Journal of

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

Fatigue of Dental Ceramics in a Simulated Oral Environment

R. Morena, G.M. Beaudreau, P.E. Lockwood, A.L. Evans and C.W. Fairhurst
J DENT RES 1986 65: 993
DOI: 10.1177/00220345860650071901
The online version of this article can be found at:
http://jdr.sagepub.com/content/65/7/993

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/65/7/993.refs.html

>> Version of Record - Jul 1, 1986

What is This?

Downloaded from jdr.sagepub.com at MCGILL UNIVERSITY LIBRARY on October 28, 2014 For personal use only. No other uses without permission.

Fatigue of Dental Ceramics in a Simulated Oral Environment

R. MORENA, G.M. BEAUDREAU, P.E. LOCKWOOD, A.L. EVANS, and C.W. FAIRHURST
Dental Materials Division, School of Dentistry, Medical College of Georgia, Augusta, Georgia 30912

Fatigue in ceramics refers to the subcritical growth of cracks, aided

by the combined influence of water and stress. The dynamic fatigue
(constant stressing rate) method was used to obtain subcritical crack
growth parameters for three dental ceramics: a feldspathic porcelain,
an aluminous porcelain, and a fine-grain, polycrystalline core material. The constant stressing rate experiments were carried out at
37"C for all three ceramics in distilled water, and, for the feldspathic
porcelain, in artificial saliva as well. Considerable differences were
found in the value of the crack growth exponent (n) among the three
ceramics. The feldspathic porcelain was lowest in n-value, while the
fine-grain ceramic had the highest n-value. No differences were found
for the feldspathic porcelain with respect to n measured in water and
in the artificial saliva. Lifetime prediction curves in 37C water, constructed from the n-values and inert strengths, showed that fatigue
failure within five years is a good possibility for feldspathic porcelain
specimens at stress levels which can reasonably be anticipated to
occur in the oral environment. Little likelihood of failure was perceived for the fine-grain ceramic. The aluminous porcelain was intermediate between these two materials with respect to failure
probability.

J Dent Res 65(7):993-997, July, 1986

Introduction.
It has long been known that the presence of water will degrade
the strength of silicate glasses and many other ceramic materials (Shand, 1958; Mould, 1959). Exposure to an aqueous
environment has also been found to affect the mechanical performance of dental ceramics. Sherrill and O'Brien (1974) demonstrated that the fracture stress of feldspathic and aluminous
dental porcelains decreased by nearly 30% when the samples
were broken in water. Southan and Jorgensen (1974) showed
that the ability of a dental porcelain to sustain a static load in
water decreased as the duration of load application increased.
The effect of aqueous exposure and other aspects of dental
porcelain mechanical behavior have been the subject of an
extensive review by Jones (1983), who also lists decreasing
strength with decreasing stress/strain rate as further evidence
for the detrimental role played by water.
The process of strength degradation of ceramics in aqueous
environments is termed fatigue and is believed to be caused
by a stress-corrosion process involving the stable growth of
small, pre-existing flaws (Hillig and Charles, 1965; Michalske
and Freiman, 1983). The effect of water is so pronounced that
stress-corrosion has been observed in silicate glasses at moisture levels as low as 0.017% relative humidity (Wiederhom,
1967). Failure, generally after long loading times and often at
stresses substantially less than the measured short-term fracture
stress, will ensue once these flaws attain a size equivalent to
that needed to initiate fast fracture.
The oral environment would appear to have all the factors
necessary for fatigue failure to occur in ceramic-based dental
prostheses. Water is, of course, the primary chemical species
Received for publication October 31, 1985
Accepted for publication April 10, 1986
This investigation was supported in part by USPHS Research Grant
DE07045 from the National Institute of Dental Research, National
Institutes of Health, Bethesda, MD 20892.

