Design of Functionally Graded Dental Implant in the Presence of

Cancellous Bone
H. S. Hedia
Product Engineering and M/c Design Department, Faculty of Engineering, Mansoura University, Mansoura, 35516 Egypt

Received 7 May 2004; revised 26 October 2004; accepted 10 November 2004

Published online 15 July 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.b.30275

Abstract: In a previous work by the author [Hedia HS, Mahmoud NA. Biomed Mater Eng
2004;14(2):133–143], a functionally graded material (FGM) dental implant was designed
without cancellous bone in the model. In this investigation, the effect of the presence of
cancellous bone as a thin layer around the dental implant was investigated. It is well known
that the main inorganic component of natural bone is hydroxyapatite (HAP) and that the main
organic component is collagen (Col). HAP implants are not bioabsorbable, and because
induction of bone into and around the artificially made HAP is not always satisfactory,
loosening or breakage of HAP implants might occur after implantation in the clinical appli-
cation. The development of a new material that is bioabsorbable and that has osteoconductive
activity is needed. Therefore, the aim of the current investigation was to design an implant, in
the presence of cancellous bone as a thin layer around it, from FGM. In this study, a novel
biomaterial, Col/HAP, as a FGM, was developed using the finite element and optimization
techniques that are available in the ANSYS package. These materials have a self-organized
character similar to that of natural bone. The investigations have shown that the maximum
stress in the cortical bone and cancellous bone for the Col/HAP functionally graded implant
has been reduced by about 40% and 19%, respectively, compared with currently used
titanium dental implants. © 2005 Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater 75B:
74 – 80, 2005

