See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/14970992

The Effects of Enamel Anisotropy on the Distribution of Stress in a Tooth

Article  in  Journal of Dental Research · December 1993

DOI: 10.1177/00220345930720111101 · Source: PubMed

CITATIONS READS

108 72

5 authors, including:

Iain Richard Spears Richard van Noort

Pro-football Support Ltd The University of Sheffield
68 PUBLICATIONS   1,546 CITATIONS    247 PUBLICATIONS   5,368 CITATIONS   

SEE PROFILE SEE PROFILE

Robin Huw Crompton Ian C Howard

University of Liverpool The University of Sheffield
142 PUBLICATIONS   4,191 CITATIONS    132 PUBLICATIONS   2,310 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Structural Integrity Assessment View project

mercury release View project

All content following this page was uploaded by Robin Huw Crompton on 31 May 2014.

The user has requested enhancement of the downloaded file.

 Journal of Dental Research
http://jdr.sagepub.com/

The Effects of Enamel Anisotropy on the Distribution of Stress in a Tooth

I.R. Spears, R. Van Noort, R.H. Crompton, G.E. Cardew and I.C. Howard
J DENT RES 1993 72: 1526
DOI: 10.1177/00220345930720111101

The online version of this article can be found at:

http://jdr.sagepub.com/content/72/11/1526

Published by:

http://www.sagepublications.com

On behalf of:
International and American Associations for Dental Research

Additional services and information for Journal of Dental Research can be found at:

Email Alerts: http://jdr.sagepub.com/cgi/alerts

Subscriptions: http://jdr.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Citations: http://jdr.sagepub.com/content/72/11/1526.refs.html

Downloaded from jdr.sagepub.com by guest on July 15, 2011 For personal use only. No other uses without permission.
The Effects of Enamel Anisotropy on the Distribution of Stress in a Tooth
I.R. SPEARS, R. van NOORT1, R.H. CROMPTON, G.E. CARDEW2, and I.C. HOWARD2
Department of Human Anatomy and Cell Biology, University of Liverpool, Liverpool L69 3BX, United Kingdom;
'Department of Restorative Dentistry and 2Department of Mechanical and Process Engineering, University of Sheffield,
S10 2SZ, United Kingdom