in saliva. A dental restoration would also be exposed to water

from a cementing agent as well as from the dentin tubules.
Stresses, both masticatory-related and also associated with
thermal expansion mismatches between the various components of the restoration, would be present to provide the driving
force for fatigue. Flaws in a dental ceramic may be introduced
by grinding during occlusal adjustment or may be intrinsic and
related to microstructural features (Jones and Wilson, 1975;
Corbitt et al., 1985). Finally, the range of temperatures that
would be encountered by a restoration during service in the
oral cavity is sufficiently elevated from ambient as to exacerbate any slow crack growth process (Ritter et al., 1985).
Failure of dental ceramic restorations caused by fatigue does
appear to be a possibility, as judged by several studies reported
by Lehman (1967) on long-term failures of feldspathic porcelain jacket crowns. McLean (1983) lists a failure rate of
15.2% within three years following placement in the mouth
for platinum-bonded, aluminous porcelain crowns in molar sites.
Similar studies documenting failures for porcelain-fused-to-metal
(PFM) crowns and bridges could not be found, although Warpeha and Goodkind (1976) list several types of persistent clinical failures observed with PFM crowns that were all associated
with the porcelain component. Delayed failures which occurred up to six weeks after firing have been observed by
Walton and O'Brien (1985) for model PFM crowns and were
attributed to a probable fatigue interaction. The large number
of commercial in vivo porcelain repair materials currently
available provides indirect testimony for the existence of longterm failures in PFM restorations (Ferrando et al., 1983).
Although the several studies cited have given evidence for
the fatigue susceptibility of various dental porcelains, no data
could be found in the literature relating to the fundamental
material parameters which determine the kinetics of slow crack
growth for dental ceramics. In the past 10 years, the discipline
of fracture mechanics has become firmly established for analyzing the fracture properties of ceramics and predicting their
fatigue behavior (Wiederhorn, 1974; Ritter, 1978). Slow crack
growth is described in terms of a fracture mechanics analysis
by a fundamental parameter the crack growth exponent, n
which is a material constant for a given environment. Assuming a power law expression for subcritical crack growth
(i.e., the rate of crack growth x K, where K, is the opening
mode stress intensity factor), the failure time of a ceramic
component experiencing slow crack growth-induced fatigue is
given (Pletka and Wiederhorn, 1982) by:
,

BSn -2t-an

(1)

failure time under a constant applied stress, Ua;

constant dependent on n, and also on crack geometry and the fracture toughness of the material;
and
S = inert or initial strength in the absence of subcritical crack growth.
Note that of two materials with comparable values of initial
strength that are candidates for a given application in which a
stress must be sustained without failure over a long time period, the material with the higher n value will be more likely
to survive.

where tf
B

Downloaded from jdr.sagepub.com at MCGILL UNIVERSITY LIBRARY on October 28, 2014 For personal use only. No other uses without permission.

993

994

J Dent Res July 1986

MORENA ET AL.

It was the objective of this study to determine the crack

growth exponent using standard fracture mechanics tests for a
number of restorative ceramics, including a polycrystalline core
material. Additionally, predicted lifetime curves, constructed
from the fatigue data using Eq. (1), will be presented so that
the survivability of various dental ceramics can be compared
under different levels of applied stress and fatigue conditions.
In general, lifetime prediction curves are constructed based on
values of n and B measured in controlled laboratory fatigue
tests, and with S, the inert strength of the component, obtained
on a geometry as representative as possible of the actual enduse.

Materials and methods.

The fatigue data were obtained by the dynamic or stressing
rate technique, in which fracture stress in the environment
under study is measured as a function of stressing rate. The
materials used in the study consisted of a feldspathic porcelain
(Ceramco Gingival, Ceramco, Inc., East Windsor, NJ), an
aluminous porcelain (Vitadur-N, Vident Corp., Baldwin Park,
CA), and a fine-grain polycrystalline "shrink-free" ceramic
(CerestoreT, Johnson & Johnson, East Windsor, NJ). These
latter two materials are intended for use as the core component
in all-ceramic replacement crowns. All materials were obtained
from regular commercial lots supplied by the vendor.
Specimens of both the feldspathic and aluminous porcelains
were vibration-condensed into thin disks by means of a split
brass mold. Unfired dimensions of these disks were typically
15 mm in diameter and 2.0 mm in thickness. Following removal from the mold, the specimens were permitted to air-dry
for at least two hours prior to being fired. Firing was carried
out in a commercial porcelain furnace (System 8, J.M. Ney
Co., Bloomfield, CT), following the firing schedule recommended by the respective porcelain manufacturer.
The fine-grain, polycrystalline core material was processed
into thin disks by a variant of the lost wax technique, with
commercial equipment supplied by the vendor. The time/temperature schedule recommended by the manufacturer was followed in firing the test specimens using the programed furnace
supplied as part of the processing unit.
Following firing, all specimens were ground and polished
on both surfaces using a sequence of steps culminating in 15pum diamond paste on a nylon cloth-backed bronze wheel. One
surface of each specimen received an additional polishing sequence through 1-pLm diamond paste. Final dimensions of the
polished disks for all three ceramics were typically 12.5 mm
in diameter and 1.0 mm in thickness. For all three materials
in this study, a sample size of 10-12 specimens was used for
each stressing rate and for the determination of inert strengths.
All specimens were tested such that the finest polished surface
was the tensile surface during fracture.
Prior to the strength-testing procedure, a single Vickers indentation was made in the center of the 1-pRm diamond polished surface of each specimen of the aluminous porcelain and
the fine-grain, polycrystalline core material. This was done to
ensure that a single dominant flaw would be present in the
material to minimize any variation in strength caused by the
presence of random handling damage. The load selected for
the indentation (4.9 N for the aluminous porcelain, 9.8 N for
the fine-grain ceramic) was the minimum necessary to produce
a significant decrease in strength (p < 0.05 by Student's t test)
relative to that of unindented specimens. No such indentation
was made for the feldspathic porcelain, since a previous study
(Corbitt et al., 1985) had indicated that the strength of this
material was controlled by large, intrinsic microcracks.
Evans (1974) has shown that the dependence of strength on

