
HA composites with predominant calcium and phosphorus peaks.
The insert shows the quantification of the P and Ca atoms in the
composites. [Color figure can be viewed in the online issue, which
is available at www.interscience.wiley.com.]
Mechanical Properties
To observe the mechanical properties of the PCL-HA scaf-
folds, a tensile load was applied to each sample. Constant
crosshead speed was exerted until final failure occurred.
Solid PCL-HA composites were also tested for comparison.
The plot of stress–strain curves indicates that deformation
of PCL-HA scaffolds occurred through three distinct
regions: a linear-elastic region followed by a plateau of
roughly constant stress which leaded into a final region of
decreasing stress (Figure 4). Within the groups using the
same size of porogen (groups A or B), increasing the NaCl
concentration reduced both tensile strength and Young’s
modulus (Figure 5): A1 \ A2 \ A3 \ C and B1 \ B2 \
B3 \ C (p \ 0.001). The solid PCL-HA composites (group
C) showed the highest strength without exception. Under
the conditions of the same porogen concentration, the parti-
cle size of NaCl markedly affected the tensile strength and
Young’s modulus. Young’s modulus behaved in a reverse
manner to tensile strength as porogen size was varied. With
the exception of the comparison between group A1 and B1,
the larger NaCl particles (groups B2 or B3) improved the
tensile strength (p \ 0.05) but reduced the Young’s modu-
lus (p \ 0.05) as opposed to the smaller particles at same
NaCl ratio (groups A2 or A3) (Figure 5).
DISCUSSION
This study evaluates the influence of porous features on the
mechanical properties of PCL-HA composite scaffold pro-
posed for use in bone tissue engineering. Poly-e-caprolac-
tone (PCL) and HA are nontoxic to cells and commonly
used in clinics. But major drawbacks on mechanical fea-
tures limit the application of both types of materials as
scaffolds for engineered bone. For instances, a reduction in
mechanical strength as degradation leads to distortion and
cracking of PCL structure.21 Because of the extreme stiff-
ness and brittleness as well as the low flexibility and mold-
ability,8 a porous HA scaffold cannot sustain the
mechanical loading for the larger bone repairing. By effec-
tively controlling the volume fractions and global arrange-
ment of each component, a PCL-HA composite have been
attempted to improve the mechanical strengths to suit the
physiological conditions of the repaired bone.
Journal of Biomedical Materials Research Part B: Applied Biomaterials
545
In this study, we have examined the correlation of
porosity, mean thickness of pore wall, connectivity of pores
and walls, and mechanical strength of the PCL-HA compo-
sites as engineered-bone scaffold. A high porosity (large
surface area/volume ratio) is thought to facilitate cell
attachment, proliferation, and subsequent matrix deposition.
The porosity of PCL-HA scaffolds generated was strictly
dependant upon the proportion and dimension of NaCl used
in the manufacturing process. The highest weight ratio of
NaCl to the PCL-HA (4:1 w/w) generated the composite
material with the highest porosity of 84.32% 6 5.68%
(group B1). In addition, higher porosities were obtained
using larger NaCl particles (355–600 lm) at the same
concentration of NaCl. These differences were achieved
possibly because the higher concentrations and larger par-
ticulates of NaCl contributed to increasing interconnections
between pores and NaCl leaching. EDX spectrum showed
NaCl residue in the scaffolds fabricated with lower concen-
tration and smaller particle sizes of NaCl (data not shown),
suggesting incomplete NaCl leaching due to the deeper
entrapment in the PCL-HA structures.
Although controversy exists in the literature regarding
optimal pore sizes for tissue engineered scaffolds, a pore
size of 200–600 lm is acceptable in most studies to pro-
vide adequate space for osteoblasts and vascular tissue
ingrowth.22–25 More importantly, the ingrowth of the osteo-
blasts and neovascularization into a scaffold can be attrib-
uted to the width of interconnection between pores. As an
important microarchitecture parameter for tissue engineer-
ing scaffolds, the connectivity of pore and wall were eval-
uated quantificationally by computing the Euler number on
3D microCT images. After normalized to the ROI volume,
the higher EuN/ROI volumes account for the higher level
of connectivity of pores; while lower values for more con-
nectivity of solid scaffold network.26,27 The positive EuN/
ROI volumes in groups fabricated with higher amount of
NaCl porogens indicated more pores linked together, which
Figure 4. The typical strain-stress curves generated by PCL-HA
composites under tensile load at a crosshead speed of 2 mm/min.