Keywords: hydroxyapatite; collagen; bone; porosity; stress; ceramic

INTRODUCTION poor durability of hydroxyapatite (HAP) coatings over time

Bone is a nanoscale composite with mechanical properties
that cannot be duplicated by monolithic materials. Composite
systems for bone replacement have been developed using
various bioactive ceramic or glass reinforcements in a bioin-
ert polymer matrix. The mechanical behavior of these com-
posites is dependent on the interface between the composite
components, which in turn affect the interface with the bio-
logical system.1
So, in the field of biomedical implants, attempts to design
suitable material systems have been made both from conven-
tional engineering and tissue engineering directions. Current
bone implants research is primarily focused in four direc-
tions: (1) polymeric coatings on metallic implants, (2) com-
posites, (3) tissue engineering, and recently (4) functionally
graded material (FGM). The coated metallic implants have
provided a superior mechanical implant anchorage and im-
proved ceramic stability over other bioactive coatings. Some
of the problems associated with the current coated metallic
implants include stress shielding of the surrounding bone and
Correspondence to: H. S. Hedia (e-mail: hedia@mans.edu.eg)
© 2005 Wiley Periodicals, Inc.
74
resulting in severe biocompatibility issues.2
The mechanical reliability of pure HAP ceramics is low,
therefore it cannot be used as artifical teeth or bones. For
these reasons, various HAP-based composites have been fab-
ricated, but only HAP-coated titanium (Ti) alloys have found
widespread application. The microstructurally controlled
HAP ceramics such as fibers/whiskers-reinforced HAP, fi-
brous HAP-reinforced polymers, or biomimetically fabri-
cated HAP/collagen (Col) composites seem to be the most
suitable ceramic materials for future hard tissue replacement
implants.3
However, the original concept of a ceramic reinforced
polymer composite to be used as a bone substitute was
introduced in 1981. The idea was to develop bone-analog
composites using bone itself as the template to copy. HAP,
the major inorganic constituent of bone, was selected as the
reinforcement. The use of HAP allows simultaneously for the
mechanical reinforcement of the polymer and for a bioactive
behavior of the prosthesis (inducing bone-bonding and favor-
ing osteointegration). Col has been used in several biomate-
rial applications. Polyethylene (PE) was selected as the duc-
tile matrix to try to ensure the necessary ductility after the
incorporation of the brittle HAP reinforcement. For the last
 FUNCTIONALLY GRADED DENTAL IMPLANT
15 years, PE/HAP composites have been proposed for a range
of modest load-bearing applications such as cranial-facial,
maxillofacial, and middle ear implants and orbital floor re-
construction.4
From the standpoint of hard tissue response to implant
materials, calcium phosphate is probably the most compatible
of materials known. In recent years, much attention has been
given to HAP and b-tricalcium phosphate as potential bio-
materials for bone substitute. Good implantation of biomate-
rials, such HAP and b-tricalcium phosphate, in the skeleton is
evidenced by an ability to reach full integration of the non-
living implant with living bone.5
Attempts have been made to form high-strength consoli-
dated HAP ceramics. Calcium phosphate ceramics, especially
HAP and b-tricalcium phosphate, are widely used for hard
tissue replacement because of their biocompatibility and os-
teoconductive properties. These ceramics are used in powder
and block forms as bone defect fillers. The applications of
calcium phosphates in the body have been limited by the low
strength and low fracture toughness of the synthetic phos-
phates. HAP is used in composites to retain useful bioactive
properties as well as provide enhancement in mechanical
properties. The composites include fiber reinforcement of
HAP, HAP/PE, and HAP/polyethyl ester. Composites of
HAP with bone phosphoprotein, bone Gla protein, and Col
have also been attempted. Mechanical properties in the above
composite systems are still inadequate for the potential use of
these composites for total bone replacement.2
Several current research efforts have thus focused on com-
posites of synthetic polymers and HAP formed using in situ
mineralization techniques. A few polymeric additives such as
poly(acrylic acid) and acrylic acid and methacryloyl-
poly(oxyethylene)phosphate copolymer have been investi-
