ARTICLE IN PRESS

Biomaterials 26 (2005) 953–960

Increased osteoblast functions on theta+delta nanofiber alumina

Thomas J. Webstera,b,*, Elaine L. Hellenmeyera, Rachel L. Pricea
a
Department of Biomedical Engineering, Purdue University, Potter Building, West Lafayette, IN 47907-1296, USA
b
School of Materials Engineering, Purdue University, Potter Building, West Lafayette, IN 47907-1296, USA
Received 31 December 2003; accepted 30 March 2004
Available online 19 June 2004

Abstract

Nanophase materials, or materials with grain sizes less than 100 nm in at least one direction, are promising materials for various
implant applications since our tissues are composed of nanometer components (i.e., proteins and/or inorganics). Specifically, bone is
comprised of nanostructured hydroxyapatite and collagen fibers which continuously provide an extracellular matrix surface to bone-
forming cells (osteoblasts) with a high degree of nanometer roughness. Despite this fact, materials currently utilized for orthopedic
implants, whether metallic or ceramic, have constituent grain sizes in the non-biologically inspired micron regime. For this reason,
the objective of the present in vitro study was to determine osteoblast functions on one classification of nanomaterials for orthopedic
applications: nanofiber alumina. Various crystalline forms of nanofiber alumina were tested in this study. To obtained different
crystalline structured nanofiber alumina, boehmite nanofiber alumina was sintered at either 400
C, 600
C, 800
C, 1000
C, or
1200
C for 2 h in air. X-ray diffraction results provided evidence that boehmite nanofiber alumina remained boehmite when sintered
at 400
C but changed crystalline phases to gamma, gamma+delta, theta+delta, and alpha when sintered at 600
C, 800
C, 1000
C,
and 1200
C, respectively. Moreover, compared to any other alumina formulation tested in this study, osteoblast functions (as
measured by alkaline phosphatase activity and calcium deposition) were the greatest on theta+delta crystalline phase nanofiber
alumina after 14 days of culture. Boehmite had the next greatest amount of calcium deposition by osteoblasts followed by
gamma+delta. Gamma crystalline phase then followed and was greater than alpha crystalline phase nanofiber alumina which
promoted osteoblast functions the least of all the compacts with the exception of borosilicate glass (reference substrate). For this
reason, this study suggests that theta+delta nanofiber alumina should be further investigated in orthopedic applications.
r 2004 Elsevier Ltd. All rights reserved.

Keywords: Alumina; Nanophase; Nanofiber; Nanotechnology; Osteoblasts; Orthopedic

1. Introduction engineering applications, to date, relatively few advan-

Nanotechnology can be broadly defined as the use of
materials whose components exhibit novel and signifi-
cantly changed properties when control is gained at the
atomic, molecular, and supramolecular levels [1].
Several critical research fields (such as for processing,
catalytic, optical, actuation, electrical, mechanical, etc.)
have already started to benefit from new technological
advancements in the area of nanotechnology, particu-
larly through the use of nanomaterials [1–9]. Although
showing promise for these traditional science and
*Corresponding author. Department of Biomedical Engineering,
Purdue University, Potter Building, West Lafayette, IN 47907-1296,
USA. Tel.: +1-765-494-2995; fax: +1-765-494-1193.
E-mail address: twebster@ecn.purdue.edu (T.J. Webster).
tages of nanophase materials have been described for
biological applications (specifically, those involving
bone implant applications). Yet, since nanotechnology
embraces a system whose core of materials is in the
range of nanometers (109 m), there are many simila-
rities between nanophase materials and components of
biological organs [10].
For example, living systems are clearly governed by
molecular behavior at nanometer scales [10]. The
molecular building blocks of life (i.e., proteins, nucleic
acids, lipids, carbohydrates, etc.) are all examples of
materials that possess unique properties determined by
the size, folding, and patterns at the nanoscale.
Specifically for bone, hydroxyapatite (the major inor-
ganic component of bone) is between 2 and 5 nm in
width and 50 nm in length while Type I collagen (the

