December 1, 2008
11:16
00202
SPI-131-J084-JMR
Journal of Musculoskeletal Research, Vol. 11, No. 3 (2008) 135–143
© World Scientific Publishing Company
ASSESSMENT OF MECHANICAL PROPERTIES OF HUMAN
OSTEON LAMELLAE EXHIBITING VARIOUS DEGREES OF
MINERALIZATION BY NANOINDENTATION
Sabine Bensamoun∗,§ , Zaifeng Fan† , Ilharreborde Brice‡ , Jae Young Rho†
and Marie-Christine Ho Ba Tho∗
∗Laboratoire
de Biomécanique et Bioingénierie, CNRS UMR 6600
Université de Technologie de Compiègne
BP 20529, F-60205 Compiègne Cedex, France
†Department of Biomedical Engineering
University of Memphis, Memphis, TN, USA
‡ Département de Chirurgie Orthopédique Pédiatrique
Hôpital Robert Debré, 75935 Paris Cedex 19, France
§ [email protected]
Accepted 8 September 2008
ABSTRACT
Background: Cortical bone analysis has been investigated at the macroscopic level with mechanical
tests and imaging techniques, but few studies have been done at the microscopic level (osteons). The
purpose of this study is to measure the elastic modulus of thick lamellae of osteons exhibiting different
degrees of mineralization. This study aims to provide clinicians with a better understanding of bone
remodeling and help in assessing the different stages of bone healing. Methods: Six femoral human
samples (5 mm × 5 mm × 5 mm) were cut transversally along the length of a human femur. Scanning
electron micrographs were produced to reflect the composition of the microstructure. Three types of
osteons were selected: white (high mineralization), gray (intermediate mineralization), and dark (low
mineralization) osteons. Nanoindentation tests were performed on three locations of the thick lamella
located in the middle of each osteon. The mechanical test induced three holdings and unloadings
with a constant holding of 10 s. The maximal force was 2500 µN, which induced a maximal depth of
about 400 nm. Results: Elastic modulus (E) and hardness (H) for the white (N = 61), gray (N = 17),
and dark (N = 39) osteons were Ewhite = 21.30 GPa ± 3.00 GPa and Hwhite = 0.55 GPa ± 0.15 GPa,
Correspondence to: Sabine Bensamoun, PhD, Laboratoire de Biomécanique et Bioingénierie, CNRS UMR 6600, Université
de Technologie de Compiégne, BP 20529, F-60205 Compiègne Cedex, France.
135
December 1, 2008
136
11:16
00202
SPI-131-J084-JMR
S. Bensamoun et al.
Egray = 19.27 GPa ± 1.78 GPa and Hgray = 0.41 GPa ± 0.09 GPa, and Edark = 12.95 GPa ± 2.66 GPa
and Hdark = 0.30 GPa ± 0.10 GPa, respectively. The variation of elastic properties within a lamella was
approximately 2.6 GPa, depending on the level of mineralization. Conclusions: These results demonstrate the inhomogeneity of the lamella, suggesting that both the orientation of collagen fibers and
the degree of mineralization may vary within the lamella. Our study shows a large range of elastic
properties and hardness, reflecting different degrees of osteon mineralization.
Keywords: Nanoindentation; Cortical human bone; Bone remodeling; Osteon lamellae; Elastic
modulus.
INTRODUCTION
The mechanical properties of human cortical bone
have been investigated at the tissue (Haversian system) and microstructural (osteon, osteon
lamella) levels. At the microstructural level, three
types of osteons have been identified based on the
collagen fibers’ orientation and classified as follows: transversal (type I), alternate (type II), and
longitudinal (type III) osteons.1−4 The mechanical properties of isolated osteons have been determined using different mechanical tests such as
compressive, torsion, and bending tests.
Compressive tests were performed on osteons
from two human femoral bones (30−80 years
old) with two different degrees of calcification.
The authors found that the values of the elastic properties for the osteons increased from type
III to type I for calcified osteons and for osteons
at the initial stage of calcification. Furthermore,
the range of values for osteons at the initial
stage of calcification was found to be lower than
for the fully calcified ones.4 Tensile tests were
performed on dry and wet osteons (of various
types) of a human femur (30 years old) with two
stages of calcification. The dry specimen exhibited higher mechanical properties and the degree
of calcification increased the elastic properties of
osteons, which were more significant for the wet
specimen.3 Torsional tests applied to fully calcified femoral osteons showed that the shear elastic
modulus for type III osteons was higher than for
type II osteons.2 The bending properties of single osteons (types II and III) were investigated,
and the elastic properties of type II osteons were
found to be higher than those of type III osteons.1
Thus, the alternate (type II) structure seems to
the authors to be more adapted to the bending
stresses. These initial results indicate that each
osteon type exhibits specific mechanical properties related to the orientation of the collagen fibers,
which appears to be well adapted for handling
different types of stresses.