gated as suitable organic additives because of their calcium
binding properties. Some of these polymers are known to
accelerate the nucleation of HAP from calcium and phosphate
solutions. Mechanical properties of such composites are still
inadequate for their potential use for bone replacement al-
though some improvement in the mechanical properties of the
composites was achieved over the simple mixing of organic
and HAP. The specific knowledge of the fundamental mech-
anism of the association of organic functionalities with HAP
during HAP nucleation and growth is lacking.2
Itoh et al.6 demonstrated that the HAP/Col composite has
a structure that resembles natural bone. The finding suggests
that the HAP/Col composite has a self-organized character
similar to that of natural bone, and that HAP/Col is the first
material found to induce these osteogenic cells and bone
remodeling units.6 Rigid union between HAP/Col and recip-
ient bone can be obtained as early as 8 weeks after implan-
tation, showing that it has very high osteoconductive activity.
Furthermore, it is suggested that this composite may be a
good carrier of rhBMP and that the quantity of rhBMP used
for bone induction can be reduced.
The merge of materials science with the biological sci-
ences has produced some advanced materials which are
called functionally graded materials (FGMs) in order to pro-
vide the desired material characteristics. The FGMs concept
75
originated in Japan in 1984 during the space plan project.
FGMs are materials or structures in which the material prop-
erties vary with location in such a way as to optimize some
functions of the overall FGMs.7
Ti/HAP FGM was fabricated by spark plasma sintering
and the mechanical properties and biocompatibility as an
implant were evaluated. Maturation of bone formed at the
implant surface proceeded earlier in the HAP-rich part than in
the Ti-rich part. The results showed that Ti/HAP FGM im-
plants made by the spark plasma sintering method had
strength, excellent biocompatibility, and controllability for
graded bioreaction.8
Boughton and Ruys9 developed a polymer-fiber FGM
using ceramic fibers (HAP) in a thermoplastic polymer ma-
trix. The proposed application of this novel material is as a
spinal disk prosthesis.
The main goal of the current investigation was to introduce
the optimal design of FGM dental implant in the presence of
cancellous bone as a thin layer around the implant, in order to
achieve an excellent biocompatibility while maintaining long-
term chemical and mechanical stability as results of matching
the properties between the implant and the bone. The finite
element method and optimization technique were used to
achieve the required implant design. The rules of mixture and
the volume fraction relations that represent the FGM of the
dental implant were used. Different variations of the composi-
tions were considered through the optimization technique.
FGM MODELING
Materials Models
Identification of material properties is an essential step in all
engineering fields. There are two different models, functions, for
representation of the continuous gradation of the material prop-
erties of FGMs. These models are the analytical solution and the
volume fraction and rules of mixture. In analytical solution
of FGM problems, exponential functions for continuous grada-
tion of the material properties were considered10 –16 as:
E ⫽ Eoe␤x,␯ ⫽ ␯oe␤x,␣ ⫽ ␣oe␥x, and k ⫽ koe␦x where E ⫽
Young’s modulus, ␯ ⫽ Poisson’s ratio, ␣ ⫽ coefficient of linear
thermal expansion, and k ⫽ heat conductivity. Eo, ␯o, ␣o, and ko
are the material properties of the constituent materials at x ⫽ 0.
␤, ␥, and ␦ are material nonhomogeneous parameters that rep-
resent the continuous variation of the material properties along x
directions. Exponential functions for representation of the ma-
terial properties usually facilitate the analytical solution but do
not give real representation for material properties, except for the
upper and lower surfaces of FGM, where the effect of volume
fractions distributions are not taken into account. A more real-
istic way for representation of the continuous gradation of the
material properties of FGM is the volume fraction and rules of
mixture. The use of volume fraction and rules of mixture over-
come the drawbacks of exponential functions for representation
of the material properties. Unfortunately, the use of volume
fraction and rules of mixture makes the solution of the FGM
problems very complicated and perhaps impossible. However,
76 HEDIA

