Annealing.
as a mechanism
of ncreas ng wear resistance of composites
A.J. de Gee
P. Pallav
A. Werner
C.L. Davidson
Department of Dental Materials Science
ACTA
Louwesweg 1
1066 EA Amsterdam
The Netherlands
Received October 19, 1989
Accepted May 1, 1990
Dent Mater 6:266-270, October, 1990
Abstract-The objective of this study was
to examine the influence of a short-term
exposure to heat (125°C) on the wear
resistance of composites. Both light- and
chemically initiated materials improved by
20-60%. However, the improvement was
also attained in the course of time if the
materials had not been exposed to heat.
The heat-induced improvement could not
be explained by a continuation of
polymerization but rather by stress relief,
which is common for annealing
processes. Polymerization shrinkage
stresses, Initially concentrated mainly
around the filler particles, became more
homogeneously distributed by the heat
treatment. The long-term improvement of
the non-heat-treated materials was based
on the same mechanism, but proceeded
more gradually.
288 DE GEE et aL/IMPROVED WEAR RESISTANCE B Y ANNEALING COMPOSITES
he use of composites for stress-
T bearing restorations in the pos-
terior region revealed essential
shortcomings of these materials with
respect to wear resistance (Swift,
1987; Vann et al., 1988) and sealing
capacity at the margins (Browne and
Tobias, 1986; Gross et aL, 1985). The
quality of both characteristics de-
pends on, among other things, the
degree of conversion of the resin.
However, improving the curing ef-
ficiency to benefit the wear resis-
tance (Ruyter and ~yssed, 1987) will
at the same time increase the risk of
marginal debonding as a result of a
higher polymerization shrinkage
(Asmussen, 1975; Davidson, 1985).
A solution to these conflicting inter-
ests was sought in the use of com-
posite direct inlays (Lutz et aL, 1987).
An inlay can be cured to a higher
degree under conditions which can-
not be tolerated in the mouth. This
results in a more wear-resistant res-
toration (Wendt, 1987a) as well as in
improved mechanical p r o p e r t i e s
(Bausch et al., 1981; Cook and Jo-
hannson, 1987; Wendt, 1987b). In
addition, the problem of polymeri-
zation shrinkage is restricted to the
luting cement only.
The purpose of this study was to
investigate the effectiveness of post-
curing composites in a fight/heat-oven
on the wear resistance in relation to
materials that functioned without this
treatment.
MATERIALS AND METHODS
The composite materials tested in this
study are listed in Table 1. Only
Brilliant DI is marketed as a direct
composite inlay material. All the
other materials are posterior com-
posites of both the chemically cured
(CC) and light-cured (LC) types.
The m a t e r i a l s w e r e placed di-
rectly into a wear-testing wheel (Fig.
1), covered by a matrix, and cured
in place at 37°C: the photo-initiated
composites by illumination with a
Visilux 2 light source (3M Co., St.
Paul, MN) for at least 120 s, while
the light guide was moved over the
sample surface, and the chemically
initiated composites a f t e r being
mixed for 30 s. Immediately after
being set, the samples w e r e re-
moved from the wear-testing wheel
and placed in a light/heat-oven (DI-
500 Colt~ne AG, Altst~itten, Swit~
zerland). This oven, especially de-
signed for the preparation of direct
composite inlays, exposed the sam-
ples to the unfiltered light of a hal-
ogen lamp at 125°C for seven min.
As a control, a series of samples was
also prepared without this treat-
ment. All samples were glued into
the sample-holding wheels and wet-
ground with grit 240, 320, 400, and
600 SiC paper so that a perfect cy-
lindrical outer surface would be ob-
tained. A layer of approximately 200
~m was removed by this procedure.
From this moment on, the materials
were kept continuously wet. The
wheels were mounted in the wear
machine (Fig. 1), which simulates in
vivo occlusal wear as described ear-
lier in detail (de Gee et al., 1986; Pal-
lav et aL, 1988).
Eight wear tests of 200,000 rev-
olutions each were performed in
succession over a period of 70 days.
The loss of material after each wear
test was determined in ~m depth
by p r o f i l o m e t r i c m e a s u r e m e n t s
(Perthometer type C3A, Perthen,
H a n n o v e r , W e s t Germany). The
first two data points of each mate-
rial were obtained five and 10 days
after initiation. Each represents the
average of three surface tracings (n
= 3). The results from the third
and fourth test runs were com-
l~ined, as were those for the fifth
and sixth and the seventh and eighth
tests, since they were performed in
close succession. T h e i r a v e r a g e
values (n = 6) represent, respec-
tively, data points 25, 40, and 70
days after initiation.
 TABLE 1
MATERIALS INVESTIGATEDIN THIS STUDY
Materials Code CC/LC, Batch Numbers Manufacturer
Brilliant D.I.* BDI LC 180288-78 Y Colt~ne AG, A]tst~itten,
Switzerland
Occlusin OCC LC UN 60 MAR 87 ICl Dental Div., Macclesfield, UK
P30 P30 LC 7X5 Universal 3M Co., St. Paul, MN
P50 P50 LC 72P Universal 3M Co.
PIO PIO CC A:7AB1 B:7AD1R 3M Co.
Clearfil Posterior New Bond~ CPNB CO PU-2256 P0-2156 Kuraray Co. LTD., Osaka, Japan
Clearfil Posterior 3 CP3 CC KU-0206 KC-0106 Kuraray Co. LTD
Clearfil Photo Posteriors CPP LC HPS 1006 Universal Kuraray Co. LTD
Visio Molar VM LC 0003 Espe GmbH, Seefeld,
Oberbayern, FRG
Visio Dispers VD LC 0095 Espe GmbH
Herculite HC LC 7 1252 U66 Enamel Kerr Mfg. Co., Romulus, MI
• CC = chemically-cured and LC = light-cured.
• Marketed as direct composite inlay material.
Also marketed under the name Clearfil Posterior.
s Also marketed under the name Clearfil Ray Posterior.