Enamel is thought to have highly anisotropic stiffness to our understanding of the mechanical function of re-
characteristics, because of its prismatic structure. It is stored teeth to know how the natural materials in teeth-
probable that the enamel is stiffer in the prism direction that is, dentin and enamel-behave under the application
compared with a direction perpendicular to it. The prisms ofan external load. In other words, we need to know which
are thought to run approximately perpendicular to the of the various alternative models of their biomechanical
enamel-dentin junction. The curvilinear anisotropy that characteristics can adequately describe their behavior.
will result can readily be modeled by TOMECH, a finite The human dentition is replaced once in a lifetime, but
element program developed at the University of Sheffield, during this lifetime, the permanent teeth undergo cycli-
since curvilinearity of mechanical properties is available cal loading many times. Any mechanical structure un-
as an automated feature of this program. The patterns of dergoing this kind of repetitive loading can deteriorate,
stress due to an external load were investigated in two- both from stress overload and from propagation of cracks
dimensional abstract models, and in a model of a man- as a consequence of mechanical fatigue. Thus, it is im-
dibular second premolar, for both anisotropic and isotro- perative that stresses be kept to a minimum. In order to
pic enamel. Results were compared with the commercial withstand the test of time, teeth must therefore be both
code ANSYS and good agreement obtained. Enamel with tough and resistant to wear. These two properties do not
anisotropic properties was found to have a profoundly necessarily go hand in hand. Hard structures, which are
different stress distribution under load when compared resistant to wear, are usually brittle and offer little
with models with isotropic enamel. For isotropic enamel, resistance to crack propagation, whereas tough materi-
the load path is directed through the stiff enamel shell, als tend not to be as resistant to wear. The structure of
while for anisotropic enamel, the load path is directed into human teeth appears to have evolved so as to incorporate
the dentin, as the load path follows the stiff direction of the both of these properties. The outer shell of enamel acts
enamel prisms. Thus, if enamel is indeed anisotropic, its as a hard, high-stiffness, wear-resistant shell, being
function differs greatly from that suggested in previous approximately five times harder than dentin (Caldwell
hypotheses. Enamel with anisotropic material character- et al., 1957; Craig and Peyton, 1958), but it is also
istics would provide a hard-wearing protective surface- extremely susceptible to fracture, in a direction parallel
coating while simultaneously diverting the load away to the prisms. In contrast, dentin appears to be about four
from this brittle, low-tensile-strength phase, thus reduc- times tougher than enamel (Rasmussen et al., 1976) and
ing the potential for tooth fracture. of considerably lower stiffness.
The dissipation of stress in any structure is affected
J Dent Res 72(11):1526-1531, November, 1993 considerably by its shape and the distribution of stiffness
within it. One of the fundamental laws of stress analysis
Introduction. is that stiff materials attract stress. This fact is illustrated
in Fig. 1, where the material to the right has a stiffness
Several studies have attempted to gain some understand- eight times greater than that of the material to the left.
ing of the process of stress dissipation in teeth and how it When a vertical load is applied at the junction between the
is affected both by the design of restorations, and by the two materials, it can be seen that the bulk of stress is
use of restorative materials ofwidely different behavioral carried by the stiffer material. Thus, the following fea-
properties, with finite element techniques (Yettram et al., tures would appear to be desirable if human teeth were to
1976; De Vree et al., 1984; Williams et al., 1987; Khera et be optimally "designed" with regard to stress distribution:
al., 1988; Morinet al., 1988; van Noortet al., 1988). Finite
element stress analysis is a form of mathematical model- (1) The hard, stiff, but relatively weak enamel should be
ing which examines the deformations under load of a used as little as possible for dissipation of stress.
model composed of a meshwork of elements with given (2) The tougher dentin should be used as much as pos-
mechanical properties. While the structure of enamel has sible for stress dissipation.
been elucidated, from which it may be assumed that (3) Tensile stresses should be minimized in any part of
enamel has elastic properties which are highly anisotro- the tooth, especially in the enamel because of its poor
pic (Yettram et al., 1976), the biomechanical reasons for tensile strength.
such an arrangement have not been adequately addressed
(Howard et al., 1990; van Noort et al., 1991). Yet, it is vital Yettram et al. (1976) suggested that, since enamel is
considerably stiffer than dentin, the masticatory forces
Received for publication October 27, 1992 tend to flow around the 'enamel cap' of the tooth and that,
Accepted for publication June 25, 1993
RHC's laboratory is funded by the UK Science and Engineering as a consequence, the enamel near the enamel-cementum