0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2004.03.040
 ARTICLE IN PRESS
954 T.J. Webster et al. / Biomaterials 26 (2005) 953–960
major organic component of bone) is a triple helix
300 nm in length, 0.5 nm in width, and has a periodicity
of 67 nm [11]. It is because of this that it stands to reason
that bone cells are naturally accustomed to interacting
with surfaces with a large degree of nanometer rough-
ness. This is in contrast to the surfaces provided by
traditional metallic or ceramic implants which have
constituent particles sizes that provide non-biologically
inspired topographies rough at the micron scale yet
smooth at the nanoscale [11].
One such interesting nanomaterial formulation, alu-
mina nanofibers, hold much promise for orthopedic
applications [12–18]. Interest in alumina nanofibers has
been growing exponentially due to their unique cataly-
tic, mechanical, and surface properties [12–18]. Despite
this promise, few studies have elucidated interactions of
living cells pertinent to orthopedic prostheses with
alumina nanofibers. For this reason, the objective of
the present in vitro study was to determine osteoblast
functions (specifically, adhesion, alkaline phosphatase
activity and ability to deposit calcium) on alumina
nanofibers of various crystalline phases.
2. Materials and methods
2.1. Materials
Alumina nanofibers with dimensions approximately
2 nm in diameter and greater than 50 nm in length were
obtained from Argonide, Corporation (Sanford, FL),
who synthesized such fibers using proprietary sol–gel
methods, followed by aging, filtration, and subsequent
drying and heat treatment at 400
C [18]. Upon receipt,
alumina nanofiber powders were compacted serially in a
steel-tool die via a uniaxial pressing cycle (2–6 tons over
a 7 min period) at room temperature. Some of the
resulting compacts were then sintered in air at either
400
C, 600
C, 800
C, 1000
C, or 1200
C (10
C/min
ramp to the final temperature which was held for
120 min). Resulting compacts were approximately 1–
1.5 mm thick and 1.25 cm in diameter. Borosilicate glass
coverslips (Fisher) etched in 1 n NaOH were used as
reference substrates since there is extensive research on
osteoblast interactions with etch glass coverslips [14,15].
All substrates were autoclaved for sterilization purposes
at 250
C for 30 min.
2.2. Material characterization
The materials of interest to the present study were
characterized for crystalline phase, chemistry, and
surface roughness according to standard techniques
[14]. For crystalline phase analysis, X-ray diffraction
(XRD) was utilized. XRD graphs were obtained using a
Siemens Kristalloflex Diffractometer with a copper Ka
radiation at 40 kV and 25 mA. Scanning was recorded
on a KEVEX detector between 20
(2y) and 80
(2y)
with a 10
/min. scan speed and standard JCPDS XRD
patterns (from Diffrac Plus XRD Commander Software
by Bruker Advanced X-ray Solutions) were used to
confirm phases present in nanofiber alumina compacts.
For chemical analysis, electron spectroscopy for
chemical analysis (ESCA) was performed on all
nanofiber alumina compacts using a Surface Science
Instruments (SSI) X-Probe instrument. An aluminum
Ka1;2 monochromatized X-ray source was used to
stimulate photoemissison of the inner shell electrons of
the sample. The energy of this electron was then
recorded and analyzed for information concerning
chemical composition of the alumina samples. Wide
scans of the sample were used to generate low-resolution
spectra that identified and quantified the percentages of
different elements on the surface of the materials.
For qualitative surface roughness analysis, scanning
electron microscopy (SEM) was performed on all
nanofiber alumina compacts. Compacts were first
sputter-coated with a thin layer of gold–palladium using
a Hummer I Sputter Coater (Technics) in a 100 mTorr
vacuum in an argon environment for a 3-min period
using 10 mA of current. Images were taken using a
JEOL JSM-840 SEM at a 5 kV accelerating voltage at
15,000 times magnification. Digital images were re-
corded using the Digital Scan Generator Plus (JEOL)
software.
2.3. Cells
Non-transformed human osteoblast cells (CRL-
11372) were purchased from the American Type Culture
Collection and were used at population numbers less
than 5. Osteoblasts were cultured in Dulbecco’s
modified Eagle medium (DMEM; Hyclone), supple-
mented with 10% fetal bovine serum (FBS; Gibco) and
1% penicillin/streptomycin (P/S; Gibco), under stan-
dard cell culture conditions (sterile chamber maintained
at 37
C and a humidified environment: 5% CO2/95%
air).
2.4. Cell adhesion
As a first attempt to investigate how osteoblasts
would interact with the various nanofiber alumina
compacts, adhesion assays were performed. For this
purpose, osteoblasts were seeded (3500 cells/cm2) onto
the compacts in DMEM supplemented with 10% FBS
and 1% P/S and were allowed to adhere in standard cell