RESULTS
Fig. 3 shows the wear in ~m plotted
against the ages of the samples, that
is, the time in days elapsed from ini-
tiation to the moment at which the
wear was determined. The curves
connect the five wear-data points
obtained over a period of ?0 days.
Solid lines represent the materials
cured at 37°C and broken lines the
materials after an additional treat-
ment in the DI-500 light/heat oven.
Table 2 compiles the results of the '. steel
statistical evaluation according to
Student's t test at the 5% level of
significance. Significant (s) and non- /~l
~unworn
\ w()rn reference
significant (ns) differences are shown /,~l \ area plane
between various data points A to F,
as indicated in Fig. 2, in which A and ~111 unworn
~1 reference
D represent the wear at five days, plane
B and E that at 10 days, and C and
F that at 70 days.
DOWI wl[n mlllgt-
DISCUSSION seed/water slurry
Application of the DI-500 apparatus, Fig. 1. Left: Stainless steel sample.holding wheel with a number of samples, polymerized in place. Middle:
which is usually done for the prep- Sample-holding wheel and "antagonist" wheel rotating against each other in a food slurry. Right: Sudaoe
tracing to record loss of material in p.m depth relative to the two unworn references.
aration of direct composite inlays,
significantly improved the wear re-
sistance of all the materials investi-
gated in this study. Although the TABLE 2
apparatus emits its full light inten- SURVEY OF SIGNIFICANT(s) AND NON-SIGNIFICANT(ns) DIFFERENCESBETWEENVARIOUS DATA
POINTS AT THE 5% LEVEL OF SIGNIFICANCEACCORDINGTO STUDENT'S tTEST (cf. Fig. 2)
sity unfiltered onto the samples, the
improvement is entirely induced by Data Material Code
the heat of the light bulb, as was Points BDI OCC P30 P50 PIO CPNB Cp3 CPP VM VD HC
d e m o n s t r a t e d for the t r a n s v e r s e A,B s s s s s s ns s s s s
strength and elastic modulus (Rein- B,C ns ns s s s s s s s s s
hardt and Smolka, 1988). Improve- A,C s s s s s s s ns s s s
ment of mechanical properties and A,D s s s s s s s s s s s
wear resistance by heat for both B,E ns s s s s s ns s s s s
chemically and photo-initiated com- C,F ns ns ns ns ns ns ns ns ns s s
posites was also shown by Bausch et D,F ns ns ns s s s s s s ns ns