the use of the finite element method in such problems is the most where E and ␯ are modulus of elasticity and Poisson’s ratio,
effective tool and overcomes such difficulties. Many authors respectively. Subscripts m and c indicate metal and ceramic,
(see e.g. refs. 17 and 18) analyzed FGM problems, in different respectively. Eo is the modulus of elasticity without porosity
cases, using volume fractions and rules of mixture. effect (when porosity p ⫽ 0).

Volume Fractions and Rules of Mixture of FGM

First, let us consider an implant of FGM with porosity that FINITE ELEMENT ANALYSIS AND NUMERICAL
functionally graded from ceramic and metal. The expression OPTIMIZATION
of the material composition of the nonhomogeneous implant,
volume fractions of metal (Vm) and ceramic (Vc), are distrib- The geometry of the dental implant is shown in Figure 1.23
uted over the y direction according to the following rela- The finite element analysis and optimization were performed
tions19: using the ANSYS package.24 Axisymmetric eight-node har-
monic elements were used to form the mesh shown in Figure
2. This type of element is capable of supporting some non-
V c ⫽ (y/h) m (1)
axisymmetric loading on geometrically axisymmetric struc-
tures in addition to axial loading25 and has been used in a
V m ⫽ (1 ⫺ V c) (2) pervious study.26,27 The model consists of cortical and sleeve
of cancellous bone around the implant. Fully bonded inter-
where y and h are vertical position in y axis and implant faces were assumed. In the model, most of the bone mesh was
height, respectively. Also, m is a constant that represents the cortical of stiffness (16 GPa),28 except for a thin layer of
composition variations between ceramic and metal along the cancellous bone (E ⫽ 1 GPa) lining the sides of the im-
y axis. plant28,29 as shown in Figure 1. These bone distributions are
If m ⫽ 1, the variation of the composition of ceramics and the same as used in the previous study.26 An axial loading
metal is linear. The composition is ceramic-rich when m ⬍ 1 condition on the implant was applied. The magnitude of this
and ceramic-poor when m ⬎ 1. load was 100 N.23
The porosity p of the FGM can be represented by the A numerical optimization procedure was applied in com-
following equation19,20: bination with the above FE model using the ANSYS FE
program in order to optimize the materials of the implant and
p⫽A 冉 冊冋 冉 冊册
y
h
n
1⫺
y
h
z
(3)
the variation of the composition of the materials. A computer
program was developed using ANSYS Parametric Design
Language (APDL) to calculate the modulus of elasticity and
where A is given as Poisson’s ratio, using Equations (1)–(7) at the different po-
sitions of the implant height.
n⫹z n
n 冉 冊 The design objective is to minimize the maximum nodal
von Mises stress in the bone. The design variables for this

冉 冊 z ⱖAⱖ0 (4)
n
1⫺
n⫹z

where n, z are arbitrary constants.