culture conditions for 1 h. After the prescribed time
period, substrates were rinsed three times using phos-
phate buffered saline (PBS) to remove non-adherent
cells. The adhered cells were fixed with formaldehyde
(Fisher), stained with Hoechst 33258 dye (Sigma), and
 ARTICLE IN PRESS
T.J. Webster et al. / Biomaterials 26 (2005) 953–960
counted using fluorescence microscopy (365 nm excita-
tion and 400 nm emission wavelengths). Five random
fields were counted per substrate. Experiments were
performed in triplicate and repeated at least three times.
2.5. Total intracellular protein synthesis
Functions of osteoblasts after adhering to the
nanofiber alumina compacts were also determined in
the present study. For this purpose, osteoblasts (40,000
cells/cm2) were seeded onto the compacts and were
cultured in DMEM supplemented with 10% FBS, 1%
P/S, 50 mg/ml l-ascorbate (Sigma), and 10 mm b-
glycerophosphate (Sigma) under standard cell culture
conditions for 14 days. Medium was replaced every
second day. At the end of the prescribed time period,
supernatant medium was removed and the remaining
osteoblasts were lysed using distilled water and three
freeze–thaw cycles. This protocol only removes intra-
cellular as well as cell membrane-bound proteins (and
does not analyze proteins contained in the super-
natant media or the extracellular matrix [15,19]).
Sodium azide (Sigma) was used to prevent degradation
of proteins. Total protein content in the cell lysates was
determined spectrophotometrically using a commer-
cially available kit (Pierce Chemical Co.) and following
the manufacturer’s instructions. Total intracellular
protein synthesized by osteoblasts cultured on the
compacts of interest to the present study was determined
from a standard curve of absorbance versus known
concentrations of albumin run in parallel with the
experimental samples [14,15]. Experiments were re-
peated six times.
2.6. Intracellular alkaline phosphatase activity
Osteoblasts were seeded (40,000 cells/cm2) onto the
compacts and were cultured in DMEM supplemented
with 10% FBS, 1% P/S, 50 mg/ml l-ascorbate, and
10 mm b-glycerophosphate under standard cell culture
conditions for 14 days. Medium was replaced every
second day. The method of Lowry et al. [19] was used to
assay alkaline phosphatase activity in the cell lysates
(prepared as described in the total intracellular protein
synthesis section). Alkaline phosphatase is a marker of
osteoblast differentiation from non-calcium depo-
siting to calcium depositing cells [11]. Alkaline phos-
phatase synthesized by osteoblasts cultured on the
compacts of interest to the present study was determined
from a standard curve of absorbance versus known
concentrations of p-nitrophenol run in parallel with
experimental samples. The alkaline phosphatase activity
was normalized by total intracellular protein synthesis
and substrate surface area. Experiments were repeated
six times.
955
2.7. Calcium deposition
Osteoblasts were seeded (40,000 cells/cm2) onto the
compacts and were cultured in DMEM supplemented
with 10% FBS, 1% P/S, 50 mg/ml l-ascorbate, and
10 mm b-glycerophosphate under standard cell culture
conditions for 14 days. Medium was replaced every
second day. At that time, cells were lysed for the
purpose described in the total intracellular protein
synthesis and intracellular alkaline phosphatase activity
sections. The extracellular matrix synthesized on each
compact of interest to the present study was then
treated with 0.6 n HCl at 37
C overnight. After the
prescribed time period, the amount of calcium present in
the acidic supernatant was spectrophotometrically
quantified using a commercially available kit (Calcium
Quantification Kit #587-A; Sigma) and following
the manufacturer’s instructions; light absorbance of
the samples was measured spectrophotometrically at
575 nm. Total calcium (mg/dl) was calculated from
standard curves of absorbance versus known concentra-
tions of calcium measured in parallel with the experi-
mental samples. Calcium precipitation on the compacts
in the absence of cells was also determined using
the same protocols as just described but without
cells. Acellular calcium precipitation results were sub-
tracted from results obtained in this section to ac-
quire an accurate measurement of calcium deposited
by osteoblasts. Calcium concentration values were
then normalized by total protein synthesis and
substrate surface area. Experiments were repeated six
times.
2.8. Statistical analysis
Results were analyzed using analysis of variance
(ANOVA) techniques; statistical significance was con-
sidered at po0:01:
3. Results
3.1. Material characterization
3.1.1. Chemistry
It was found that the nanofiber alumina compacts
synthesized in this study possessed differences in
chemical composition, crystalline phase, and topogra-
phy. Specifically, Table 1 outlines the different chemical
compositions of the compacts as determined from
