Journal of Dental Research

http://jdr.sagepub.com/

The Competition between the Composite-Dentin Bond Strength and the Polymerization Contraction
Stress
C.L. Davidson, A.J. De Gee and A. Feilzer
J DENT RES 1984 63: 1396
DOI: 10.1177/00220345840630121101

The online version of this article can be found at:

http://jdr.sagepub.com/content/63/12/1396

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/63/12/1396.refs.html

>> Version of Record - Dec 1, 1984

What is This?

Downloaded from jdr.sagepub.com at Bobst Library, New York University on October 15, 2014 For personal use only. No other uses without permission.
The Competition between the Composite-Dentin Bond Strength
and the Polymerization Contraction Stress
C. L. DAVIDSON, A. J. de GEE, and A. FEILZER
Department of Dental Materials Science of the University of Amsterdam, Louwesweg 1, 1066 EA Amsterdam, The Netherlands
The influence of contraction stresses, developed during the poly- ment provides a flat rectangular dentinal surface (2.0 x
merization of composites, on adhesion to dentin treated with a den- 3.0 mm) for the adhesion tests. No special attention was
tin adhesive was studied for a chemically- and a light-activated given to the orientation of the dentinal tubules.
microfilled composite, in both linear and 3-D models. Bovine dentin was used, rather than human dentin,
The linear model consisted of an arrangement set up in a tensil- since preparation of the samples from bovine teeth is more
ometer in which the composites could be applied to a flat dentin
surface fixed to the stationary cross-head at one end, and mechanic- easily carried out and, according to Nakamichi et al. (1983),
ally clamped to the cross-head connected to the load cell at the it may serve as a substitute for human dentin in adhesion
other end. The increase of the bond strength was measured at tests.
different time intervals from the start of mixing and was compared The samples were washed, dried, and placed tightly into
with the developing contraction stress. Throughout the complete the sample holder, which was connected to the lower head
polymerization process, the adhesion survived the contraction stress, of a tensilometer§ . Dentin adhesive was then applied to the
which is explained by flow relaxation, which can occur sufficiently rectangular surface of the sample according to the manu-
in this configuration. facturer's instructions. Half a dumb-bell-shaped mould
In the three-dimensional model, the composites are attached was connected to the upper head of the tensilometer and
to more than two dentin walls. In this situation, flow is severely
limited, and contraction stress values can exceed the bond strength, separated from the lower part by a Teflon spacer. Sufficient
leading to separation. This was demonstrated in Class V cavities. composite material was then injected into the mould to
The shape of the cavity is considered to be of great importance in fill its space (Fig. 1). Directly thereafter, the tensilometer
conservation of the composite-dentin bond. was switched on for a short while to create a small crevice
(50-100 jum) between the upper mould and the Teflon
J Dent Res 63(12):1396-1399, December, 1984 spacer, in order to allow the upper cross-head, which is
connected to the load cell, to be displaced by the contrac-
Introduction. tion forces developed during polymerization, without
counteraction of the stationary lower cross-head. The
It has been demonstrated, in several studies, that bonding
of a composite to dentin can be achieved (Masuhara, 1982; § Zwick 1463, Uln, West Germany
Bowen et al., 1982; Asmussen and Munksgaard, 1983).
Bond strengths in the range of 1.6 -17.9 MPa have been
reported (Causton, 1984; Bowen and Cobb, 1983). The
main value of a relatively high bond strength is the con- T
servation of the marginal seal during polymerization. If,
in a cavity, the composite can shrink without creating con-
traction forces, no problems may be expected in this
respect. However, since contraction forces are always built
up in the polymerizing resin (Hegdahl and Gjerdet, 1977),
the composite-dentin bond may be disrupted. If this occurs,
the composite will be attached only to the enamel, and
there will be a gap along the dentinal wall (Davidson and
De Gee, 1984). The purpose of this study was to investi-
gate the development of the composite-dentin bond strength C-
in relation to the polymerization shrinkage stress as a func-
tion of the polymerization time.
-S
Materials and methods.
The materials used were the chemically-initiated compos-
ite, Silar*, the light-initiated composite, Silux**, and the
dentin adhesive, Scotchbond+.
Freshly extracted bovine teeth were cut into 2.0-mm-
thick slices with a water-cooled diamond sawt and ground
with waterproof SiC-paper (grit 600) to a shape which pre-
cisely fits into a sample holder (D in Fig. 1). This arrange-
Received for publication June 4, 1984
Accepted for publication September 28, 1984
*Batch 122183, A 3G1 and B 3H1, 3M Co., St. Paul, MN
**Batch 072083, univ. 3J2, and 071183, univ. 3DD3, 3M Co.,
T
St. Paul, MN Fig. 1 - Schematic representation of the device for composite-
+Batch 121983, no. 7422 L 3TT1 and 7422 R 3J2, 3M Co., dentin bond testing by tensile loading. C = half a dumb-bell-shaped
St. Paul, MN mould filled with composite, S = Teflon spacer, D = dentin sample,
Tlsomet 11-1180, Buehler, Evanston, IL and T = connections to tensilometer.
1396 Downloaded from jdr.sagepub.com at Bobst Library, New York University on October 15, 2014 For personal use only. No other uses without permission.
Vol.
VoL63 WAo
Alo. 1212~~~~~~STRESS
O\ THE COMPOSITI--E%,DI,\
T BOND 1397
19