Initial investigations into the mechanical properties of osteon lamellae (width between 3 µm and
7 µm) were performed using a nanoindentation
technique by Rho et al.13 The Young’s modulus of
the osteon and the interstitial lamellae were measured using different human bones (femur and
tibia) and different categories of bone (cortical and
spongious).11−15 Values of the interstitial lamellae were found to be significantly higher (around
26 GPa) than those of the secondary osteon lamellae (19−22.5 GPa). The hardness range found
for the secondary osteon in the literature varied
between 0.42 GPa and 0.65 GPa.11,12 The elastic properties of osteon lamellae also varied by
about 2 GPa from the center of the osteon to the
outer, with a higher value in the center. This result
was not found to be consistent for all osteons.14
In addition to the osteon area, the elastic properties of osteon lamellae were dependent on the
anatomical location. For instance, values measured from the diaphysis to the femoral neck varied by about 16%.16 According to the authors,
the elastic moduli found for the femoral neck
were lower than those of the mid-diaphysis. This
December 1, 2008
11:16
00202
SPI-131-J084-JMR
Mechanical Assessment of Osteons
last observation could explain the fragility of the
femoral neck at the macroscopic scale.
Thick and thin osteon lamellae underwent
nanoindentation tests with different experimental
configurations (different indentation depths) and
different physiological conditions (dry and wet).7
The thick lamellae showed high mechanical properties (indentation modulus and hardness) for low
depth indentation, while the mechanical properties decreased at higher indentation depths (the
change was less significant for the thin lamellae).
The results were consistent with both physiological conditions. According to the authors, this supports the hypothesis that thick and thin lamellae
have different compositions and ultrastructural
properties.
The question of the influence of age on the elastic properties of lamellae was analyzed by Rho
et al.15 They observed no significant difference in
elastic properties between young and old people,
probably due to the low number of samples.15
The reported nanoindentation measurements
have shown variations in the elastic properties of
osteon lamellae related to the anatomical location,
experimental conditions, and anisotropy.5 Moreover, different types of lamellae (thick, thin, and
interstitial) have shown specific elastic properties
related to different chemical compositions. At the
osteon level, the influence of decalcification was
demonstrated for all osteon types, but no investigation has been performed on osteon lamellae.
The purpose of this study is to identify the
degree of mineralization in order to measure the
mechanical properties of osteon lamellae using a
nanoindentation technique.
MATERIALS AND METHODS
Specimen Preparation
A cadaver femur (natural death, male, 70 years of
age) was obtained from the anatomical laboratory
of the University Hospital of Amiens, France. Six
137
samples (5 mm×5 mm×5 mm) were cut parallel
to the axis of the femur with a low-speed diamond saw (MICROCUT 2). These samples were
extracted from the lateral and medial sides of
the femur between 40% and 70% of the total
length. Each sample was ground with a different size of sandpaper (#600, #800, and #1200)
in order to obtain a smooth surface and then
polished with 0.5 µm aluminum powder. Control of the surface was performed through an
optical microscope in reflection (NIKON 198477,
Nikon, Japan) with a magnification of ×10 at
the Integrated Microscopy Center, University of
Memphis, USA.
Calibration
Fused silica, an elastic isotropic material, was
used to calibrate the tip shape function of the
indenter.10 This material is currently used for
this kind of calibration technique because the
ratio between the modulus and the hardness was
very low.
Scanning Electron Microscopy
A backscattered environmental scanning electron
micrograph (Philips XL-30 ESEM) was produced
under high vacuum at 20 kV with a magnification of ×45. The surfaces of all the samples
were scanned with an electron beam in order
to have a mapping of the backscattered electrons, which reflects the mineralization of the surface. This study was performed at the Integrated
Microscopy Center, University of Memphis, USA.
The advantage of this technique was that none of
the specimen surfaces were treated.6 Then, three
different types of osteons were identified based
on image preprocessing. Histogram analysis on
the gray levels (0−255) of the native images identified three gray-level ranges (89−107, 110−122,
December 1, 2008
138
11:16
00202
SPI-131-J084-JMR
S. Bensamoun et al.
osteons characterized by darker gray levels (representing an intermediate degree of mineralization) and a well defined external border. Dark
osteons (having the lowest degree of mineralization) were characterized by the darkest gray
levels.