stressing rate, caused by subcritical crack growth, can be described by:

af

where

afo

(&/6 )1/n+ l

(2)

of = fracture stress determined at a given

stressing

rate,

(Ufo = a constant for a given environment,

& = stressing rate, and
&O = a normalizing constant (1 MPa/s).
The parameters n and coy were obtained from the stressing
rate experiments by fitting the dynamic fatigue data to a ln-ln
relationship, and carrying out a linear regression analysis. All
data points were used in these calculations. The value of n
obtained from Eq. (2) was corrected for the two materials
which had been indented by multiplying by a constant term,
1.31, to correct for residual stress effects associated with the
indentation (Lawn et al., 1981). The standard error of n was
determined from a statistical analysis of the linear regression,
using the approach derived by Pletka and Wiederhorn (1982).
The residual stress correction factor was included in this error

analysis.
The B term in Eq. (1) can be obtained (Ritter, 1978) from:

Un+lIi(3)
n +l()
where Si = inert or initial strength of laboratory specimens
in the absence of subcritical crack growth.
The strength values at each stressing rate were obtained with
a constant displacement rate mechanical testing machine (Model
TT-B, Instron Corp., Canton, MA). Test specimens were broken in bi-axial flexure by means of the piston-on-3-ball technique described by Wachtman et al. (1972). This technique
was selected since it eliminates spurious edge failures (which
occur in flexure tests) associated with stress concentrations at
contact points. This is because tensile stress in the bi-axial
flexure apparatus decays rapidly with increasing radial distance
from the center of the disk. The disks were concentrically
supported on three ball bearings (each 1.6 mm in diameter)
that were evenly spaced around a support circle having a radius
of 3.5 mm. The specimen-holder is shown in Fig. la. The
specimens were loaded to failure by a compressive load applied
by a right circular cylinder with a diameter of 1.6 mm. The
fracture stress was calculated (Wachtman et al., 1972) from
the load at failure by:
B

3P

of

l + 2 In
47r t2 (I+v)Lb
+

I+V
I+
v

lI

b2

S2a2

4a2l
2

(4)

where P = failure load

t = thickness of specimen
D = diameter of specimen
v = Poisson's ratio of specimen (assumed to be 0.25)
a = radius of support circle
b = radius of right circular cylinder
All stressing rate tests were carried out in a simulated oral
environment consisting of distilled water maintained at 370C
+ 0.50C by a thermostatically-controlled liquid circulator (Model
FE2, Haake Instruments Co., Saddle Brook, NJ). Specimens
were allowed to reach thermal equilibrium in the liquid meDownloaded from jdr.sagepub.com at MCGILL UNIVERSITY LIBRARY on October 28, 2014 For personal use only. No other uses without permission.

Vol. 65t No. 7

v~ ~

FATIGUE OF DEINTAL CERAMICS

995

that had received a relatively coarse final polish of 30 vm in

order to duplicate the actual surface condition of a ceramic
component in a dental restoration. All specimens for inert
strength determination were chosen from the same sample population from which test specimens for the stressing rate experiments were selected.