TABLE 1
THI' COMPOSITE-DENTIN BOND STRENGTH (GrB) AND THL 4
LINEARLY-RESTRICTED POLYMERIZATION SHRINKAGE M Pa SI LUX
STRFSS (oS) AT DIFFVRENT TIMES AYTFR THiE ONSFT OF
POLYMERIZATION
7.1 bond strength

SiOar S ilux 6'

Curing time Ogn cr5 a13 itS I--

(min) (MPa) (MPa) (MPa) (MPa) 5-S'ILAFt

5.
t
0.17 0 0 1.1I(0. 1) 0.5(0.1) i
0.33 0 0 2.6(0.9) 1.1I(0. 1) 4.
0.67 0 0 4.5(1.0) 1.6(0.3) .0

1.0 0 0 4.9(0.8) i.6(o.2)

3.0 0. 7(0.0o) 0 5.1(0.9) 1.8(0.3) 31 .1
4.0 1.8(0.8) 0.4(0.2) SILOX
shrinkage stress
5.0
6.0
8.0
3.2(0.8)
2.5(0.7)
1.3(0.3)
1.6(0.3)
6.0(0.4) 2.0(0.3)
2-
",
""
/... - SILAR

10.0 3.9(1.5) 1.7(0.4) 1- II

15.0 4.6(0.7) 2.2(0.3) 6.6(1.5) 2.3(0.3)
20.0 5.8(0.7) 2.3(0.4)
30.0 6.9(3.4) 2.2(0.6) 7.8(4.1i) 2.4(0.3) -
1 5 10 1's 20 30
min
n 5.
=

Standard deviation between parentheses.

Teflon spacer was used to create a defined volume of corn
posite, which was only hindered in the axial hut not in the
transverse direction by adhesion forces. Polymerization of
the light-cured composite was initiated with a halogen light
source*. The created shrinkage torce was recorded contin-
uously, while the lower cross-head of the tensilometer was
kept stationary. At different times from the start of the
curing process, the bond strength was determined by ten-
sule loading at a speed of 0.2 mm/mmn. Each experiment
was repeated five times.
In a second experiment, 24 Class V cavities were pre-
pared in bovine teeth in such a way that the floors were cut
into dentin. The enamel of the cavities was etched for 60
sec with ortho-phosphoric acid (40 wt l) and washed for
4Optilux, Demetron Research Co., Danbury, CT
1i1g. 2 Comiposite-dentin bond strength and shrinkage stress as a
fu nction of the curing timie, determiined for the linear arrangemnent
in the tensilonieter.
45 sec. Dentin adhesive was applied to the cavities accord-
ing to the manufacturer's instructions. The cavities were
then filled, 12 with Silar and 12 with Silux. In these experi-
ments, the polymerization shrinkage was hindered in a
radial and in an axial sense. After storage for 48 hours in
water (370C), the teeth were sectioned down the centers
of the cavities and placed in a solution of methylene blue
for 10 min. The cross-sections were examined tor staining
under a stereo-microscope to check the integrity of the
bond. Control cross-sections were made of small amiounts
of the composites bonded to Hlat dentin surfaces and
immersed in the dye solution, to check whether the sec-
tioning procedure affectedl the adhesion.