The effective values of the material properties for FGM,
with porosity and continuously graded profile, are determined
by using the suspended spherical grain model (see e.g. refs.
21 and 22). It was derived based on the assumption that the
granular phase is in a matrix phase. The expression of the
material properties of the nonhomogeneous implant material,
modulus of elasticity and Poisson’s ratio, are expressed over
y direction, depending on ceramics and metal volume frac-
tions, according to the following relations:

Eo (1 ⫺ p)
E⫽ (5)
1 ⫹ p(5 ⫹ 8␯)(37 ⫺ 8␯)/兵8(1 ⫹ ␯) (23 ⫹ 8␯)其

Eo ⫽ Ec 冋 Ec ⫹ (Em ⫺ Ec )Vm2/3
Ec ⫹ (Em ⫺ Ec )(Vm2/3 ⫺ Vm ) 册 (6)

Figure 1. Schematic diagram of Leeds University implant design and

␯ ⫽ ␯mVm ⫹ ␯c Vc (7) bone model (dimensions in millimeters).23
 FUNCTIONALLY GRADED DENTAL IMPLANT
Figure 2. Finite element mesh for full model (a) and for separate
model (b).
problem are the elastic moduli of the basic constituent mate-
rials of the FGM implant (Ec and Em), minimize the maxi-
mum von Mises stress in the cancellous bone and the varia-
tion of the composition, Vc and Vm, which are described by
the constant m. Thus, the objective function for this problem
is to minimize the maximum nodal von Mises stress in the
bone under the following constrains:
● To maintain the maximum nodal von Mises stress in the
cancellous bone less than or equal to the initial value
obtained for Ti implant (initial design).
Si ⱕ Sb max
where Si is the maximum nodal von Mises stress in cancel-
lous bone, and Sb max is the initial maximum von Mises
stress in cancellous bone.
● To maintain the values of the design variable Em and Ec
within the values used in literature, that is, 1 GPa ⱕ Ec and
Em ⱕ 210 GPa, which cover a great range of bioceramic
and biometal materials.
● To maintain the values of the design variable m within the
limits used in literature16,17 as 0 ⱕ m ⱕ 10.
RESULTS
In the current investigation, stress analysis and optimization
processes for the dental implant were performed. The pro-
cesses were started from Ti implant material as initial de-
signs, equal elastic moduli for the basic constituents (Ec ⫽ Em
⫽ 110 GPa) and zero value for ceramic volume fraction and
unity for metallic volume fraction at the whole of the implant.
Then, the optimization process was continued using different
values of elastic moduli of ceramics and metal constituents,
that lay in the range 1 GPa ⱕ Ec and Em ⱕ 210 GPa and
different values of m in the range 0 ⱕ m ⱕ 10, that represent
the variation of the composition between Vc and Vm. Eleven
77
iterations with minimal CPU time were required to determine
the optimal elastic moduli and the parameter of variation of
the composition, m. From a strength and biocompatibility
point of view, the optimal values are found to be 2 and 80
GPa, which represents by Col (ECol) and HAP (Ec) materials
and m ⫽ 0.5.
The effect of the parameter m, which represents the com-
position variation between ceramic and Col, on the volume
fractions distributions is shown in Figure 3. It shows the
ceramic volume fractions distributions (Vc) versus the vertical
position along the implant height for various values of the
parameter m, according to Equation (1). The optimal m value,
which minimizes the maximum nodal von Mises stresses in
the cancellous and cortical bone at the interfaces, is found to
be m ⫽ 0.5. It is observed that, when m ⫽ 1, the variation of
the composition of ceramics and Col is linear. The composi-
tion is ceramic-rich when m ⬍ 1 and ceramic-poor when m
⬎1.
Figure 4 shows the variations of the elastic modulus along
the implant height for different values of m for the Col/HAP
FGM dental implant, according to Equations (5)–(7). There-
fore, the Col/HAP FGM implant with HAP-rich composition,
m ⫽ 0.5, is the optimal design from the excellent biocom-
patibility point of view. According to the current investiga-
tion, it can be concluded that the Col/HAP FGM implant with
HAP-rich composition, m ⫽ 0.5, is the optimal design that
minimizes the maximum nodal von Mises stresses in the bone
and produces excellent biocompatibility, between implant
and cancellous bone, at the same time.
To check the reliability of our introduced optimal design
of the FGM implant, it was compared with Ti dental implant
(initial design). Therefore, comparisons of nodal von Mises
stress for the Ti (initial design) and Col/HAP (FGM) optimal
were performed.
Figure 3. Ceramic volume fraction distribution along the implant
height for various values of parameter m.
78
Figure 4. Elastic modulus distribution for FGM implant materials
along the implant height for various values of parameter m.
The nodal von Mises stress in the implant along the
implant/cancellous bone interface is shown in Figure 5. The
stresses for optimal (FGM) design are everywhere lower
compared with those in the initial design. Moreover, the
maximum nodal von Mises stress for optimal design is re-
duced by 33% compared with the initial design.
A comparison of nodal von Mises stress in cancellous
bone along the implant/cancellous bone interface for initial
and optimal design (FGM) is shown in Figure 6. It is ob-
served that the maximum stress, which is responsible for
bone deposition and microfracture, for optimal design is