Research Council, The Medical Research Council, the Erna and junction is very highly stressed. They found that the
Victor Hasselblad Foundation, Tektronix UK Ltd, and the Oral and anisotropic characteristics of enamel had a negligible
Dental Research Trust. effect on this stress concentration. Since this result is
1526
Downloaded from jdr.sagepub.com by guest on July 15, 2011 For personal use only. No other uses without permission.
Vol. 72 No. I I ANISOTROPIC BEHAVIOR OF ENAMEL
counter-intuitive-the prismatic, anisotropic nature of
enamel surely serving some adaptive, and presumably
mechanical purpose-in this study, we re-examine the
effects of the anisotropic nature of enamel on the mode of
stress distribution in a tooth under load. Two different
finite elements analysis software programs were used in
order to reduce the possibility that the method of analysis
contributes substantially to the result obtained.
Materials and methods.
Properties of enamel and dentin.-A value of 10 GPa was
used in this study for the elastic modulus of dentin. For
isotropic enamel, a value of 80 GPa was used. For aniso-
tropic enamel, a value of 80 GPa was again used for the
elastic modulus in the direction of the prisms, which are
assumed to be perpendicular to the DEJ. It is also as-
sumed, as did Yettram et al. (1976), that low values of
enamel stiffness were obtained from measurements taken
across the prism direction, where stiffness should be
lower. Thus, in the transverse direction, the same modu-
lus is used as that for dentin, giving an enamel anisotropy
ratio of 8:1, which is probably at the upper end of the
possible range. Since the actual ratio of enamel anisotropy
is not known, it was decided that the effects of different
ratios should be compared. By use of the elastic moduli
presented in the Table, ratios of 4:1 and 2:1 were therefore
also employed.
Haines (1968) found the Poisson's ratio of enamel to be
0.3, although the actual effect of this ratio in two-dimen-
sional modeling is negligible (Yettram et al., 1976). In
I/ Applied load
.. +
. %
ss * t ,t
s-ss-I i
, t
f 4ft ft
J %-
m- m a.. *~ 11 II u mIIIIIZ.
-- ]11zl*
0^**. .^ I 1I1 I % .,
Fig. 1-The stress distribution, as represented by the principal
stress vectors, in a combined low- and high-stiffness material result-
ing from a vert *1 Ied loadA the interface between the two
materials. (The low-stiffness material is on the left.)
Downloaded from jdr.sagepub.com by guest on July 15, 2011 For personal use only. No other uses without permission.
15;27
order to find out how important Poisson's ratio is in two-
dimensional modeling, we substituted different values of
Poisson's ratio in the tooth model. Varying these values
did not have an effect on the stress distribution, and
Yettram's assumption was found to be correct.
The finite element system used.-This work has been
undertaken by two research groups with similar interests
in the structural performance of teeth. Two different finite
element systems were used for the stress analysis: the
commercial code ANSYS, and the in-house research code
TOMECH, developed at the University of Sheffield. Al-
though the results in this paper are presented primarily in
terms ofthe TOMECH output, the research was carried out
in parallel-the Liverpool group using ANSYS, the
Sheffield group TOMECH-so that the results for the two
finite element systems could be compared and verified at
each stage. Such verification is a vital element of our study,
since our results disagree with those of previous work
(Yettram et al., 1976), and the disagreement might other-
wise be attributed to inadequacies of our analytic method.
Effect of anisotropy on stress distribution in a simple
model-.Before dealing with the complexity of a tooth, we
tested a simple model containing anisotropic elements, in
order to observe how anisotropy affects stress dissipation.
The model consisted of a semicircle of dentin supporting
an arch of enamel. This is an idealized representation of
a tooth cusp, where the load is applied in a direction
parallel to the stiff direction of the anisotropic enamel.
The model is shown in Fig. 2, which indicates the point of
load application and the direction of anisotropy assigned
to the individual elements making up the enamel. The
model was constructed from 8-noded isoparametric ele-
ments by use of TOMECH. A corresponding model was
produced by use of ANSYS. A vertical load of 10 N was
applied at the apex, and the reaction forces in the Y-
direction (R ) at the fixed boundary (XX) of the model
were compuied for both isotropic and anisotropic enamel.
% % v Effect of anisotropy in a tooth model.-A finite element
model of a buccal-lingual section through a mandibular
second premolar, subjected to an occlusal load of 10 N, was
produced by TOMECH and ANSYS. The shape and charac-
teristics ofthe tooth were similar to those used by Yettram
et al. (1976). The meshed model produced with TOMECH is
shown in Fig. 3, which also indicates the loading point and
the direction of anisotropy in the enamel. The force was
*
applied to the model at the tip of the lingual cusp.
- a O Enamel { Applied load
Q Dentin
* direction vectors
x
Fig. 2-Model of a semicircle showing the load application point
and the fixed boundary along XX. The direction vectors show the
direction of the prisms within each of the enamel elements.
1528 SPEARS et al. J Dent Res November 1993