3
2
1
Nanoindentation Technique
Fig. 1 Backscattered image of the surface of the cortical sample. The three different types of osteons are represented by
white osteons (1), gray osteons (2), and dark osteons (3).
120−127) corresponding to the white, gray, and
dark osteons, respectively (Fig. 1). White osteons
were characterized by the lightest gray levels (representing a high degree of mineralization) with
an unclear outer border, compared to the gray
Measurements of Young’s modulus were determined through a mechanical technique which
applied compressive tests at the microstructural
level. The mechanical equipment (Hysitron Inc.,
Minneapolis, MN, USA) was composed of a
Berkovich diamond indenter that has a pyramidal shape and an optical microscope (Fig. 2).
The regions to be indented were selected and
located through the optical microscope, and the
indenter was then optically driven to the target
location. The nanoindentation technique used a
load−time sequence (Fig. 3) composed of two preconditioning cycles followed by a constant holding (for 10 s), which leads to reduced effects due
Optical
microscope
Berkovich
diamond indenter
Cortical sample
Fig. 2 Representation of the coupling between the optical microscope and the indenter.
December 1, 2008
11:16
00202
SPI-131-J084-JMR
Mechanical Assessment of Osteons
139
The elastic modulus of the sample (ES ) was
related to the stiffness (S) and the contact area (A)
by the method of Oliver and Pharr10 :
−1
√
1 − νi2 1 − νs2
S
π
×√ =
+
,
(1)
2
Ei
Es
A
Fig. 3 The load cycle applied to the indenter.
to viscoelasticity14 and improves the convergence
to the steady-state response of the material. The
load−displacement data corresponding to the last
unloading phase were used to determine the elasticity of the material (Fig. 4). The maximum peak
load used was 2500 µN, which resulted in an
indentation depth of about 400 nm, and the stiffness was determined from the data between 50%
and 95% of the peak load.
where ν is Poisson’s ratio, the subscript ‘‘s’’ corresponds to the sample (νs = 0.3), and the subscript
‘‘i’’ refers to the indenter. The indenter is characterized by Ei = 1140 GPa and by νi = 0.07. Hardness (H) was measured as the ratio between the
maximum load performed and the contact area
(A). Equation (1) is usually applied at the macroscopic level for homogeneous and isotropic materials but it is well known that bone material is
anisotropic. Therefore, the effective modulus is an
average of the anisotropic elastic constants.
Nanoindentation tests were performed on the
three types of osteons. Each indentation was performed three times on the thick lamella (defined
by a lighter gray level compared to the darker, thin
lamella) located in the center of an osteon, considered to be the most representative area of the
95% of the peak
load
50% of the
peak load
Slope = S
Fig. 4
Representation of a typical curve representing the load−displacement performed.
December 1, 2008
140
11:16
00202
SPI-131-J084-JMR
S. Bensamoun et al.
RESULTS
Calibration of the Nanoindentation
Technique
×
×
Indentation around the
same thick lamella
×
The elastic modulus of the fused silica measured
with the nanoindentation technique was about
72.1 GPa ± 0.5 GPa. This experimental value is
close to the reference value of 72 GPa. Thus,
proper calibration enables confirmation of future
results found by this technical experimentation.
The reproducibility of the elastic modulus and
hardness obtained from the three repeated measurements performed at different times on the
fused silica was about E = 1.5 GPa and H = 0.05
GPa, respectively.
Fig. 5 Localizations of the indentations.
entire osteon7 (Fig. 5). A total of 351 indentations
from dry samples were performed on 61 white, 17
gray, and 39 dark osteons. The reproducibility of
the technique was assessed by performing three
repeated tests executed three different times on
fused silica.
Statistical Analysis
Analysis of variance (ANOVA) was performed
with the software Statgraphics 5.0 (Sigma Plus,
Maryland, USA) in order to study firstly the variation of the elastic properties (E) and hardness (H)
within the same thick lamella, as well as the difference of the elastic modulus and hardness between
different types of osteons.
Variation of Elastic Properties and
Hardness within a Thick Lamella
The intralamellar variations of the mean elastic modulus measurements for the white, gray,
and dark osteons were 2.93 GPa ± 1.83 GPa,
2.69 GPa ± 1.57 GPa, and 2.45 GPa ± 1.70 GPa,
respectively (Table 1). The mean hardness measurements for the white, gray, and dark osteons
were 0.15 GPa ± 0.10 GPa, 0.10 GPa ± 0.08 GPa,
and 0.09 GPa ± 0.08 GPa, respectively (Table 1).