ESCA measurements. It can be seen that the chemical
composition of the nanofiber alumina compacts unsin-
tered and sintered at 400
C closely matched the
stoichiometry of boehmite (AlOOH). Specifically, the
nanofiber alumina compacts unsintered and sintered at
400
C had an O/Al ratio of 1.9 which would be expected
 ARTICLE IN PRESS
956 T.J. Webster et al. / Biomaterials 26 (2005) 953–960
Table 1
A chemical composition analysis of the nanofiber alumina compacts as determined by ESCA
Nanofiber alumina sintering temperature (
C) Chemical elements
Oxygen
Unsintered and at 400 59
600, 800, 1000, and 1200 52
if the majority of the compact was boehmite (AlOOH).
These data confirm previous studies that have shown the
primary chemistry of the unsintered alumina nanofibers
to be boehmite [18]. In contrast, nanofiber alumina
sintered at or above 600
C closely matched the
stoichiometry of alumina (Al2O3). Specifically, an O/Al
ratio of 1.4 was found for all of the nanofiber alumina
compacts sintered at or above 600
C, which is close to
the 1.5 ratio that would be expected for a chemistry
of Al2O3.
3.1.2. Crystal phase
Differences in crystalline phases between nanofiber
alumina compacts can be seen in the XRD results (Fig.
1, Table 2). The unsintered nanofiber alumina spectrum
was characteristic of boehmite alumina, which emits
distinct peaks at 2y values of 28
, 38
, 45
, 49
, 55
, and
72
. A similar characteristic boehmite spectrum
was observed for nanofiber alumina sintered at 400
C.
This validates the ESCA data previously presented
in Table 1. In contrast, when boehmite nanofiber
alumina was sintered at 600
C, gamma crystalline
phase alumina was prevalent as indicated by two large
peaks at a 2y of around 45
and 67
, three close peaks
at around 32
, 37
, and 39
, and a large peak at around
61
. Boehmite nanofiber alumina continued a phase
transformation when sintered at 800
C as indicated
by the introduction of a delta crystalline phase into
the gamma crystalline phase; this was noticed by
peaks at 2y values of 33
and 34
. When sintered at
1000
C, boehmite nanofiber alumina compacts were
mostly theta phase with smaller amounts of delta
phase as indicated by the introduction of sharp
peaks at 2y values of 31
, 33
, 36
, 44
, 47
, 60
, 62
,
and 66
. Lastly, boehmite nanofiber alumina compacts
transformed to alpha alumina as indicated by strong
peaks at 2y values of 25
, 35
, 37
, 43
, 52
, 58
, and
61
when sintered at 1200
C. Such XRD data confirm
literature reports that demonstrated alumina undergoes
a delta to theta to alpha phase transformation in the
temperature range of 1000–1200
C [20]. This present
data also support ESCA results given in Table 1 that
demonstrated a chemical change in boehmite alumina
when sintered at either 600
C, 800
C, 1000
C, or
1200
C.
Carbon Aluminum O/Al ratio
7 31 1.9
9 37 1.4
3.1.3. Topography
Finally, SEM images portray visible differences in the
surface roughness between all the nanofiber alumina
compacts (Fig. 2). Individual and agglomerated nano-
fiber alumina particles were observed at this magnifica-
tion in all compacts (individual alumina fiber sizes are
2 nm in diameter with lengths greater than 50 nm [18]).
Moreover, while micron size porosity was evident on
gamma and gamma+delta nanofiber alumina, micron
size porosity was not observed on unsintered boehmite,
boehmite sintered at 400
C, theta+delta nanofiber
alumina, and alpha nanofiber alumina.
3.2. Cellular studies
3.2.1. Adhesion
Results did not show any statistically significant
differences in osteoblast adhesion when comparing the
nanofiber alumina compacts of interest to the present
study (Fig. 3). However, osteoblast adhesion was
significantly (po0:01) greater on all alumina compacts
compared to glass after 1 h.
3.2.2. Alkaline phosphatase activity
In contrast to adhesion results, there were significant
differences in alkaline phosphatase activity when osteo-
blasts were cultured on the various nanofiber alumina
compacts of interest to this study (Fig. 4). Specifically,
alkaline phosphatase activity by osteoblasts on theta+
delta nanofiber alumina was greater (po0:01) than on
any other nanofiber alumina formulation after 14 days
of culture. In addition, osteoblast alkaline phosphatase
activity was greater (po0:01) when cultured on boeh-
mite nanofiber alumina sintered at 400
C compared to
gamma, gamma+delta, and alpha nanofiber alumina.
In fact, no alkaline phosphatase activity was deter-
mined by the techniques used in this study when
osteoblasts were cultured on gamma, gamma + delta,
and alpha nanofiber alumina. Most importantly, osteo-
blasts produced more than two times the amount of
alkaline phosphatase when cultured on theta+delta
compared boehmite nanofiber alumina. Alkaline phos-
phatase activity was similar when osteoblasts were
cultured on boehmite nanofiber alumina and glass
(reference material).
 ARTICLE IN PRESS
T.J. Webster et al. / Biomaterials 26 (2005) 953–960 957