Results.

The dynamic fatigue curves are shown in Fig. 2 for the three
tested in water. Dynamic fatigue curves are compared
for the feldspathic porcelain evaluated in both water
and the artificial saliva environments. The error bars shown in
these Figs. for fracture stress correspond to the Mean + 1
S.D. No error bars are shown for stressing rate, since the
coefficient of variation (S.D./mean) was typically 5% or less.
Also shown in both Figs. is the coefficient of determination,
r2, for each test situation.
The results of the linear regression analyses on the dynamic
fatigue data are listed in Table 1. The value of the crack growth
exponent, measured in distilled water, varied from approximately 14.5 for the feldspathic porcelain to nearly 81 for the
fine-grain core material. The differences in n-value for the
three dental ceramics were found to be highly significant (p <
0.01) by the Tukey Multiple Range Test. No differences (p >
0.50 by Student's t test) were found in the crack growth exponent for the feldspathic porcelain when tested in the water
and artificial saliva environments. This agrees with stress-corrosion studies of other bioceramics where no differences were
recorded in the n-value between water and other biologic media, such as Krebs-Ringer solution and bovine serum (Rockar
and Pletka, 1978). The higher strength values measured at each
stressing rate for the feldspathic porcelain tested in water relative to the artificial saliva media are mirrored in the higher
ceramics
in Fig. 3

:X

.(
i..

B3i-axial flexure apparatus: A. Specimen-holder; B. EnvironFig. I

mental chamber.

Stressing Rate, MPals

dium for 10 min prior to breaking. The environmental chamber

with the specimen-holder in place is shown in Fig. lb. Data
were also obtained with an artificial saliva solution (Salivart,
Westport Pharm., Westport, CT), held at 370C + 0.50C for a
second set of feldspathic porcelain specimens to determine
whether differences in crack growth kinetics occur between the
two aqueous media. The stressing rates were determined directly from the chart recorder of the mechanical testing machine and spanned nearly four decades, from almost 0. 10 to
approximately 85 MPa/sec.
The inert strengths used to obtain the B term in Eq. (3) and
to predict lifetime behavior from Eq. (1) were obtained in biaxial flexure following the procedure outlined by Ritter et al.
(1985). Test specimens were broken in mineral oil at high
stressing rates ( 100 MPa/sec) with a thin layer of plastic
placed within the support cycle on the tensile surface to minimize further the possibility of exposure to water. Note that
two different measures of inert strength are used in Equations
(I) and (3). Si (Eq. 3) refers to the inert strength of laboratory
test specimens, which, for this study, were finely-polished disks
that, where specified, had received indentations on their tensile
surfaces. S (Eq. 1) relates to the inert strength of engineering
specimens which, ideally, should represent the actual size,
shape, manufacturing conditions, stress state, etc., of the component (Pletka and Wiederhorn, 1982). Since the stress distribution in a restoration such as a replacement crown is complex
even in relatively simple loading situations (Farah and Craig,
1975), lifetime curves for comparison of the three dental ceramics were constructed with S measured on unindented disks

0.20

1.0
.

48

100

10

120

Fine Grain Ceramic

0(
(r 068)
=

4.6

ffi 4A4
2

100

80

(r2=0 69)

0<n

IN 37 C H20

- 42
c;4.

60 u
4.01

Feldspathic Porcelain

3.8-

40
3.6
--

-1

0
+2.0
In (Stressing Rate, MPals)

Fig. 2in water.

Dynamic fatigue

curves at

+4 0

37C for the three dental ceramics

Downloaded from jdr.sagepub.com at MCGILL UNIVERSITY LIBRARY on October 28, 2014 For personal use only. No other uses without permission.

996

J Dent Res July 1986

MORENA ET AL.

Applied Stress(1 03psi)

TABLE 1
DYNAMIC FATIGUE RESULTS
n-value
(mean + S.E.)
In Water
Feldspathic Porcelain
Aluminous Porcelain
Fine-grain Ceramic
In Artificial Saliva
Feldspathic Porcelain

1o01o Feldspathic

ufo (MPa)

44.0
78.7
105

1.8

II1

\~

Surface
Condition
72.7 + 8.2
Feldspathic Porcelain
147.9 + 18.2*
Aluminous Porcelain
Fine-grain Ceramic
131.6 8.2*
*Indented prior to being broken.