The values of the Young’s modulus were determined by the slope of
a line drawn tangential to the low-strain portion (3%) of the curve.
Groups A1–A3 were samples with smaller porogen while groups
B1–B3 were with larger NaCl particles.
546 YU ET AL.
Figure 5. Summary of mechanical properties of PCL-HA composite with variety of porogen con-
centration and particles size (*p \ 0.05).
should favor cell ingrowth and angiogenesis inside of scaf-
fold. However, extreme high EuN/ROI volume implies
detached scaffold network and intolerance to mechanical
loading. The negative EuN/ROI volume in group A3 and
B3 represented the broad connected scaffold, which ration-
alizes the enhanced mechanical properties, but also sug-
gests the presence of isolated pores. SEM assessment also
suggested that only the highest concentration of NaCl
(group A1 and B1) resulted in sufficient pore interconnec-
tions, while the lower concentrations of NaCl resulted in
few interconnections.
Surface microtopography or texture has been shown to
affect cell and biomaterial interaction. An increase in cell
number and attachment force has been reported on textured
polymer substrates. A higher percentage of osteoblasts
were inclined to attach to the rougher surface, synthesize
extracellular matrix and mineralize.25,28,29 In this study,
rough surfaces due to the presence of HA particles were
apparent on pore walls of PCL-HA composites. These HA
particles were closely bonded to the PCL, in both clusters
and single HA particles. Interestingly, exposed HA particles
were more predominant on high porosity scaffolds (groups
A1 and B1) than on low porosity scaffolds. The gradient
differences on the microtopography indicated that the den-
sity of exposed HA particles on PCL-HA scaffold surface
may be controlled by adjusting the NaCl concentration to
some extent. Thus the optimal surface microtopography
and energy could be achieved by manipulating the appro-
priate porogen ratio in order to optimize the cell growth
behavior.
The scaffold is designed as bone graft substitute to
replace the trabecular bone. Besides the sufficient porosity
to accommodate cell and vascular ingrowth, the scaffolds
should also aim at the mechanical strength of the trabecular
bones. Whereas the strength and modulus of trabecular
bone intently correlates with its density and direction which
varies over a wide range, the midrange value of trabecular
bone strength (5–10 MPa) and Young’s modulus (50–100
MPa)30 may serve as criteria. Under the circumstance with-
out cell involvement and biological mineralization, the ten-
sile strength of group B2, A3, and B3 matched the
trabecular bone strength, while only Young’s modulus of
group A3 falls into the target range. The mechanical prop-
erties of PCL-HA scaffolds changed considerably during
alterations in the scaffold pore sizes and porosities. Using
equivalent NaCl particle sizes, increasing the porogen con-
centration led to increased void space or porosity, thinner
walls and disconnected solid network, which significantly
weakened tensile strengths and Young’s moduli of the
scaffolds. Using equivalent NaCl ratios to PCL-HA, the
differences in porogen sizes resulted in variable network
morphologies, characterized by distinction in the thickness
of pore walls (Table II). Compared with the groups A2 and
A3 (porogen size 212–355 lm) respectively, groups B2
and B3 (porogen size 355–600 lm) showed significantly
low Young’s moduli and high tensile strengths, possibly
due to the distinguishing thickness of pore wall, the mean
thickness in groups B2 and B3 being markedly thinner than
those in groups A2 and A3 (p \ 0.001). The difference in
the wall thickness influenced the amount of HA involved
in the single PCL pore-wall, and in turn these HA was
likely sufficient to impact the mechanical property of this
wall locally by interrupting the continuity and plastic effect
of PCL. Resultantly the global effects ensued, evidenced
by the distinct mechanical behaviors including Young’s
moduli and tensile strengths. These rationale were consist-
ent with the previous finding that increasing the HA con-
tent provided higher moduli and brittleness to composites,
while plasticity and the percent elongation of composites
was decreased.1,10,31
Overall, the study indicated that varying the size and
ratio of porogen effectively altered the microarchitecture,
microtopography, and mechanical properties of HA-PCL
composites. Raising the size and concentration of NaCl
Journal of Biomedical Materials Research Part B: Applied Biomaterials
 POLY-e-CAPROLACTONE-HYDROXYAPATITE COMPOSITES
porogen positively correlated the porosity, pore intercon-
nection, and microtopography of the scaffolds that benefits
cell ingrowth and angiogenesis, and negatively associated
with the material mechanical strengths as bone graft substi-
tutes. Although the dilemma on proper PCL-HA scaffolds
remained unsolved in this study, other tools to balance the
advantage of porosity and the preservation of reasonable
mechanical strength are in our advanced studies, for
instance, accelerating osteoblasts proliferation and minerali-
zation via prevascularization in these PCL-HA scaffolds.