300 Unsintered Alumina

Boehmite Standard
250
Alumina Sintered at 400˚ C
200 400
Intensity

Boehmite Standard
150 300

Intensity
100 200

50 100

0 0
20 30 40 50 60 70 80 20 30 40 50 60 70 80
2 Theta Angle 2 Theta Angle

Alumina Sintered at 800˚C

Alumina Sintered at 600˚C 400
Gamma Alumina Standard
400
Gamma Alumina Standard Delta Alumina Standard
300
300
Intensity
Intensity

200
200

100
100

0 0
20 30 40 50 60 70 80 20 30 40 50 60 70 80
2 Theta Angle 2 Theta Angle

Alumina Sintered at 1200˚C

2000
Alumina Sintered at 1000˚C Alpha Alumina Standard
5 00 Delta Alumina Standard 1600

4 00 Theta Alumina Standard

Intensity

1200
Intensity

3 00
800
2 00

400
1 00

0 0
20 30 40 50 60 70 80 20 30 40 50 60 70 80
2 Theta Angle 2 Theta Angle
Fig. 1. A comparison of the XRD spectra of the various nanofiber alumina compacts. JCPDS standards are given for alumina crystalline phase
comparisons.

3.2.3. Calcium deposition highest substrate. Boehmite nanofiber alumina sintered

Similar to alkaline phosphatase results, osteoblasts at 400
C had the next largest amount of calcium
deposited the most (po0:01) calcium when cultured on deposition which was greater (po0.01) than that
theta+delta nanofiber alumina compared to any other deposited on gamma, gamma+delta, and alpha nano-
substrate tested in this study after 14 days (Fig. 5). fiber alumina. Calcium was deposited by osteoblasts in
Almost two times the amount of calcium was measured significantly (po0:01) higher amounts on gamma+delta
on theta+delta nanofiber alumina compared to the next than gamma nanofiber alumina. Lastly, osteoblasts
 ARTICLE IN PRESS
958 T.J. Webster et al. / Biomaterials 26 (2005) 953–960

Table 2
A crystal phase analysis of the nanofiber alumina compacts as
determined by X-ray diffraction

Nanofiber alumina sintering temperature (
C) Crystalline phase

Unsintered and at 400 Boehmite

600 Gamma
800 Gamma+delta
1000 Theta+delta
1200 Alpha

Fig. 3. Similar osteoblast adhesion on all nanofiber alumina compacts

after 1 h. Values are mean7SEM, n ¼ 3; and po0:01 (compared to
the reference material, glass). It is important to note that the same
trend was observed for each n:

Not sintered Sintered at 400 oC

Boehmite Boehmite

Sintered at 600 oC Sintered at 800 oC

Gamma Gamma + Delta

Fig. 4. Enhanced osteoblast alkaline phosphatase activity on theta+

o o delta nanofiber alumina compacts after 14 days. Values are mean
Sintered at 1000 C Sintered at 1200 C
7SEM, n ¼ 6; po0:01 (compared to boehmite nanofiber alumina
Theta + Delta Alpha
compacts), and po0:01 (compared to gamma, gamma+delta, and
Fig. 2. SEM images at 15,000 times magnification of the various alpha nanofiber alumina compacts). It is important to note that the
nanofiber alumina compacts. The scale bar (—) represents 1 mm. same trend was observed for each n:

deposited more (po0:01) calcium on gamma than alpha

nanofiber alumina after 14 days. No calcium deposition
was detected by the techniques used in this study when
osteoblasts were cultured on glass (reference material).
4. Discussion
Bonding of juxtaposed bone to a biomaterial surface
is a property shown to be a common indicator for
successful orthopedic implants [21,22]. Analysis of failed
implants uncovers several familiar themes: biomaterial
mechanical failure, soft-fibrous tissue encapsulation,
and/or osteolysis [23–26]. Many of the successful
prostheses promote bone ingrowth containing active
osteoblasts (bone-forming cells), osteoclasts (bone-re-
sorbing cells), and osteocytes (mechano-transduction
cells) [21,27]. It is because of these promising studies that
several researchers are examining how to control events
at the bone/biomaterial interface with the hopes of
improving the efficacy of implants [28]. One of the focal
points of such studies is improving material properties
of implants to create an environment more conductive
for osteoblast function and, subsequently, bone in-
growth.
For example, recent research has been conducted to
determine bone cell functions on novel formulations of
alumina [29–33]. This is because although traditional
large (micron) spherical particle size alumina has been
used in total hip arthroplasties [33] and single-crystal
 ARTICLE IN PRESS
T.J. Webster et al. / Biomaterials 26 (2005) 953–960
Fig. 5. Enhanced calcium deposition by osteoblasts on theta+delta
nanofiber alumina compacts after 14 days. Values are mean 7SEM,
n ¼ 6; po0:01 (compared to boehmite nanofiber alumina com-
pacts), po0:01 (compared to gamma+delta nanofiber alumina
compacts), zzpo0:01 (compared to gamma nanofiber alumina
compacts), and zpo0:01 (compared to alpha nanofiber alumina
compacts). It is important to note that the same trend was observed
for each n:
alumina in dental implants [25,26] with some success,
such materials have occasionally led to limited fixation
with bone and, consequently, eventual clinical failure.
For these reasons, researchers are currently investigating
new formulations of alumina for use in orthopedic
applications.
One such new alumina formulation, nanophase
spherical alumina, has been investigated [29–32].
Throughout these studies it was found that osteoblast
and osteoclast function increased on alumina with
spherical nanophase compared to conventional micron
grain sizes. However, nanofiber (not nanospherical)
alumina more closely simulates the physical geometry of
hydroxyapatite crystals and collagen fibers in bone that
osteoblasts are accustomed to interacting [11]. Studies
have indeed shown that osteoblast function (specifically,
viable cell adhesion, proliferation, and deposition of
calcium containing mineral) was enhanced on nanofiber
compared to nanospherical particulate alumina [13–15].
The present study utilized the same nanofiber alumina
investigated in those studies [13–15] (identified as
boehmite nanofiber alumina sintered at 400
C in the
present study), but sought to determine osteoblast
functions on a wide-range of crystal structured nanofi-
ber alumina. In doing so, this study provided the first
evidence of increased osteoblast functions on theta+
delta compared to boehmite nanofiber alumina.
It was an important objective of the present study to
elucidate material properties of alumina nanofibers that
influenced osteoblast behavior. There are several possi-
bilities: nanometer surface features, chemistry, and
crystallinity. Since all of these material properties
959
changed when comparing boehmite to theta+delta
nanofiber alumina, it is unclear at this time what
influenced osteoblast function between these substrates.
However, correlations can be made between the
remaining gamma, gamma+delta, theta+delta, and
alpha nanofiber alumina compacts investigated in this
study since the chemistry was similar between these
substrates. In this manner, this is one of the few studies
to delve into the influence of nanofiber alumina crystal-
line phase on osteoblast behavior. In doing so and when
considering similar alumina (Al2O3) chemistries (i.e.,
omitting boehmite), this study provided the first