The approximate variation of the elastic properties and the hardness within a lamella (intralamellar) was 2.6 GPa (with a range of variation of
0.2−8 GPa) and 0.11 GPa (with a range of variation of 0.01−0.43 GPa), respectively. These values
were dependent on the types of osteons.
Table 1 Variability of Elastic Modulus and Hardness Measured within a Thick Lamella from White, Gray,
and Dark Osteons.
White Osteons
(N = 61)
Range
Median
Mean ± standard deviation
Gray Osteons
(N = 17)
Dark Osteons
(N = 39)
E (GPa)
H (GPa)
E (GPa)
H (GPa)
E (GPa)
H (GPa)
0.19−7.76
2.81
2.93 ± 1.83
0.02−0.43
0.11
0.15 ± 0.10
0.93−5.93
2.42
2.69 ± 1.57
0.02−0.26
0.06
0.10 ± 0.08
0.24−8.48
2.12
2.45 ± 1.70
0.01−0.31
0.05
0.09 ± 0.08
December 1, 2008
11:16
00202
SPI-131-J084-JMR
Mechanical Assessment of Osteons
141
Table 2 Comparison of Elastic Modulus and Hardness Measured within a Thick Lamella from White, Gray,
and Dark Osteons.
White Osteons
(N = 61)
Range
Median
Mean ± standard deviation
H (GPa)
E (GPa)
H (GPa)
E (GPa)
H (GPa)
13.87−27.06
21.59
21.30 ± 3
0.26−0.85
0.56
0.55 ± 0.15
16.88−22.63
19.15
19.27 ± 1.78
0.25−0.6
0.43
0.41 ± 0.09
7.04−17.59
13.33
12.95 ± 2.66
0.09−0.52
0.29
0.30 ± 0.10
White
The interlamellar variations of the elastic modulus and hardness are summarized in Table 2.
White osteons had a higher modulus and hardness compared to gray osteons (E = 2 GPa and
H = 0.13 GPa, P < 0.05) and dark osteons
(E = 8 GPa and H = 0.27 GPa, P < 0.001)
(Figs. 6 and 7). The elastic modulus and hardness
of the gray osteons were also significantly higher
(E = 6 GPa and H = 0.14 GPa, P < 0.001) than
those of the dark osteons.
DISCUSSION
In the present study, the variation of the mechanical properties within a lamella was approximately E = 2.6 GPa (Ewhite_osteons = 2.81 GPa,
White
Dark
*
*
*
* *
* *
*
7
11
15
Dark Osteons
(N = 39)
E (GPa)
Comparison of Elastic Properties and
Hardness between Different Types of
Osteon Lamellae
Gray
Gray Osteons
(N = 17)
19
23
27
31
E (GPa)
Fig. 6 Range of values of elastic modulus for different types
of osteons. ***: P < 0.001; **: P < 0.05.
*
Gray * *
* *
*
Dark
0
*
*
0.2
0.4
0.6
0.8
1
H (GPa)
Fig. 7 Range of values of hardness for different types of
osteons. ***: P < 0.001; **: P < 0.05.
Egray_osteons = 2.42 GPa, Edark_osteons = 2.12 GPa)
with a range of variation from 0.2 GPa to 8 GPa,
and the variation of the hardness was about H =
0.11 GPa (Hwhite_osteons = 0.11 GPa, Hgray_osteons =
0.06 GPa, Hdark_osteons = 0.05 GPa) with a range
of variation from 0.01 GPa to 0.43 GPa. This result
demonstrates the inhomogeneity of the lamella,
suggesting that both the orientation of collagen
fibers and the degree of mineralization may vary
within the lamella. This last result was not in
agreement with the study of Ascenzi and Bonucci,
who considered the lamellae as homogeneous for
all osteon types (I, II, and III).4
The three different categories of osteons were
characterized by their gray level, reflecting different degrees of mineralization. The quantitative
measurements showed an increase in the elastic modulus and hardness with the gray level of
December 1, 2008
142
11:16
00202
SPI-131-J084-JMR
S. Bensamoun et al.
the osteon, confirming the presence of different
degrees of mineralization. The observed increase
of the mechanical properties is in agreement with
the data found by Ascenzi and Bonucci4 for all
osteon types. Our ratio (factor 2) between the elasticity of dark osteons and white or gray osteons
is close to that found by Ascenzi and Bonucci4 for
the alternate osteon (type II)4 . The ratio for the
other types of osteons (I and III) would be around
1.3. Our data are statistically significant and provide a quantitative range of elasticity that could
be used as reference values for fully mineralized
and less mineralized osteon lamellae.