Fig. 3 -A typical Class V restora-

tion with a dentin tloor, shiowing
disruption of the comiposite-den-
tin bond (dark line) in favor of
polymerization contraction to-
ward the composite-enamel bond.
C =comnposite, D dentin, and
=

L = enamel.

Downloaded from jdr.sagepub.com at Bobst Library, New York University on October 15, 2014 For personal use only. No other uses without permission.
1398 DA VIDSONETAL. J Dent Res December 1 984

TABLE 2
DEFECTIVE ADHESION OF COMPOSITES IN CLASS V CAVITIES EXPRESSED AS THE PERCENTAGE
WITH OBVIOUSLY VISIBLE STAINING* (n = 12)
Cervical wall Incisal wall Enamel- Cervical wall Incisal wall
Floor of cavity of cavity in of cavity in dentin of cavity in of cavity in
Material in dentin dentin dentin junction enamel enamel
Silar 96% 100% 96% 40% 10% 16%
Silux 96% 92% 84% 96% 0% 0%
*0% means no staining, thus perfect adhesion.
100% means complete staining, thus no adhesion.

Results. can be deduced from the analogously developing slopes of

The polymerization shrinkage stress and the bond
strengths of the two composites to dentin after different
curing times are listed in Table 1 and represented graphic-
ally in Fig. 2. Since the control cross-sections of the com-
posites bonded to flat dentin surfaces did not show defec-
tive adhesion caused by the sectioning procedure, observed
staining at dentin-composite borders in the Class V restora-
tions was ascribed to contraction forces developed during
polymerization. In Table 2, the results are given regarding
the bond between the composite and various sites of the
Class V cavities. Fig. 3 is a representative example of such a
the contraction stress and bond strength curves for each
kind of polymerization initiation. Significant differences in
ultimate bond strength values for both materials were not
demonstrated. Although the goal of having a composite-
dentin bond at least during curing of the resin is not achieved
with the bonding agent investigated, the initial adhesion is
still of importance. The pre-treatment of the dentin floor
t SILAR