Direction vectors
has a low elastic modulus until this process is finished.
Nevertheless, this kind of deformation occurs only at the
very onset of loading and can therefore be neglected when
a reasonable amount of force is applied to the tooth.
Thus, to summarize, the assumptions made in the
production of this model were that:

Lingual (1) the direction ofenamel anisotropy is perpendicular to

the EDJ;
(2) the stiffness of enamel perpendicular to the prism
A- A direction is similar to that of dentin;
(3) the fixed boundary in the tooth model can be set along
Y
the x-axis at about a third of the way down the root;
(4) loads are applied vertically along the stiff direction of
t-v-. x the enamel prisms;
(5) the transfer of stress around the tooth is instanta-
neous, thus enabling the model to be studied by
elastic static analysis;
Fig. 3-Model of a buccal-lingual section of a mandibular second (6) the Poisson's ratio does not significantly affect the
premolar. AA represents the cervical section analyzed in this study. stress distribution in two-dimensional analysis;
(7) apart from the different orientation of enamel prisms,
The direction vectors indicate the prism direction within each of the
enamel elements. all the properties of enamel are identical at all loca-
tions in the enamel; and
During mastication of soft food items, the applied load (8) dentin is isotropic.
in a real situation would rarely be a point force, since the
food particles, when comminuted, would be spread over an Results.
area of the tooth, thus exerting a pressure rather than a
force. However, in occlusal loading due to tooth-to-tooth The principal stress vectors for the semi-circular model
contact, or when the food substrate includes small, hard are presented in Figs. 4a and b for both the isotropic and
elements, the load is virtually a point force. In this anisotropic cases. In isotropic enamel, the load path is
situation, stress in the tooth is likely to be much more directed through the stiff enamel arch, virtually as if the
damaging. Therefore, analysis of a tooth model loaded by dentin were absent, whereas for the anisotropic enamel
a point force will yield results more representative of the the loading path is into the dentin, and relatively little
high-risk loading situations, which will attract the greater load is being carried by the enamel at the base ofthe arch.
selective influences. The "rigid displacement" which we This is all the more apparent from the reactions at the
have assigned at the level of the tooth root (that is, its fixed fixed boundary XX, shown in Fig. 5. Anisotropy is a
and undeformable character) is, of course, not a true beneficial property of enamel (in this illustrative model)
representation of a real tooth, but it is far enough away in that it "carries" stress toward the dentin. It produces a
from the crown not to affect the crown's overall response, stress distribution that dissipates stress more effectively
which in this study is the matter of interest. than would isotropic enamel.
It has been shown by Granath and Svensson (1991) Published comparison between the performances of
that below loads of 20 N, even on teeth with stress- different stress analysis codes is generally rare, and we
concentrating cavities, enamel and
* 5!
S dentin behave elasti- believe that this is the first instance of it in the field of
cally and linearly. However, it has also been shown, by dental research. An example of this comparison for aniso-
Haines (1968), that enamel has very low stiffness at the tropic enamel in the semicircular model is presented in
onset of loading at low loads. This is because water held Fig. 5, showing the calculated reactions at the fixed
between prisms permeates out at low loads, so that enamel boundary for the two finite element codes for the case of