Dental MaSerials/October 1990 267


£
5 t0 70
age Hays)
F/g. 2. General representative picture of the
illustrations in Fig. 3. A and D, B and E, and C and F
represent wear-data points five, 10, and 70 days
after initiation, respectively, materials cured
at 37°C, - ..... materials additionally heat-treated in
the DI-500 oven.
al. (1981), Cook and J o h a n n s o n
(1987), and Wendt (1987a, b). In the
present study, a 20-60% improve-
ment was shown for the wear resis-
tance. This is evident from Fig. 3,
where the wear at five days of the
DI-treated materials is compared
with that of their controls (cured in
the normal way). However, when the
wear rates were followed over a pe-
riod of 70 days, the difference ap-
peared to decrease with time to an
insignificant value for most of the
materials (Fig. 3 and Table 2), i.e.,
the curves for the heat-treated ma-
terials and the controls approached
one another. This is due mainly to a
significant decrease of wear rate of
the normally cured materials be-
tween five and 10 days (Table 2, in-
terval A-B) and more gradually from
10 to 70 days ~Table 2, interval B-
C). To a small extent, the approach-
ing of the curves can also be attrib-
uted to an increase in wear rate for
many of the heat-treated materials
(Table 2, interval D-F).
Although it has been well-docu-
m e n t e d for both chemically and
photo-initiated composites that me-
chanical properties still develop with
time by a continued cross-linking re-
action (Leung et al., 1983; Hansen,
1983; Ferracane, 1985; Braem et al.,
1987; Lloyd and Adamson, 1987), this
process usually ceases within one day
after initiation (Watts et al., 1987).
After this period, the post-curability
at ambient temperature of the re-
maining double bonds, usually pres-
ent in the order of 25-45% (Fen'acane
et al., 1985), is greatly reduced, as
is already the case under thermally
induced conditions (Eliades et al.,
1987). Therefore, it seems unlikely
268 DE GEE et aL/IMPROVED WEAR RESISTANCE BY ANNEALING COMPOSITES
t h a t the observed significant im-
provement of wear resistance over
much longer periods resulted from a
continued cross-linking reaction. A
mechanism which could better ac-
count for the improvement (in par-
ticular, stress-bearing wear) may be
a gradual relaxation of polymeriza-
tion contraction stresses in the ma-
trix concentrated around the filler
particles into a more homogenized
distribution throughout the entire
polymer matrix. The total stress in
the matrix around and underneath
the particles during mechanical load-
ing may therefore decrease with time,
reducing the formation of micro-
cracks and, consequently, w e a r
(Heymann et aL, 1986). Equally, this
very same mechanism may be mainly
responsible for the improved wear
resistance due to heat treatment in
the oven. Under such conditions, the
relaxation of local stress concentra-
tions around the filler particles into
a more homogenized distribution will
only be accelerated and will be main-
rained after cooling. This process can
be compared with the annealing of
local stresses in amorphous glass
constructions. It must be appreci-
ated that additional cross-linking also
occurs if heat is applied directly after
initiation (Eliades et al., 1987; Cook