Figure 5. Comparison of stress in dental implant along implant/
cancellous bone interface for initial and optimal design.
HEDIA
Figure 6. Comparison of stress in cancellous bone along implant/
cancellous bone interface for initial and optimal design. [Color figure
can be viewed in the online issue, which is available at www.
interscience.wiley.com.]
reduced by 19% compared with the initial design. The min-
imum stress, which leads to net bone resorption, is increased.
Therefore, the bone is continuously under mechanical loading
which helps it to stay healthy and remodel.
The nodal von Mises stress in cancellous bone along the
cancellous/cortical bone interface is shown in Figure 7. The
stresses for optimal design everywhere are higher than those
for the initial design except for the proximal end. However,
the maximum stress for the optimal design in this interface is
lower than that for the implant/cancellous bone interface as
shown in Figure 6.
Figure 8 shows the nodal von Mises stress in cortical bone
along the cortical/cancellous bone-implant interface. The
maximum stress along this interface is reduced by 40%
compared with the initial design of the implant. This reduc-
tion of maximum stress prevents microfracture and fatigue
failure of the bone.
DISCUSSION
No artificial implant has biomechanical properties equivalent
to the original tissue; the implants cannot repair themselves as
Figure 7. Comparison of stress in cancellous bone along bone/
cancellous bone interface for initial and optimal design.
 FUNCTIONALLY GRADED DENTAL IMPLANT
Figure 8. Comparison of stress in cortical bone along cortical/can-
cellous bone and implant interface for initial and optimal design.
the living tissues can. Consequently, a man-made spare part
may never be as good as the living tissue it replaces. Bone is
a typical example of a material with an adequate balance of
biochemical and mechanical properties that is very hard to
reproduce. Bone strength, stiffness, and toughness are the
result of its composite and anisotropic structure of organic
(Col) and inorganic (HAP) materials.4 Furthermore, the ma-
terials for replacements should not be too stiff in order to
avoid the stress-shielding of bone. In fact, it is well estab-
lished4 that the bone must be continuously under mechanical
loading to stay healthy and remodel. As a consequence, an
ideal load-bearing implant material must match as closely as
possible bone mechanical behavior, especially its stiffness.
The HAP/Col composite has a structure that resembles
that of natural bone. These findings suggest that the HAP/Col
composite has a self-organized character similar to that of
natural bone, and that HAP/Col is the first material found to
induce these osteogenic cells and bone remodeling units.6
Rigid union between HAP/Col and recipient bone can be
obtained as early as 8 weeks after implantation, showing that
it has very high osteoconductive activity.6
Therefore, the use of Col and HAP for FGM is a unique
way to produce suitable stiffness and a biocompatible mate-
rial that may be used in biomedical applications. Col and
HAP, the major organic and inorganic constituent of bone,
were selected as the materials for the FGM of the implant.
The use of HAP allows simultaneously for the mechanical
reinforcement of Col and for a bioactive behavior of the
prosthesis inducing bone-bonding and favoring osteointegra-
tion. Col was used as a second material because of the
similarity of stiffness with cancellous bone. The similarity of
stiffness reduced the stress levels in the adjacent cancellous
bone and produced favorable stress levels.
The effects of stress on the precise mechanisms of bone
remodeling, microfracture and fatigue, and bone resorption
are complex and not fully understood.30 It is clear, however,
that a particular level of stress is required to maintain the
bone density around an implant, and that this level will also
be dependent on the physiological loading that occurs in the
different parts of the body.31 Levels of stress below this
maintenance level will lead to net bone resorption whereas
79
elevated levels will lead to net bone deposition.31 For very
high levels of stress, bone microfracture and fatigue can
occur which can also lead to rapid bone resorption.30 In the
interpretation of the stress in this study, von Mises stresses
were used as an indication of bone remodeling and when a
microfailure can occur.23,26,27 The von Mises stresses were
compared with the yield stress of cortical bone and cancellous
bone which have a predicted yield stress of 115 and 20 MPa,
respectively.32 The level of stress in the cancellous and cor-
tical bone for the FGM implant was approximately 11% and
13% of the predicted yield stress, respectively.
CONCLUSIONS
1. The optimal design of the Col/HAP FGM implant re-
duces the stresses concentration in the cortical bone,
cancellous bone, and implant compared with conven-
tional Ti. This is an indication of preventing microfrac-
ture and fatigue failure of bone.
2. Col/HAP FGM is an optimal implant from excellent
biocompatibility and controllability for graded bioreac-
tion points of view.
3. The use of the FGM concept in introducing dental im-
plant materials achieves full integration of the implant