Thus the current study laid important foundation and one
step further to optimize composite scaffold for repairing
mass bone defects.
REFERENCES
1. Calandrelli L, Immirzi B, Malinconico M, Luessenheide S,
Passaro I, Pasquale RD, Oliva A. Natural and synthetic hy-
droxyapatite filled PCL: Mechanical properties and biocom-
patibility analysis. J Bioact Compat Polym 2004;19:301–313.
2. Dunn AS, Campbell PG, Marra KG. The influence of polymer
blend composition on the degradation of polymer/hydroxyapa-
tite biomaterials. J Mater Sci Mater Med 2001;12:673–677.
3. Ignatius AA, Betz O, Augat P, Claes LE. In vivo investiga-
tions on composites made of resorbable ceramics and poly
(lactide) used as bone graft substitutes. J Biomed Mater Res
2001;58:701–709.
4. Kikuchi M, Cho S-B, Suetsugu Y, Tanaka J. In vitro tests and
in vivo tests developed TCP/CPLA composites. Bioceramics
1997;10:407–410.
5. Kikuchi M, Koyama Y, Yamada T, Imamura Y, Okada T,
Noriaki S, Akita K, Takakuda K, Junzo T. Development of
guided bone regeneration membrane composed of b-trical-
cium phosphate and poly(L-lactide-co-glycolide-co-e-caprolac-
tone) composites. Biomaterials 2004;25:5979–5986.
6. Kim H-W, Knowles JC, Kim H-E. Development of hydroxy-
apatite bone scaffold for controlled drug release via poly (e-
caprolactone) and hydroxyapatite hybrid coatings. J Biomed
Mater Res 2004;70B:240–249.
7. Kim H-W, Knowles JC, Kim H-E. Hydroxyapatite/poly(e-cap-
rolactone) composite coatings on hydroxyapatite porous bone
scaffold for drug delivery. Biomaterials 2004;25:1279–1287.
8. Kim H-W, Lee E-J, Kim H-E, Salih V, Knowles JC. Effect of
fluoridation of hydroxyapatite in hydroxyapatite-polycaprolac-
tone composites on osteoblast activity. Biomaterials 2005;26:
4395–4404.
9. Marra KG, Szem JW, Kumta PN, DiMilla PA, Weiss LE. In
vitro analysis of biodegradable polymer blend/hydroxyapatite
composites for bone tissue engineering. J Biomed Mater Res
1999;47:324–335.
10. Ramakrishna S, Mayer J, Wintermantel E, Leong KW. Bio-
medical applications of polymer-composite materials: A
review. Compos Sci Technol 2001;61:1189–1224.
11. Rizzi SC, Heath DJ, Coombes AGA, Bock N, Textor M,
Downes S. Biodegradable polymer/hydroxyapatite composites:
Surface analysis and initial attachment of human osteoblasts.
J Biomed Mater Res 2001;55:475–486.
12. Verheyen CCPM, de-Wijin JR, van-Blitterswijk CA, de-Groot
K. Evaluation of hydroxyapatite/poly (L-lactide) composites:
Mechanical behavior. J Biomed Mater Res 1992;26:1277–
1296.
13. Kweon HY, Yoo MK, Park IK, Kim TH, Lee HC, Lee H-S,
Oh J-S, Akaike T, Cho C-S. A novel degradable polycapro-
Journal of Biomedical Materials Research Part B: Applied Biomaterials
547
lactone networks for tissue engineering. Biomaterials 2003;24:
801–808.