and Johannson, 1987). However, from
p r e l i m i n a r y experiments, t h e r e
seems to be no evidence that this is
the determining factor for the im-
provement of wear resistance. In
these experiments with P30 and P50,
it was found that the improvement
of wear resistance attained the same
level, D (Fig. 2), regardless of the
moment of heat treatment, whether
this was done directly after initia-
tion or only after eight days. Thus,
despite secondary cross-linking-
which does occur under the former
conditions, but can almost be ex-
cluded under the latter (Eliades et
a/,, 1987)-an equal (improved) wear
resistance was obtained. Appar-
ently, secondary cross-linking did not
contribute to the improvement of
wear for the two materials men-
tioned, which supports the proposed
mechanism of heat-induced stress
relaxation. On this basis, it can also
be understood that most of the nor-
mally cured materials could finally
reach the same wear level as their
heat-cured versions by long-term
stress homogenization. 0nly for the
materials HC and VD may addi-
tional cross-linking by heat have
contributed to the improvement of
wear resistance, since a significant
difference in wear between the nor-
mally- and heat-cured samples still
remained after 70 days (Table 2).
The slow decrease in wear resis-
tance of many of the heat-cured (fully-
stress-annealed) matel"ials during the
70-day period (Fig. 3 and Table 2)
must be ascribed to surface-soften-
ing by various chemicals (Wu et al.,
1984; McKinney and Wu, 1985; As-
m u s s e n , 1984; Montes-G. and
Draughn, 1986) which are present in
the millet slurry and to hydrolytic
degradation (SSderholm et ~ , 1984).
For the normally cured materials,
this process is masked by the oppo-
site effect from stress homogeniza-
tion and therefore is not reflected in
the wear-rate curves. It may even
be that the stress homogenization is
stimulated by water sorption, com-
parable to the strain release as de-
scribed by Smith and Schoonover
(1953) in their water sorption exper-
iments. This may partly explain the
wear behavior of VM cured in the
normal way, showing an increased
wear rate during the first 10 days.
This extremely hydrophobic mate-
rial exhibits only a very slow and very
small water sorption (Feilzer et al.,
1990), which delays stress homoge-
nization and thereby provides op-
portunity for superficial chemical-
softening. Only after some time,
when water has diffused deeper into
the matrix, does stress homogeni-
zation become predominant again. A
second explanation for the delay in
stress homogenization can be sought
in a reduced mobility in the polymer
network caused by the relatively rigid
nature of the tricyclodecane struc-
ture in the Espe monomer. The same
may hold for CPP, with a polymer
network build-up from tetra-func-
tional urethane monomers.
The present findings show that
composites cured in the normal way
will, within a few weeks, attain a
wear resistance comparable with that
of their additionally light/heat-ta'eated
equivalents. Thus, the light/heat
procedure, advocated as a secondary
cure to improve the wear resistance
of composites, offers this advantage
only for a short, early period of the
 50 50, 50
Brilliant O.I. P-tO Visio-Nolar
4C 40F 4O
l=