\\ \\

41.1

-1 yr.\

TABLE 2
INERT STRENGTH VALUES (Mean -+ S.D.)

20

Porcelain Aluminous Porcelain FFine Grain Ceramic

1 08

Si (MPa)
Polished through
1-pum diamond

10

I~~~~~

14.6 + 1.4
28.9 + 5.8
80.8
20.7

14.9

'I'

lI
II

II'

en

S(MPa)
Polished through
30-pum diamond
72.9 + 5.3
147.1 + 11.3
151.4 + 9.9

-1 mo.

106

h._v
(U

IIA

\~~~~~~~~~~~~~~~~~Il

day

--1

104_

Stressing Rate, MPals

\\

-1 hr.

1.0

U;
a
P

~~~~~~~~~~~~I
A

IL02

,I\ I,

II

'.

I1

102

zi
0

.2U

10

25

50

75

100

Applied Stress(MPa)

U.

Fig. 4

Lifetime curves for the three dental ceramics in 370C water.

value is comparable with that of silicate glasses (Wiederhorn

and Bolz, 1970), which are among the most fatigue-susceptible
of all ceramics. The close agreement with respect to n-value
between feldspathic porcelain and silicate glasses also pertains
Fig. 3
Dynamic fatigue curves at 37TC for feldspathic porcelain in
to other aspects of mechanical performance, such as strength
water and in artificial saliva.
and fracture toughness (Morena et al., 1984). These similarities indicate that the mechanical properties of a feldspathic
(tf0 value obtained for water. This effect is presumably due to
porcelain are controlled largely by the matrix glass and not by
a larger initial flaw size for the set of feldspathic porcelain
any crystalline phase(s).
specimens evaluated in the artificial saliva.
(b) The very high value measured for the fine-grain, "shrinkfree" ceramic core material agrees with data cited on fineThe various values of inert strength obtained on the test
materials are listed in Table 2. Lifetime curves, constructed
grain, ot-AI203 ceramics (Pletka and Wiederhorn, 1982). This
from the inert strengths and n-values using Equations (1) and
material should be relatively unaffected by the oral environment except in instances where a high tensile stress is main(4), are presented in Fig. 4. Also shown for each curve is the
90% confidence band, represented by dashed lines, and contained for a long time period.
structed according to a propagation of errors technique outlined
(c) The middle value recorded for the aluminous porcelain
by Pletka and Wiederhorn (1982).
places this material at a position intermediate between the feldspathic porcelain and the fine-grain core material with respect
to resistance to subcritical crack growth. The aluminous porDiscussion.
celain, despite a glassy content comparable with that of feldCrack growth exponents. - The widely different n-values
spathic porcelain (30-40 vol %), has a crack growth exponent
obtained in water for the three dental ceramics indicate that
significantly higher than that of a silicate glass. This indicates
these materials behave quite differently with respect to the
that direct interaction is occurring between cracks and the diseffect that an aqueous environment has on subcritical crack
persed phase (a-A1203), a behavior manifested in other fracgrowth and mechanical degradation:
ture properties as well (Morena et al., 1985).
(a) The low value obtained for the feldspathic porcelain inLifetime curves. - As noted, the lifetime curves apply strictly
dicates a material whose mechanical performance would be
to thin disks that are being subjected to a static applied stress.
appreciably adversely affected when exposed to water. This nThe slopes of these curves are governed by the n-value, which
Downloaded from jdr.sagepub.com at MCGILL UNIVERSITY LIBRARY on October 28, 2014 For personal use only. No other uses without permission.
-2.0

0
2.0
In (Stressing Rate, MPa/s)

4.0

997

FATIGUE OF DENTAL CERAMICS

Vol. 65 No. 7

is a material constant for a given environment. However, the

relative position of each curve is dependent on the inert strength,
S. Thus, if anything, the lifetime curves in Fig. 4 may well
underestimate actual fatigue degradation. This is because the