evidence of increased deposition of calcium containing
mineral by osteoblasts according to the following trend
of crystalline phases in nanofiber alumina: theta+
delta>gamma+delta>gamma>alpha.
Increased osteoblast functions on gamma compared
to alpha nanofiber alumina as shown in this study
confirm previous investigations which have shown larger
amounts of osteoblast adhesion, proliferation, alkaline
phosphatase activity, and calcium deposition by osteo-
blasts on gamma compared to alpha nanospherical
alumina [29,30]. Another study by Nishio et al. [34]
reported that osteoblastic differentiation on composites
containing conventional grain size gamma-crystalline
phase alumina allowed direct bone formation, whereas
the composite containing the conventional grain size
alpha crystalline phase alumina did not. The present
study confirmed these same trends of increased osteo-
blast functions on gamma compared to alpha crystalline
phases but transferred such trends to novel nanofiber
alumina samples.
Certainly, future studies will need to be performed to
determine how osteoblasts recognize such changes in
alumina nanofiber crystal structure; for example, by
elucidating initial protein interactions that mediate
subsequent osteoblast functions on these alumina
compacts. In fact, while it has been reported that
osteoblast adhesion is enhanced on nanometer com-
pared to conventional alumina of the same crystalline
phase (but decreased grain size) in the presence of
serum, in the absence of serum, correlations between
grain size and osteoblast adhesion were not observed
[29]. In this manner, changes in initial protein interac-
tions are likely candidates for altered osteoblast
response to the materials of interest to the present study.
5. Conclusions
The present study implicated alumina nanofibers
(specifically, theta+delta crystalline phase) as an
important material in the future design of orthopedic
materials with increased osteoblast cytocompatibility
properties. The ability of such alumina fibers to
approximate the nanometer dimensions of constituent
 ARTICLE IN PRESS
960 T.J. Webster et al. / Biomaterials 26 (2005) 953–960
components of bone offers exciting possibilities in the
design of orthopedic implants with greater efficacy that
should not be overlooked.
Acknowledgements
The authors would like to thank the National Science
Foundation for financial support through the Research
Experience for Undergraduates (REU Grant 0097696)
program as well as Argonide, Corporation for supplying
the boehmite nanofiber alumina particles. The authors
would also like to thank Dr. Stephen Golledge at the
University of Washington for the ESCA data in this
publication.
References
[1] Siegel RW. Creating nanophase materials. Sci Am 1996;275:42–7.
[2] Baraton MI, Chen X, Gonsalves KE. FTIR study of nano-
structured alumina nitride powder surface: determination of the
acidic/basic sites by CO, CO2, and acetic acid adsorptions.
Nanostruct Mater 1997;8:435–9.
[3] Bohn R, Haubold R, Birringer R, Gleiter H. Nanocrystalline
intermetallic compounds—an approach to ductility. Scr Met
Mater 1991;25:811.
[4] Carry C, Mocellin A. Structural superplasticity in single phase
crystalline ceramics. Ceram Int 1987;13:89–98.
[5] Catledge S, Vohra Y. Effect of nitrogen addition on the
microstructure and mechanical properties of diamond films grown
using high-methane concentrations. J App Phys 1999;86:698–700.
[6] Ciftcioglu M, Mayo MJ. Processing of nanocrystalline ceramics.
In: Mayo MJ, Kobayashi M, Wadsworth J, editors. Super-
plasticity in Metals, Ceramics, and Intermetallics Symposium
Proceedings. Pittsburgh, PA: Materials Research Society; 1990.
p. 77–86.
[7] Cui Z, Hahn H. Tensile deformation of nanostructured TiO2 at
low temperatures. Nanostruct Mater 1992;1:419–23.
[8] Nieman GW, Weertman JR, Siegel RW. Mechanical behavior of
nanocrystalline Cu and Pd. In: Van Aken DC, editor. Micro-
composites and nanophase materials. Warrendale, PA: TMS;
1991. p. 15–7.
[9] Mayo M, Siegel RW, Liao YX, Nix WD. Nanoindentation of
nanocrystalline ZnO. J Mater Res 1992;7:973.
[10] Ayad L. The extracellular matrix factsbook. New York, NY:
Plenum Press; 1994.