30 .................. °..o-- .o°°..- .........................

~o 20~ 20
I
lO to o/o .................................................
#,

I ~ i i i i
o i i i I a I

tO 20 30 40 50 60 70 o i0' '
20 3"g '
40 '
50 '
60 70 tO 20 30 40 50 60 70
age (days) age (days) age (days)

50 5O 5O
Occlusin Clearfil Posterior New BonO
4~ 4O 4O

--=' 30 ~' 30 i 30
~'-- ............. ° ...°°""°'° .........................

2O == 2o
\ 2C
.....°,.°.-° ...........................

,°o .................. .

to tO

I i i i i i i i | I i i
o
10 20 30 40 50 60 70 tO 20 30 40 50 60 70 iO 20 30 40 50 rio 70
age (days) age (days) age (days)

50 5C 5O
P-30 Clearfil Posterior 3
4O 4G 4O
"i
30 ~ 3c t 3Q

20 20
.o°O° ............ , ....................................

t0 10

I ~ I I I I
0 I I
0
tO 2 30 40 50 BO 70 0 ,o 7o 0
age (days} age (days) age (deys)

50 50
P-50 Clprfil Photo Posterior
40 4O

' 30
"i Fig. 3. Wear (~m/200,O00 rev) of composites at
various ages after initiation over a period of 70
I days. materials cured at 37°C, -....
20 20
o°°o° ........... ~ ........... ~ .......... materials additionally heat-treated in the DIe500
.oO,, ............ • ........... o ....... - ..... - ..... - ....
10 oven.
t0
~0

ago (dlyl) alia (deya}

lifetime of the restoration. Yet the REFERENCES Properties of Dental Composites, J

method may create a more mechan- Oral Rehabil 8: 309-317.
ically stable situation during the early ASMUSSEN, E. (1975): Composite Re- BRAEM, M.; LAMBRECHTS, P.;
lifetime of the inlay with respect to storative Resins. Composition versus VANHERLE, G.; and DAVIDSON, C.L.
marginal sealing (Reinhardt and Wall-to-Wall Polymerization Contrac- (1987): Stiffness Increase During the
tion, Acts Odontol Scand 33: 337-344. Setting of Dental Composite Resins, J
Smolka, 1988). Recent studies (Feil- Dent Res 66: 1713-1716.
zer et al., 1989), however, indicated ASMUSSEN, E. (1984): Softening of Bis-
GMA-based Polymers by Ethanol and BROWNE, R.M. and TOBIAS, R.S. (1986):
that shrinkage stress in thin com- by Organic Acids of Plaque, Scand J Microbial Microleakage and Pulpal In-
posite cement layers is extremely Dent Res 92: 257-261. flammation: A Review, Endod Den~
high. Therefore, the application of the BAUSCH, J.R.; DE LANGE, C.; and Trauma$ol 2: 177-183.
adhesively bonded inlay deserves DAVIDSON, C.L. (1981): The Influence COOK,W.D. and JOHAZ~NSON,M. (1987):
further study. of Temperature on Some Physical The Influence of Postcuring on the