In the present study, the ranges of the elastic
modulus and hardness for the white and gray
osteons are within the same range as those published in the literature.11−16
The elastic properties measured in the present
study exhibit a variation of about 40% (from
13.33 GPa to 21.59 GPa) between the different
types of osteons. It is interesting to note that
this variation of modulus at the microscopic
scale is more important than the variation of
modulus found at the macroscopic level, i.e.
at the tissue level (11%−20%).9 Previous work
performed on macroscopic and microscopic measurements using ultrasound and nanoindentation techniques showed that the elastic modulus
measured at the macroscopic scale represents the
homogenized elastic properties measured at the
microstructural level.8
To summarize, the mechanical properties of
thick lamellae of different types of osteons in
the process of mineralization were characterized.
Our data provided a range of values for the
elastic properties of bone at the microstructural
level, reflecting the remodeling process. The bone
quality assessment using nanoindentation provided helpful information about the level of mineralization and maturation of the osteons. The
cortical bone analysis described in this study
might give clinicians more knowledge about bone
remodeling as well as help in assessing the different stages of bone healing. Further studies using
the nanoindentation technique will include investigations into the pathogenesis of various bone
diseases.
ACKNOWLEDGMENTS
This work was performed in collaboration with
the University of Memphis (Department of
Biomedical Engineering) and is particularly dedicated to the memory of Professor Jae-Young Rho,
with thanks to the contribution of Mrs Lou G.
Boykins from the Integrated Microscopy Center.
References
1. Ascenzi A, Baschieri A, Benvenuti A. The bending properties of single osteons. J Biomech 23: 763−771, 1990.
2. Ascenzi A, Baschieri A, Benvenuti A. The torsional properties of single selected osteons. J Biomech 27: 875−884,
1994.
3. Ascenzi A, Bonucci E. The tensile properties of single
osteons. Anat Rec 158: 375−386, 1967.
4. Ascenzi A, Bonucci E. The compressive properties of single osteons. Anat Rec 161: 377−392, 1968.
5. Fan Z, Swadener JG, Rho JY, Roy ME, Pharr GM.
Anisotropic properties of human tibial cortical bone as
measured by nanoindentation. J Orthop Res 20: 806−810,
2002.
6. Harris JR. Electron Microscopy in Biology: A Practical
Approach, CIRL Press, Oxford, 1991.
7. Hengsberger S, Kulik A, Zysset JP. Nanoindentation discriminates the elastic properties of individual human
bone lamellae under dry and physiological conditions.
Bone 30: 178−184, 2002.
8. Ho Ba Tho MC, Bensamoun S, Rho JY. Macro-micro characterization of mechanical properties of human bone, in
Proceedings of the 11th International Congress of Biomedical
Engineering, p. 109, 2002.
9. Ho Ba Tho MC, Rho JY, Ashman RB. Atlas of mechanical
properties of human cortical and cancellous bone, in Van
Der Perre G, Lowet G, Borgwardt A (eds.), In Vivo Assessment of Bone Quality by Vibration and Wave Propagation
Techniques. Part II, ACCO, Leuven, Belgium, pp. 7−32,
1991.
10. Oliver WC, Pharr GM. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J Mater Res
76: 1564−1583, 1992.
December 1, 2008
11:16
00202
SPI-131-J084-JMR
Mechanical Assessment of Osteons
11. Rho JY, Pharr GM. Effect of drying on the mechanical
properties of bovine femur measured by nanoindentation. J Mater Sci Mater Med 10: 485−488, 1999.
12. Rho JY, Roy ME, Tsui TY, Pharr GM. Elastic properties
of microstructural components of human bone tissue as
measured by nanoindentation. J Biomed Mater Res 45:
48−54, 1998.
13. Rho JY, Tsui TY, Pharr GM. Elastic properties of human
cortical and trabecular lamellar bone measured by
nanoindentation. Biomaterials 18: 1325−1330, 1997.
14. Rho JY, Zioupos P, Currey JD, Pharr GM. Variation in
the individual thick lamellar properties within osteons
by nanoindentation. Bone 25: 295−300, 1999.
143
15. Rho JY, Zioupos P, Currey JD, Pharr GM. Microstructural
elasticity and regional heterogeneity in human femoral
bones of various ages examined by nanoindentation.
J Biomech 35: 189−198, 2002.
16. Zysset PK, Guo XE, Hoffler CE, Moore KE, Goldstein
SA. Elastic modulus and hardness of cortical and trabecular bone lamellae measured by nanoindentation in the
human femur. J Biomech 32: 1005−1012, 1999.