inert strength of a ceramic in a dental restoration would actually be less than that obtained on the laboratory specimens.
Design factors, such as stress concentrations at margins, and
interfacial stress - arising, for example, from differences in
thermal contraction between the components of a restoration
all would be present in a ceramic prosthesis to lower strength.
The usefulness of the lifetime curves can be illustrated by
employing them to predict the level of applied stress that would
cause failure within five years for the three dental ceramics.
Using the regression analyses for the lifetime curves, we find
that the five-year failure stress for these ceramics is:

feldspathic porcelain
aluminous porcelain
fine-grain ceramic

13.0 MPa (1900 psi)

42.1 MPa (6100 psi)
95.2 MPa (13,800 psi)

Stress of the magnitude needed to cause failure within five

years may well exist for PFM restorations in the oral environment. Finite element analyses of two-dimensional PFM model
central incisors indicate conservative estimates of masticatory
stress, ranging from 6.5 to 17.5 MPa on the upper incisor edge
(Farah and Craig, 1975; Anusavice et al., 1985). In addition,
residual mismatch stresses, arising from thermal contraction
differences between porcelain and metal in a PFM restoration,
can be substantial and attain a value as high as 50 MPa, as
shown by Anusavice et al. (1982) for model PFM crowns with
porcelain facial coverage.
The lifetime curves for the aluminous porcelain and finegrain ceramic core materials both suggest that long-term failure
should not be a problem for anterior applications. Although
no values of masticatory stress could be found in the literature
for posterior crowns, preliminary data (Anusavice, personal
communication) indicate a value of approximately 30 MPa.
Thus, should even moderate levels (15 MPa) of residual thermal mismatch stress exist for the aluminous porcelain, fatigue
failure could result. It is possible that the platinum backing
used in aluminous porcelain crowns could minimize moisture
contact, thus improving the fatigue resistance of aluminous
porcelain restorations. The fine-grain ceramic would appear to
be relatively immune to fatigue failure, even for posterior
placement. However, the steep slope of the lifetime curve could
pose a potential problem for restorations where a large tensile
value of residual mismatch stress exists (caused perhaps by a
gross compositional error for the overglaze), since rapid fatigue failure would occur once stresses much exceeded 95 MPa.
Future work. - Future investigations on fatigue will utilize
cyclic stressing to simulate mastication and will also determine
whether aging or crack tip blunting can occur at low stress
levels to mitigate, to some extent, the fatigue process.

Acknowledgments.
The authors wish to thank the Johnson & Johnson Company
for providing commercial lots of their dental ceramics. Thanks
are also expressed to Skelley Dental Arts, Inc., Aiken, SC,
for their assistance in fabricating the fine-grain ceramic core
specimens. Appreciation is also expressed to S.W. Twiggs for
assistance with statistical analyses.

REFERENCES
ANUSAVICE, K.J.; TWIGGS, S.W.; DEHOFF, P.H.; and FAIRHURST,
C.W. (1982): Correlation of Thermal Shock Resistance with Thermal Compatibility Data for Porcelain-Metal Systems, J Dent Res 61:419-422.
ANUSAVICE, K.J.; HOJJATIE, B.; and DEHOFF, P.H. (1985): Influence of
Metal Thickness in Stress Distribution on Metal Ceramic Crowns, J Dent
Res 64 (Spec Iss):246, Abst. No. 641.
CORBITT, G.V.; MORENA, R.; and FAIRHURST, C.W. (1985): Fracture
Stress of a Commercial Dental Porcelain and its Components, J Dent Res
64 (Spec Iss): 296, Abst. No. 1089.
EVANS, A.G. (1974): Slow Crack Growth in Brittle Materials Under Dynamic
Loading Conditions, Int J Frac 10:251-259.