Dental Materials~October 1990 268

 Fracture Properties of Photo-cured
Dimethacrylate Based Dental Com-
posite Resin, J Biomed Mater Res 21:
979-989.
DAVIDSON, C.L. (1985): Conflicting In-
terests with Posterior Use of Com-
posite Materials. In: Posterior
Composite Resin Dental Restorative
Materials, G. Vanherle and D.C.
Smith, Eds., The Netherlands: Peter
Szulc, pp. 61-65.
DE GEE, A.J.; PALLAV, P.; and
DAVIDSON, C.L. (1986): Effect of
Abrasion Medium on Wear of Stress-
bearing Composites and Amalgam in
vitro, J Dent Res 65: 654-658.
ELIADES, G.C.; VOUGIOUKLAKIS, G.J.;
and CAPUTO, A.A. (1987): Degree of
Double Bond Conversion in Light-cured
Composites, Dent Mater 3: 19-25.
FEILZER, A.J.; DE GEE, A.J.; and
DAVIDSON, C.L. (1989): Increased
Wall-to-Wall Curing Contraction in
Thin Bonded Resin Layers, J Dent Res
68: 48-50.
FEILZER, A.J.; DE GEE, A.J.; and
DAVIDSON, C.L. (1990): Relaxation of
Polymerization Contraction Shear
Stress by Hygroscopic Expansion, J
Dent Res 69: 36--39.
FERm,,CANE, J.L. (1985): Correlation
Between Hardness and Degree of
' Conversion During the Setting Reac-
tion of Unfilled Dental Restorative
Resins, Dent Mater 1: 11-14.
FERRACANE, J.L.; MATSUMOTO,H.; and
0EASE, T. (1985): Time-dependent
Deformation of Composite Resins-
Compositional Considerations, J Dent
Res 64: 1332-1336.
GROSS, J.D.; RETIEF, D.H.; and
BRADLEY, E.L. (1985): Microleakage
270 DE GEE et aL/IMPROVED WEAR RESISTANCE BY ANNEALING COMPOSITES
of Posterior Composite Restorations,
Dent Mater 1: 7-10.
HANSEN, E.K. (1983): After-Polymeri-
zation of Visible Light Activated Res-
ins: Surface Hardness vs. Light Source,
Scand J Dent Res 91: 406-410.
HEYMANN, H.O.; WILDER, A.D.; MAY,
K.N.; and LEINFELDER, K.F. (1986):
Two-year Clinical Study of Composite
Resins in Posterior Teeth, Dent Mater
2: 37-41.
LEUNG, R.L.; FAN, P.L.; and JOHNSTON,
W.M. (1983): Post-irradiation Poly-
merization of Visible-light-activated
Composite Resin, J Dent Res 62: 363-
365.
LLOYD, C.H. and ADAMSON, M. (1987):
The Development of Fracture Tough-
ness and Fracture Strength in Poste-
rior Restorative Materials, Dent Mater
3: 225-231.
L~rrz, F.; KREJCI, I.; and MORMANN,W.
(1987): Die Zahnfarbene Seitenzahn-
Restauration, PhiUip J 4: 127-137.
MCKINNEY, J.E. and Wu, W. (1985):
Chemical Softening and Wear of Den-
tal Composites, J Dent Res 64: 1326-
1331.
MONIES-G, G.M. and DRAUGHN, R.A.
(1986): In vitro Surface Degradation of
Composites by Water and Thermal
Cycling, Dent Mater 2: 193-197.
PALLAV, P.; DAVIDSON, C.L.; and DE
GEE, A.J. (1988): Wear Rates of Com-
posites, an Amalgam, and Enamel un-
der Stress-bearing Conditions, J
Prosthet Dent 59: 426429.
REINHARDT,K.J. and SMOLKA,R. (1988):
Kunststoffe im Seitenzahnbereich-
Fiillung oder Inlay?, Dtsch Zahn~2rztl
Z 43: 909-913.
RUYTER, I.E. and ~YS~D, H. (1987):
Composites for Use in Posterior Teeth:
Composition and Conversion, J Biomed
Mater Res 21: 11-23.
SMITH, D.L. and SCHOONOVER, I.C.
(1953): Direct Filling Resins: Dimen-
sional Changes Resulting from Poly-
merization Shrinkage and Water
Sorption, J A m Dent Assoc 46: 540-
544.
SODERHOLM, K.J.; ZIGAN, M.; RAGAN,
M.; FISCHLSCHWEIGER, W.; and
BERGMAN,M. (1984): Hydrolytic Deg-
radation of Dental Composites, J Dent
Res 63: 1248-1254.
SWIFT, E.J. (1987): Wear of Composite
Resins in Permanent Posterior Teeth,
J A m Dent Assoc 115: 584-588.
VANN, W.F.; BARKMEIER, W.W.; and
MAHLER, D.B. (1988): Assessing
Composite Resin Wear in Primary
Molars: Four-year Findings, J Dent
Res 67: 876-879.
WATTS, D.C.; AMER, O.M.; and COMBE,
E.C. (1987): Surface Hardness Devel-
opment in Light-cured Composites,
Dent Mater 3: 265-269.
WENDT, S.L. (1987a): The Effect of Heat
used as a Secondary Cure upon the
Physical Properties of Three Compos-
ite Resins. II. Wear, Hardness, and
Color Stability, Quint Int 18: 351-356.
WENDT, S.L. (1987b): The Effect of Heat
used as a Secondary Cure upon the
Physical Properties of Three Compos-
ite Resins. I. Diametral Tensile
Strength, Compressive Strength, and
Marginal Dimensional Stability, Quint
Int 18: 265-271.
Wu, W.; TOTH, E.E.; MOFFA, J.F.; and
ELLISON, 3.A. (1984): Subsurface
Damage Layer of in vivo Worn Dental
Composite Restorations, J Dent Res
63: 675-680.