cross-sectioned restoration.
tensile strength
Discussion.
From the results of the linear dentin-composite arrange-
ment in the tensilometer (Table 1), we concluded that,
when polymerization contraction is restricted to one
direction only, a substantial composite-dentin bond is
formed. The fact that the bond could withstand the poly-
merization contraction forces in such a device can be
explained by the capacity of a composite to flow in the
early rigid state (Davidson and De Gee, 1984). In our
second experiment, where the composite was constrained __-.--.-. shrinkage stress in 3-D
in three dimensions rather than one, almost no composite-
dentin bond withstood the polymerization shrinkage forces
(Table 2). Obviously, when contraction is hindered in three
dimensions, the stress will be less compensated for by flow.
If adhesion is maintained throughout the complete poly-
merization process, an estimate of the stress can be made for
lI
I
Silar on the basis of the average dimensions of our Class oe
V cavities (0= 3.0 mm and h = 1.8 mm) and the ultimate
stress value of 32.2 MPa as calculated by Davidson and De
I
Gee (1984). In these cavities, approximately 2/3 of the
total surface of composite material contacts the tooth
I
structure, which may develop an estimated stress of 20 I
MPa. A tentative explanation for the observed disruption
of the composite from the dentin can be given, when the
curve of the estimated contraction stress is superimposed
I _._--bond strength
on the experimental bond strength curve (Fig. 4). For
Silar, we see that initially the bond strength exceeds the
contraction stress; however, shortly hereafter, the latter ,--
_- shrinkage stress in 1-D
develops so rapidly that it surpasses the bond strength at
B. Obviously, then, the bond strength is too weak to allow
flow to compensate for the contraction forces set up in a
three-dimensionally patterned polymerization as is the case
within a concave Class V cavity. In Fig. 4, the bond strength
curve after time B is only imaginary because it does not
exist in reality. Since Fig. 4 shows the tensile strength of Fig. 4 -.-.- -. and - - -- -- -- Silar-dentin bond strength and
Silar at all times to be higher than the estimated contrac- shrinkage stress as a function of curing time, determined for the
tion stress, no damage inside the composite material itself linear arrangement in the tensilometer (from Fig. 2); - - ---
is to be expected in these cases. The influence of the rate estimated contraction stress for Silar, when fully bonded to all
of polymerization on the development of the bond strength walls of a Class V cavity; diametral tensile strength of
Silar (De Lange etal., 1983).
Downloaded from jdr.sagepub.com at Bobst Library, New York University on October 15, 2014 For personal use only. No other uses without permission.
 *-.
Vol. 63 No. 12
.,,,,- ........
-w: --
;'
and walls of the cavity promotes the wetting by the still-
viscous composite, to enable it to flow readily into the cav-
ity. Using this facilitates the handling of the composite
STRESS ON THE COMPOSITE-DENTIN BOND
-- =.. .. . .,.......
materials and improves their adaptation.
Our experiments indicate that, with the present material,
a durable composite-dentin bond can only be formed in
favorable circumstances, such as on flat surfaces and in
shallow cavities (Hansen, 1984). For wide, discus-shaped
restorations (Figs. Sa and b), the contraction along the
composite-dentin interface may also damage the bond.
Lambrechts and Vanherle (1982) have described that the
arc undergoes greater linear shrinkage than the chord, tend-
ing to cause the margins to curl during polymerization.
Curling directed toward the interface is considered more
favorable for the bond than curling away from the margins.
In more concave cavities (Figs. 3 and 5c), bulk contraction
plays the most important role. Here, the composite becomes
detached from the dentin part of the cavity and will remain
attached only to the enamel walls. Not only the develop-
ment of bonding agents with bond strengths which are
stronger than the contraction stress, but also composites
with smaller shrinkage and cavity design deserve attention
to obtain a stable marginal seal of composites in dentinal
cavities.
Acknowledgments.
The authors wish to thank Miss A. Bebb, Miss J. Bees,
Mrs. M. Duppen-Cohen, and Messrs. A.D.P. Heyboer and
R.S. McCarroll for their help during this study.
Downloaded from jdr.sagepub.com at Bobst Library, New York University on October 15, 2014 For personal use only. No other uses without permission.
c
1399
Fig. 5-Schematic representation
of the various contraction patterns
in differently shaped cavities and
the respective forms of disruption
of the adhesion. C = composite,
D = dentin, and E = enamel.
REFERENCES
ASMUSSEN, E. and MUNKSGAARD, E.C. (1983): Bonding of
Restorative Resins to Dentin by means of Methacryloylchloride
and Methacryloyl-R-isocyanate, Scand J Dent Res 91:153-155.
BOWEN, R.L.; COBB, E.N.; and RAPSON, J.E. (1982): Adhesive
Bonding of Various Materials to Hard Tooth Tissues: Improve-
ment in Bond Strength to Dentin, JDent Res 61:1070.1076.
BOWEN, R.L. and COBB, E.N. (1983): A Method for Bonding to
Dentin and Enamel, JADA 107:734-736.
CAUSTON, B.E. (1984): Improved Bonding of Composite Restora-
tion to Dentine, Br Dent J 156:93-95.
DAVIDSON, C.L. and DE GEE, A.J. (1984): Relaxation of Poly-
merization Contraction Stresses by Flow in Dental Composites,
J Dent Res 63:146-149.
DE LANGE, C.; BAUSCH, J.R.; and DAVIDSON, CL. (1983):
The Influence of Shelf Life and Storage Conditions on Some
Properties of Composite Resins, J Prosthet Dent 49:349-356.
HANSEN, E.K. (1984): Effect of Scotchbond dependent on Cavity
Cleaning, Cavity Diameter and Cavosurface Angle, Scand J
Dent Res 92:141-147.
HEGDAHL, T. and GJERDET, N.R. (1977): Contraction Stresses
of Composite Resin Filling Materials, Acta Odontol Scand
35:191-195.
LAMBRECHTS, P. and VANHERLE, G. (1982): The Use of Glaz-
ing Materials for Finishing Dental Composite Resin Surfaces,
J Oral Rehabil 9:107-117.
MASUHARA, E. (1982): Die neuentwickelten haftflhigen Kunst-
stoffe und ihre klinische Anwendung, Dtsch Zahndrztl Z 37:155-
159.
NAKAMICHI, I.; IWAKU, M.; and FUSAYAMA, T. (1983): Bovine
Teeth as Possible Substitutes in the Adhesion Test, J Dent Res
62: 1076-1081.