14. Pitt CG. Poly-e-caprolactone and its copolymers. In: Chasin M,
Langer R, editors. Biodegradable Polymers as Drug Delivery
Systems. New York: Maicel Dekker; 1990. pp 71–120.
15. Ciapetti G, Ambrosio L, Savarino L, Granchi D, Cenni E,
Baldini N, Pagani S, Guizzardi S, Causa F, Giunti A. Osteo-
blast growth and function in porous poly e-caprolactone matri-
ces for bone repair: A preliminary study. Biomaterials 2003;24:
3815–3824.
16. Mikos AG, Sarakinos G, Vacanti JP, Langer RS, Cima LG.
Biocompatible polymer membranes and methods of prepara-
tion of three dimensional membrane structures. US patent,
5,514,378; 1996.
17. Rai B, Teoh SH, Ho KH, Hutmacher DW, Cao T, Chen F,
Yacob K. The effect of rhBMP-2 on canine osteoblasts seeded
onto 3D bioactive polycaprolactone scaffolds. Biomaterials 2004;
25:5499–5506.
18. Odgaard A, Gundersen HJG. Quantification of connectivity in
cancellous bone, with special emphasis on 3-D reconstructions.
Bone 1993;14:173–182.
19. Peyrin F, Mastrogiacomo M, Cancedda R, Martinetti R. SEM
and 3D synchrotron radiation micro-tomography in the study
of bioceramic scaffolds for tissue-engineering applications.
Biotechnol Bioeng 2007;97:638–648.
20. Filmon R, Retailleau-Gaborit N, Grizon F, Galloyer M, Cincu
C, Basle MF, Chappard D. Non-connected versus intercon-
nected macroporosity in poly(2-hydroxyethyl methacrylate)
polymers. An X-ray microtomographic and histomorphometric
study. J Biomater Sci Polym Ed 2002;13(10):1105–1117.
21. Griffith LG. Polymeric biomaterials. Acta Mater 2000;48:
263–277.
22. Dennis JE, Haynesworth SE, Young RG, Caplan AI. Osteo-
genesis in marrow-derived mesenchymal cell porous ceramic
composites transplanted subcutaneously: Effect of fibronectin
and laminin on cell retention and rate of osteogenic expres-
sion. Cell Transplant 1992;1:23–32.
23. Ziegler J, Mayr-Wohlfart U, Kessler S, Breitig D, Günther K-
P. Adsorption and release properties of growth factors from
biodegradable implants. J Biomed Mater Res 2002;59:422–
448.
24. Bucholz RW, Carlton A, Holmes R. Interporous hydroxyapa-
tite as a bone-graft substitute in tibial plateau fractures. Clin
Orthop 1989;240:53–62.
25. Boyan BD, Hummert TW, Dean DD, Schwartz Z. Role of mate-
rial surfaces in regulating bone and cartilage cell response. Bio-
moterials 1996;17:137–146.
26. Boutry N, Cortet B, Dubois P, Marchandise X, Cotten A. Tra-
becular bone structure of the calcaneus: Preliminary in vivo
MR imaging assessment in men with osteoporosis. Radiology
2003;227:708–717.
27. Hilfer R, Manwart C. Permeability and conductivity for
reconstruction models of porous media. Phys Rev E 2001;64:
021304.
28. Bowers KT, Keller JC, Randolph BA, Wick DG, Michaels
CM. Optimization of surface micromorphology for enhanced
osteoblast responses in vitro. Int J Oral Maxillofac Implants
1992;7:302–310.
29. Groessner-Schreiber B, Tuan RS. Enhanced extracellular ma-
trix production and mineralization by osteoblasts cultured on
titanium surfaces in vitro. J Cell Sci 1992;101:209–217.
30. Yaszemski MJ, Payne RG, Hayes WC, Langer R, Mikos AG.
Evolution of bone transplantation: Molecular, cellular, and tis-
sue strategies to engineer human bone. Biomaterials 1996;17:
175–185.
31. Ural E, Kesenci K, Fambri L, Migliaresi C, Piskin E. Poly
(D,L-cactide/e-caprolactone)/hydroxyapatite composites. Bio-
materials 2000;21:2147–2154.