l Applied load

0i X

s
s

Fig. 4-Stress distribution as represented by the principal stress vectors in the semicircular model for (a) isotropic and (b) anisotropic
enamel.
Downloaded from jdr.sagepub.com by guest on July 15, 2011 For personal use only. No other uses without permission.
Vol. 72 No. 1 1 ANISOTROPIC BEHAVIOR OF ENAMEL
anisotropic enamel. This shows that the solutions to the
same problem, formulated and solved independently at
Liverpool and Sheffield, produced results that differ from
each other only slightly. The concordance of two analyses
is a necessary check, given the importance ofour findings,
installing a greater degree of confidence in the results
presented. The variation between the two sets of results
can be ascribed to small distinctions in the meshing
produced by the two mesh generators, and to differences
in the method ofassigning the anisotropy to the individual
elements.
The results for the tooth model analyzed by TOMECH
are presented in Fig. 6, where the stresses in the Y-
direction (6yy) are plotted for the isotropic and anisotropic
enamel along the line AA, indicated in Fig. 3. As with the
semicircular model, the results were found to be virtually
the same when ANSYS was used. Regions 1 and 5 repre-
sent the buccal and lingual enamel, respectively, and
regions 2, 3, and 4 lie within the area representing the
dentin. There is twice as much compressive stress in the
isotropic model as in the anisotropic model in the enamel
at the buccal surface of region 1. This indicates that the
anisotropic model has not incurred high compressive
stresses in the enamel around the cervical region. The
anisotropic model appears to have dissipated the stress
gradually across the EDJ into the dentin. The conse-
quence is that the forces flowing down the tooth do not
cause high stresses in the enamel near the gingival
margin-the point at which the cross-sectional area ofthe
enamel is reduced. The anisotropic model also seems to
avoid interfacial stress of high potential at the EDJ and
deforms in this region with the dentin, as if a single
material, with a gradual reduction of stress across the
junction. In contrast, in the isotropic model, there is a
rapid change of stress across the EDJ, which may cause
stress-related problems along the junction. In fact, it is
difficult to identify the location ofthe EDJ from the stress
pattern in the anisotropic case. In the buccal dentin
- ANSYS (anisotropic)
- TOMECH (anisotropic)
4 TOMECH (isotropic)
1.2
1.1
1.0
0.9
008
0.7
0
111 0.6
cr
0.5
0.4
0.3
0.2
0.1
0.0
x
9 -8 -7 -6 -5 -4 -3 -2 -1 01 2 3 4
Fig. 5-Comparison of the vertical reaction R (N) at the fixed
boundary XX of the semicircular model for isotropic and anisotropic
enamel using TOMECH and anisotropic enamel using ANSYS.
Downloaded from jdr.sagepub.com by guest on July 15, 2011 For personal use only. No other uses without permission.
1529
(region 2), the stress in the isotropic model has fallen
significantly, whereas in the anisotropic model the stress
has indeed fallen, but only gradually. The anisotropic
model has therefore distributed more of the forces into the
dentin than has the isotropic model. Thus, the stress
levels within the buccal side of dentin are significantly
higher in the anisotropic model than in the isotropic
model. However, since the dentin is a tougher material,
this would not cause a problem, and may be regarded as
a beneficial consequence of enamel anisotropy. Near the
pulp (region 3), the oy stress in the two models is close to
zero. The different properties of enamel have not signifi-
cantly affected stress in dentin around the pulp chamber.
In the lingual dentin (region 4), the stress in both models
becomes tensile. Here, the stress distributions in both
models exhibit features similar to those in region 2,
although the stress is now tensile and not compressive.
The stress in the anisotropic model is higher than that in
the isotropic model, but both are still very low. Toward the
EDJ, the stress in dentin rises more steeply in the isotro-
pic model than in the anisotropic model. Finally, in the
lingual enamel (region 5), there are tensile stresses in
both anisotropic and isotropic enamel. However, the ten-
sile stress in the isotropic enamel is almost twice as high
as that in the anisotropic enamel.
The use ofdifferent ratios of anisotropy appeared to have
little effect on the stress distribution over the range investi-
gated. The biggest change in stress distribution thus arose
from the introduction ofanisotropy, whatever its level. This
may be because minimizing shear stresses across the EDJ is
more important, and has more significant effect on the
stress distribution, than the actual stiffness values used. It
should, however, be borne in mind that these findings apply
only to the particular situation investigated. More marked
differences between the mechanical behavior of enamel
with different levels of anisotropy may arise under different