[11] Kaplan FS, Hayes WC, Keaveny TM, Boskey A, Einhorn TA,
Iannotti JP. Form and function of bone. In: Simon SP, editor.
Orthopedic basic science. Columbus, OH: American Academy of
Orthopedic Surgeons; 1994. p. 127–85.
[12] www.argonide.com.
[13] Gutwein LG, Tepper F, Webster TJ. Increased osteoblast
function on nanofibered alumina. 26th Annual American Ceramic
Society Meeting, Cocoa Beach, FL, 2004, in press.
[14] Price RL, Gutwein LG, Kaledin L, Tepper F, Webster TJ.
Osteoblast function on nanophase alumina materials: influence of
chemistry, phase, and topography. J. Biomed Mater Res
2003;67(4):1284–93.
[15] Price RL, Haberstroh KM, Webster TJ. Enhanced functions of
osteoblasts on nanostructured surfaces of carbon and alumina.
Med Biol Eng Comput 2003;41:372–5.
[16] Boni O, Berger S. Dielectric properties of KDP filled porous
alumina nanocomposite thin films. J Nanosci Nanotechnol
2001;4:433–9.
[17] Luo X, Zhao Y, Chao Y, Tian J, Zhang Y, Zhang S. Effects
of alumina particle-size on mechanical properties of alumina–
glass composite. Hua Xi Kou Qiang Yi Xue Za Zhi 1999;17:
23–5.
[18] Tepper F, Lerner M, Ginley D. Nanosized alumina fibers. Am
Ceram Soc Bull 2001;80:57–60.
[19] Lowry OH, Roberts NR, Wu M, Hixon WS, Crawford EJ. The
quantitative histochemistry of the brain II. Enzyme measure-
ments. J Biol Chem 1954;207:19–37.
[20] Souza Santos P, Souza Santos H, Toledo SP. Standard transition
aluminas, electron microscopy studies. Mater Res 2000;3:105–14.
[21] Steflik DE, Corpe RS, Lake FT, Sisk AL, Parr GR, Hanes PJ,
Buttle K. Composite morphology of the bone and associated
support–tissue interfaces to osseointegrated dental implants:
TEM and HVEM analyses. Int J Oral Maxillofacial Impl
1997;12:443–53.
[22] Steflik DE, Corpe RS, Young TR, Sisk AL, Parr GR. The
biological tissue responses to uncoated and coated implanted
biomaterials. Adv Dent Res 1999;13:27–33.
[23] Anderson JM, Gristina AG, Hanson SR, Harker LA, Johnson
RJ, Merritt K, Naylor PT, Schoen FJ. Host reactions to
biomaterials and their evaluation. In: Ratner BD, Hoffman AS,
Schoen AS, Lemons JE, editors. Biomaterials science: an
introduction to materials in medicine. San Diego: Academic
Press; 1996. p. 165–214.
[24] Jones LC, Frondoza C, Hungerford DS. Immunohistochemical
evaluation of interface membranes from failed cemented and
uncemented acetabular components. J Biomed Mater Res
1999;48:889–98.
[25] Steflik DE, Parr GR, Singh BB, Lake FT, Sisk AL, Howell FV,
Shelton TW. Light microscopic and scanning electron micro-
scopic analyses of dental implants retrieved from humans. J Oral
Implantol 1994;20:18–24.
[26] Steflik DE, Corpe RS, Young TR, Parr GR, Tucker M, Sims M,
Tinley J, Sisk A, McDaniel M. Light microscopic and scanning
electron microscopic retrieval analyses of implanted biomaterials
retrieved from humans and experimental animals. J Oral
Implantol 2001;27:5–15.
[27] Steflik DE, Corpe RS, Young TR, Sisk AL, Parr GR. The
biological tissue responses to uncoated and coated implanted
biomaterials. Adv Dent Res 1999;13:27–33.
[28] Puleo DA, Nanci A. Understanding and controlling the bone–
implant interface. Biomaterials 1999;20:2311–21.
[29] Webster TJ, Siegel RW, Bizios R. Osteoblast adhesion on
nanophase ceramics. Biomaterials 1999;20:1221–7.
[30] Webster TJ, Ergun C, Doremus RH, Siegel RW, Bizios R.
Enhanced functions of osteoblasts on nanophase ceramics.
Biomaterials 2000;21:1803–10.
[31] Webster TJ, Schadler LS, Siegel RW, Bizios R. Mechanisms of
enhanced osteoblast adhesion on nanophase alumina involve
vitronectin. Tissue Eng 2001;7:291–301.
[32] Webster TJ, Ergun C, Doremus RH, Siegel RW, Bizios R.
Specific proteins mediate enhanced osteoblast adhesion on
nanophase ceramics. J Biomed Mater Res 2000;51:475–83.
[33] Hamadouche M, Meunier A, Greenspan DC, Blanchat C, Zhong
JP, La Torre GP, Sedel L. Bioactivity of sol–gel bioactive
glass coated alumina implants. J Biomed Mater Res 2000;52:
422–9.
[34] Nishio K, Neo M, Akiyama H, Okada Y, Kokubo T, Nakamura
T. Effects of apatite and wollastonite containing glass-ceramic
powder and two types of alumina powder in composites on
osteoblastic differentiation of bone marrow cells. J Biomed Mater
Res 2001;55:164–76.