with living bone which aids in increasing the life of the
implant.
REFERENCES
1. Bonfield W, Wang M, Tanner KE. Interfaces in analogue bio-
materials. Acta Mater 1998;46(7):2509 –2518.
2. Katti K, Gujjula P. Control of mechanical responses in in situ
polymer-hydroxyapatite composites for bone replacement: The
15th Engineering Mechanics Conference of the ASCE; June
2–5, 2002; Columbia University, New York.
3. Suchanek W, Yoshimura M. Processing and properties of hy-
droxyapatite-based biomaterials for use as hard tissue replace-
ment implants. J Mater Res 1998;13(1):94 –117.
4. Reis RL, Cunha AM, Oliveira MJ, Campos AR, Bevis MJ.
Relationship between processing and mechanical properties of
injection molded high molecular mass polyethylene and hy-
droxyapatite composites. Mater Res Innovations 2001;4:263–
272.
5. Jallot E. Correlation between hydroxyapatite osseointegration
and Young’s modulus. Med Eng Phys 1998;20:697–701.
6. Itoh S, Kawauchi T, Shinomiya K, Koyama Y, Matumoto H,
Takakuda K, Masanori K, Tanaka J. Osteoconductive activity of
a novel biomaterial, hydroxyapatite/collagen composite: 46th
annual meeting, Orthopaedic Research Society; March 12–15,
2000; Orlando, FL.
7. Suresh S, Mortensen A. Fundamentals of functionally graded
materials. Cambridge, UK: University Press; 1998.
8. Miyao R, Yokoyama A, Watari F, Kawasaki T. Properties of
titanium/hydroxyapatite functionally graded implants by spark
plasma sintering and their biocompatibility. J Dent Mater 2001;
20(6):344 –355.
9. Boughton P, Ruys AJ. Ceramic-polymer functionally graded
material: a novel spinal disk prosthesis. Available from: http://
www.materials.unsw.edu.au. Accessed August 2000.
10. Erdogan F, Kaya AC, Joseph PF. The crack problem in bonded
non-homogeneous material. ASME J Appl Mech 1991;58:410 –
418.
80
11. Erdogan F, Wu BH. The surface crack problem for a plate with
functionally graded properties. ASME J Appl Mech 1997;64:
449 – 456.
12. Choi HJ. An analysis of cracking in a layered medium with a
functionally graded non-homogenous interface. ASME J Appl
Mech 1996;63:479 – 486.
13. Jin Z, Noda N. Crack-tip singular fields in non-homogeneous
materials. ASME J Appl Mech 1994;61:738 –740.
14. Noda N, Jin Z. Crack-tip singularity fields in non-homogeneous
body under thermal stress fields. JSME Ser A 1995;38:364 –
369.
15. Noda N, Jin Z. Steady thermal stresses in a infinite non-homog-
enous elastic solid containing a crack. J Therm Stresses 1993;
16:181–196.
16. Wang BL, Han GC, Du SY. Thermo-elastic fracture mechanics
for non-homogenous material subjected to unsteady thermal
load. ASME J Appl Mech 2000;67:87–95.
17. Fuchiyama T, Noda N, Tsuji T, Obata Y. Analysis of thermal
stress and stress intensity factor of functionally gradient mate-
rials. In: Holt IB, Munir ZA, Koizumi M, Hirai T, editors.
Ceramic transactions: functionally gradient materials, vol. 34.
Westerville, OH: American Ceramic Society; 1993. p 425– 432.
18. Fujimoto T, Noda N. Multiple crack growths in the functionally
graded plate under thermal shock. Proceedings of the 4th Inter-
national Congress on Thermal Stresses; June 8 –11, 2001;
Osaka, Japan. p 121–124.
19. Nemat-Alla M. Reduction of thermal stresses by developing
two-dimensional functionally graded materials. Int J Solids
Struct 2003;40:7339 –7356.
20. Hedia HS, Mahmoud NA. Design optimization of functionally
graded dental implant. Biomed Mater Eng 2004;14(2):133–143.
21. Kerner EH. The elastic and thermo-elastic properties of com-
posite media. Proc Phys Soc Lond B 1956;69:808.
HEDIA
22. Kingery WD, Bowen H, Uhlmann DR. Introduction of ceram-
ics. New York: John Wiley & Sons; 1976.
23. Clif SE, Fisher J, Watson CJ. Stress and strain distribution in the
bone surrounding a new design of dental implant: a comparison
with a threaded Branemark type implant. Proc Inst Mech Eng
(H) 1993;207:133–138.
24. ANSYS user’s manual, Version 5.0A. 1992.
25. Zienkiewicz OC. The finite element method. London: McGraw-
Hill; 1977.
26. Clift SE, Fisher J, Watson CJ. Finite element stress and strain
analysis of the bone surrounding a dental implant: effect of
variation of bone modulus. Proc Inst Mech Eng (H) 1992;206:
233–241.
27. Hedia HS. Stress and strain distribution behavior in the bone
due to the effect of cancellous bone dental implant material and
the bone height. Biomed Mater Eng 2002;12(2):111–120.
28. Hayes WC. Bone mechanics: from tissue mechanical properties
to an assessment of structure behaviour. In: Schmid-Schonbein
GW, Zweifach BW, editors. Frontiers of biomechanics. New
York: Springer-Verlag; 1986. p 196 –209.
29. Brear K, Currey JD, Raines S, Smith KJ. Density and temper-
ature effects on some mechanical properties of cancellous bone.
Eng Med 1988;17:163–167.
30. Huiskes R, Nunamaker D. Local stresses and bone adaption
around orthopaedic implants. Calcif Tissue Int 1984;361:110 –
117.
31. Huiskes R. Biomechanics of bone implant interactions: In:
Schmid-Schonbein GW, Zweifach BW, editors. Frontiers of
biomechanics. New York: Springer-Verlag; 1986. p 245–262.
32. Hayes WC. Bone mechanics: from tissue mechanical properties
to an assessment of structure behaviour: In: Schmid-Schonbein
GW, Zweifach BW, editors. Frontiers of biomechanics. New
York: Springer-Verlag; 1986. p 196 –209.