loading conditions, or for different tooth models.
Discussion.
Human enamel is composed of prisms, which are built up
in an orderly fashion. These prisms run roughly parallel
from the EDJ to the tooth's surface. Although the enamel
prisms across a section of tooth do actually appear as
curved lines, particularly around the cuspal region (Luke
4 y(N)
2 region
21
A
.2
-*4
*6
*8
5 6 7 8 9 *10
Fig. 6-The a stresses in a second mandibular premolar along
a horizontal line kA indicated in the mesh model of Fig. 3 for isotropic
and anisotropic enamel.
1530 SPEARS et al.
TABLE
THE ELASTIC MODULI OF ENAMEL AND DENTIN USED IN THIS STUDY
Isotropic
Eea (GPa) 80
Eet(GPa) 80
Ed (GPa) 10
Anis. ratio (Eea/Eet) 1:1
Eea = the elastic modulus of enamel in the direction parallel to the prisms.
Eet the elastic modulus of enamel in the direction perpendicular to the prisms.
Ed = the elastic modulus of dentin.
Anis. ratio = the anisotropic ratio for enamel.
and Lucas, 1983), this effect would be very complicated to
model, and it was assumed that the prisms followed a
straight-line path from the EDJ to the enamel surface. It
has also been assumed that the general direction of the
prisms is perpendicular to the EDJ. The only previous
work on the subject (Yettram et al., 1976) made the same
assumptions. This prismatic structure is thought to cause
the enamel to behave as an anisotropic material
(Rasmussen et al., 1976), stiffer along the directions of
these prisms than perpendicular to the prisms (Lees and
Rollins, 1972; Pfretzschner, 1986). Although tubules are
present in dentin, its structure is less-well-ordered than
that of enamel. Lower variations in measurements of the
elastic modulus (Waters, 1980) and ultrasonic velocity
through dentin (Ng et al., 1989) suggest that, in contrast
to enamel, dentin can be assumed to behave isotropically.
A major problem in tooth modeling is determination of
the values to be assigned to the elastic moduli ofthe dentin
and the enamel. For enamel, measurements of the elastic
modulus range from 9.6 to 84.2 GPa (or GN/m2) and for
dentin, from 9.6 to 21 GPa, by both mechanical (Craig and
Peyton, 1958; Stanford et al., 1960; Craig et al., 1961;
Bowen and Rodriguez, 1962) and ultrasound techniques
(Lees, 1968; Gilmore et al., 1970; Lees and Rollins, 1972).
This work seems to have taken little account of the direc-
tion of the enamel prisms. Further, studies of the proper-
ties of enamel show different values of stiffness for occlu-
sal, cuspal, and side enamel (Stanford et al., 1960). Such
differences may be due to inconsistent prism orientation
along the line of measurement. It may well be that if
measurements were taken along consistent paths with
respect to prism direction, then considerably more consis-
tent values of stiffness would be found in the enamel.
Therefore, there are as yet no reliable data on the elastic
moduli for enamel and dentin in directions parallel, and
perpendicular, to the enamel prisms or to the dentinal
tubules. For enamel, we have assumed that the high
values reported for the elastic modulus were derived from
measurements which happen to have been taken along the
direction ofthe prisms, but low values from measurements
across these prisms. Although dentin is structurally aniso-
tropic, there appears to be little variation in the elastic
modulus. It is important to note that, since reliable data on
material properties are unavailable, and since a two-
dimensional modeling process is used, emphasis in this
study is placed on a relative comparison of the stress
Downloaded from jdr.sagepub.com by guest on July 15, 2011 For personal use only. No other uses without permission.
J Dent Res November 1993
Anisotropic Anisotropic Anisotropic
8 40 40
10 10 20
10 10 20
8:1 4:1 2:1
distributions in models with different material properties,
and not on any absolute distinctions in measurements.
This study shows that enamel anisotropy has a signifi-
cant effect on stress dissipation in teeth. With the excep-
tion of Yettram et al. (1976), most workers using the finite
elements approach have used isotropic material proper-
ties throughout their analyses, even though such an
assumption may be inadequate when enamel, with its
prismatic structure, is involved. Yettram et al. (1976)
could show no significant difference in the isotropic and
anisotropic case, which is at odds with our results. Al-
though the reason for this is unclear and may be associ-
ated with the finite element code used, the fact that both
TOMECH and ANSYS gave similar results shows the
importance of using multiple techniques as an essential
check on the accuracy of such models.
In our opinion, the ordered arrangement of enamel
prisms serves an important functional purpose: Simply, the
anisotropy helps to dissipate occlusal loads more effectively.
A major function of enamel anisotropy is to reduce tensile
and compressive stresses that would otherwise occur in the
enamel, and thereby reduce the risk of enamel fracture,