FARAH, J.W. and CRAIG, R.G. (1975): Distribution of Stresses in PorcelainFused-to-Metal and Porcelain Jacket Crowns, J Dent Res 54:255-261.
FERRANDO, J.P.; GRASER, G.N.; TALLENTS, R.H.; and JARVIS, R.H.
(1983): Tensile Strength and Microleakage of Porcelain Repair Materials,
J Prosthet Dent 50:44-50.
HILLIG, W.B. and CHARLES, R.J. (1965): Surfaces, Stress-dependent Surface Reactions, and Strength. In: High Strength Materials, V.F. Zackay,
Ed., New York: Wiley, pp. 682-701.
JONES, D.W. (1983): The Strength and Strengthening Mechanisms of Dental
Ceramics: In: Dental Ceramics, Proceedings of the First International
Symposium on Ceramics, J.W. McLean, Ed., Chicago: Quintessence Publ.
Co., pp. 96-98 and 110-116.
JONES, D.W. and WILSON, H.J. (1975): Some Properties of Dental Ceramics, J Oral Rehab 2:379-396.
LAWN, B.R.; MARSHALL, D.B.; ANSTIS, G.R.; and DABBS, T.P. (1981):
Fatigue Analysis of Brittle Materials using Indentation Flaws, Part I: General Theory, J Mater Sci 16:2846-2854.
LEHMAN, M.L. (1967): Stability and Durability of Porcelain Jacket Crowns,
Br Dent J 123:419-426.
McLEAN, J.W. (1983): The Future for Dental Porcelain. In: Dental Ceramics,
Proceedings of the First International Symposium on Ceramics, J.W.
McLean, Ed., Chicago: Quintessence Publ. Co., pp. 26-27.
MICHALSKE, T.A. and FREIMAN, S.W. (1983): A Molecular Mechanism
for Stress Corrosion in Vitreous Silica, J Am Ceram Soc 66:284-288.
MORENA, R.; LOCKWOOD, P.E.; MACKERT, J.R., Jr.; and FAIRHURST,
C.W. (1984): Fracture Toughness and Crack-Microstructure Interaction of
a Dental Porcelain, J Dent Res 63 (Spec Iss): 234, Abst. No. 573.
MORENA, R.; LOCKWOOD, P.E.; and FAIRHURST, C.W. (1985): Fracture
Toughness of Commercial Dental Porcelains, Dent Mat (in press).
MOULD, R.E. (1959): The Strength and Static Fatigue of Glass, Glastech Ber
32K: 18-28.
PLETKA, B.J. and WIEDERHORN, S.M. (1982): A Comparison of Failure
Predictions by Strength and Fracture Mechanics Techniques, J Mater Sci
17:1247-1268.
RITTER, J.E., Jr. (1978): Engineering Design and Fatigue Failure of Ceramics.
In: Fracture Mechanics of Ceramics, Vol. 4, R.C. Bradt, D.P.H. Hasselman, and F.F. Lange, Eds., New York: Plenum Press, pp. 667-686.
RITTER, J.E.; VIGEDOMINE, M.; BREDER, K.; and JAKUS, K. (1985):
Dynamic Fatigue of Indented Soda-Lime Glass as a Function of Temperature, J Mater Sci 28:2868-2872.
ROCKAR, E.M. and PLETKA, B.J. (1978): Fracture Mechanics of Alumina
in a Simulated Biological Environment. In: Fracture Mechanics of Ceramics, Vol. 4, R.C. Bradt, D.P.H. Hasselman, and F.F. Lange, Eds.,
New York: Plenum Press, pp. 725-735.
SHAND, E.B. (1958): Glass Engineering Handbook, 2nd ed., New York:
McGraw-Hill, pp. 50-51.
SHERRILL, C.A. and O'BRIEN, W.J. (1974): Transverse Strength of Aluminous and Feldspathic Porcelain, J Dent Res 53:683-690.
SOUTHAN, D.E. and JORGENSEN, K.D. (1974): The Endurance Limit of
Dental Porcelain, Aust Dent J 19:7-11.
WACHTMAN, J.B., Jr.; CAPPS, W.; and MANDEL, J. (1972): Biaxial Flexure Tests of Ceramic Materials, J Mater 7:189-194.
WALTON, T.R. and O'BRIEN, W.J. (1985): Thermal Stress Failure of Porcelain Bonded to a Palladium-Silver Alloy, J Dent Res 64:476-480.
WARPEHA, W.S. and GOODKIND, R.J. (1976): Design and Technique Variables Affecting Fracture Resistance of Metal-Ceramic Restorations, J Prosthet
Dent 35:291-298.
WIEDERHORN, S.M. and BOLZ, L.H. (1970): Stress Corrosion and Static
Fatigue of Glass, J Am Ceram Soc 53:543-548.
WIEDERHORN, S.M. (1967): Influence of Water Vapor on Crack Propagation
in Soda-Lime Glass, J Am Ceram Soc 50:407-414.
WIEDERHORN, S.M. (1974): Mechanisms of Subcritical Crack Growth in
Glass. In: Fracture Mechanics of Ceramics, Vol. 4, R.C. Bradt, D.P.H.
Hasselman, and F.F. Lange, Eds., New York: Plenum Press, pp. 549-580.

Downloaded from jdr.sagepub.com at MCGILL UNIVERSITY LIBRARY on October 28, 2014 For personal use only. No other uses without permission.