particularly in the cervical region. Toward the gingival
margin, the cross-sectional area of enamel decreases, re-
sulting in high tensile and compressive stresses in the
isotropic model, because the load path is channeled through
the high-stiffness isotropic enamel. Since enamel has very
poor tensile properties (Haines, 1968), tensile stresses have
been implicated in the creation ofcervical caries lesions (Lee
and Eakle, 1984). In the anisotropic model, the stresses in
the enamel are much more readily dissipated into the
dentin, and are considerably lower compared with those in
isotropic enamel. Measurements of enamel hardness
(Caldwell et al., 1957; Craig and Peyton, 1958), together
with the findings of our study, suggest that the function of
prismatic enamel is to give the tooth a hard-wearing protec-
tive surface coating, while simultaneously diverting the
load away from this brittle, low-tensile-strength phase, and
thus reducing the potential for enamel fracture.
Further work needs to be done to generate reliable data
on enamel stiffness with regard to prism orientation. Such
data on the behavioral properties of materials are basic
requirements for finite element modeling and will in-
crease the effectiveness of finite element analysis in
dentistry. Notwithstanding the lack of reliable data on
enamel stiffness characteristics, this study has shown
Vol9. 72 No2.II ANISOTROPIC BEHAVIOR OF ENAMEL
that the anisotropic characteristics of enamel cannot be
ignored in finite element modeling of teeth.
Acknowledgments.
We thank the anonymous referees for their valuable
comments.
REFERENCES
Bowen RL, Rodriguez MS (1962). Tensile strength and modulus of
elasticity of tooth structure and several restorative materials. J
Am Dent Assoc 64:378-387.
Caldwell RC, Muntz ML, Gilmore RW, Pigman W (1957). Micro-
hardness studies of intact surface enamel. JDentRes 36:732-738.
Craig RG, Peyton FA (1958). Elastic and mechanical properties of
human teeth. J Dent Res 37:710-718.
Craig RG, Peyton FA, Johnson DW (1961). Compressive properties
of enamel, dental cements and gold. J Dent Res 40:936-945.
De Vree JHP, Peters MCRB, Plasschaert AJM (1984). The influence
of modification of cavity design on the distribution of stresses in
a restored molar. J Dent Res 63:1217-1220.
Gilmore RS, Pollack RP, Katz JL (1970). Elastic properties ofbovine
dentine and enamel. Arch Oral Biol 15:787-796.
Granath L. Svensson A (1991). Elastic outward bending of loaded
buccal and lingual premolar walls in relation to cavity size and
form. Scanzd J Dent Res 99:1-7.
Haines DJ (1968). Physical properties of human teeth enamel and
enamel sheath material under load. JBiomechan 1:117-125.
Howard IC, Van Noort R, Cardew GE, Noroozi S (1990). The influ-
ence of enamel anisotropy on the stress distribution in teeth
(abstract). J Dent Res 69:956.
Khera SC, Goel UK, Chen RCS, Gurusami SA (1988). A three-
dimensional finite element model. Oper Dent 13:128-137.
Lee WC, Eakle WS (1984). Possible role of tensile stress in the
etiology of cervical lesions of teeth. J Prosthet Dent 52:374-380.
Downloaded from jdr.sagepub.com by guest on July 15, 2011 For personal use only. No other uses without permission.
View publication stats
1531
Lees S (1968). Specific acoustic impedance of enamel and dentin.
Arch Oral Biol 13:1492-1500.
Lees S, Rollins FR (1972). Anisotropy in hard dental tissues. J
Biomechan 5:557-566.
Luke DA, Lucas PW (1983). The significance of cusps. J Oral
Rehabil 10:197-206.
Morin DL, Cross M, Voller VR (1988). Biophysical stress analysis
of restored teeth: modelling and analysis. Dent Mater 4:77-84.
Ng SY, Payne PA, Cartledge NA, Ferguson MWJ (1989). Determi-
nation of ultrasonic velocity in human enamel and dentine.
Arch Oral Biol 34:341-345.
Pfretzschner HU (1986). Structural reinforcement and crack propa-
gation in enamel. In: Teet'h revisited: Proceedings of the VIIth
international symposium ofdental morphology. Paris: Memoires
de Museum Nationale D'Histoire Naturelle, 133-143.
Rasmussen ST, Patchin RE, Scott DB, Heuer AH (1976). Fracture
properties of human enamel and dentine. J Dent Res 55:154-
164.
Stanford JW, Weigel KV, Paffenbarger GC, Sweeney WT (1960).
Compressive properties of hard tooth tissues and some restor-
ative materials. J Am Dent Assoc 60:746-756.
Van Noort R, Howard IC, Cardew GE (1988). A study of the
interfacial shear and tensile stresses in a restored molar tooth.
J Dent 16:286-293.
Van Noort R, Howard IC, Cardew GE (1991). Whiy is enamel
structurally anisotropic (abstract). J Dent Res7O:455.
Waters NE (1980). Some mechanical and physical properties of
teeth. In: Vincent JVF, Curry JD, editors. The mechanical
properties of biological materials. Cambridge: Cambridge
University Press, 99-134.
Williams KR, Edmundsen JT, Rees JS (1987). Finite element
stress analysis of restored teeth. Dent Mater 3:201-206.
Yettram AL, Wright KWJ, Pickard HM (1976). Finite element
stress analysis of the crowvns of normal and restored teeth. J
Dent Res 55:1004-1011.