Chemical Softening and Wear of Dental Composites

Abstract

The purpose of this work was to determine the influence of chemical food-simulating liquids on the wear of various commercial dental composite restoratives. In many cases, pre-conditioning the restoratives in these liquids for one week produced swelling of the polymer matrix and considerable surface damage. The resulting degradation reduced the hardness and enhanced the wear as measured by a pin-and-disc apparatus. Four kinds of commercial composites were investigated: a conventional quartz-filled, a strontium-glass-filled, a visible-light-activated, and a microfilled composite. The liquids employed were heptane and several aqueous solutions of ethanol with solubility parameters, delta, ranging from delta = 1.5 to 4.8 X 10(4) J1/2m-3/2. With all restoratives, the decline in hardness during pre-conditioning maximized at about delta = 3 X 10(4), which corresponds to a 75% ethanol solution. The wear behavior was considerably more complicated and variable, as discussed in the text. For the most part, the increase in wear rate from pre-conditioning corresponded to the fall in hardness. A notable exception was for the strontium-glass-filled composite pre-conditioned in pure water. Here the wear was enhanced considerably, with no decrease in hardness. In this case, the degradation mode is assumed to be different from the others in that it is attributed to stress corrosion of the glass filler.

Full text

http://jdr.sagepub.com/
Journal of Dental Research
http://jdr.sagepub.com/content/64/11/1326
The online version of this article can be found at:
DOI: 10.1177/00220345850640111601
1985 64: 1326J DENT RES
J.E. McKinney and W. Wu
Chemical Softening and Wear of Dental Composites
Published by:
http://www.sagepublications.com
On behalf of:
International and American Associations for Dental Research
can be found at:Journal of Dental ResearchAdditional services and information for
http://jdr.sagepub.com/cgi/alertsEmail Alerts:
http://jdr.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://jdr.sagepub.com/content/64/11/1326.refs.htmlCitations:
by guest on July 10, 2011 For personal use only. No other uses without permission.jdr.sagepub.comDownloaded from

Chemical
Softening
and
Wear
of
Dental
Composites
J.E.
McKINNEY
and
W.
WU
Dental
&
Medical
Materials,
Polymer
Science
&
Standards
Division,
National
Bureau
of
Standards,
Gaithersburg,
Maryland
20899
The
purpose
of
this
work
was
to
determine
the
influence
of
chemical
food-simulating
liquids
on
the
wear
of
various
commercial
dental
com-
posite
restoratives.
In
many
cases,
pre-conditioning
the
restoratives
in
these
liquids
for
one
week
produced
swelling
of
the
polymer
matrix
and
considerable
surface
damage.
The
resulting
degradation
reduced
the
hardness
and
enhanced
the
wear
as
measured
by
a
pin-and-disc
apparatus.
Four
kinds
of
commercial
composites
were
investigated:
a
conventional
quartz-filled,
a
strontium-glass-filled,
a
visible-light-
activated,
and
a
microfilled composite.
The
liquids
employed
were
heptane
and
several
aqueous
solutions
of
ethanol
with
solubility
pa-
rameters,
8,
ranging
from
8
=
1.5
to
4.8
x
104
j1/2m-3/2.
With
all
restoratives,
the
decline
in
hardness
during
pre-conditioning
maxi-
mized
at
about
8
=
3
x
104,
which
corresponds
to
a
75%
ethanol
solution.
The
wear
behavior
was
considerably
more
complicated
and
variable,
as
discussed
in
the
text.
For
the
most
part,
the
increase
in
wear
rate
from
pre-conditioning
corresponded
to
the
fall
in
hardness.
A
notable
exception
was
for
the
strontium-glass-filled
composite
pre-
conditioned
in
pure
water.
Here
the
wear
was
enhanced
considerably,
with
no
decrease
in
hardness.
In
this
case,
the
degradation
mode
is
assumed
to
be
different
from
the
others
in
that
it
is
attributed
to
stress
corrosion
of
the
glass
filler.
J
Dent
Res
64(11):1326-1331,
November,
1985
Introduction.
As
observed
from
the
microdefect
analysis
described
by
Wu
and
Cobb
(1981),
clinically-worn
composite
restorations
re-
vealed
extensively
damaged
layers
on
non-stress-bearing
as
well
as
occlusal
surfaces
(Wu
et
al.,
1984).
Accordingly,
it
was
assumed
that
the
chemical
environment
has
an
appreciable
influence
on
the
in
vivo
degradation
of
these
materials.
The
observed
surface
damage
is
attributed
to
softening
and
possible
removal
of
portions
of
the
polymer
matrix
by
certain
chemicals
present
in
the
oral
environment.
Although
the
softening
will,
by
itself,
degrade
the
restoratives,
the
wear
on
occlusal
sur-
faces
is
expected
to
be
enhanced
considerably
by
the
softening
process.
This
hypothesis
was
corroborated
by
pin-and-disc
wear
measurements
(Wu
and
McKinney,
1982)
on
conventional-
composite
specimens
which
were
pre-conditioned
in
food-sim-
ulating
liquids
(FSL's).
The
selection
of
the
FSL's
was
based
on
recommendations
by
the
Food
and
Drug
Administration
(1976).
Some
of
these
attacked
the
polymer
matrix
and
reduced
wear
resistance
of
the
composite
considerably.
In
the
work
described
in
this
paper,
the
activity
was
extended
to
include
several
other
types
of
commercial
composites.
The
FSL's
may
be
ranked
quantitatively
by
utilizing
the
thermodynamic
solution
theory
(Hildebrand
and
Scott,
1950).
A
maximum
softening
effect
is
expected
when
the
value
of
the
magnitude
of
the
solubility
parameter
of
a
liquid
is
equal
to
Received
for
publication
October
11,
1984
Accepted
for
publication
July
30,
1985
Certain
commercial
materials
and
instruments
are
identified
in
this
report
to
specify
the
experimental
procedure
adequately.
In
no
in-
stance
does
such
identification
imply
recommendation
or
endorsement
by
the
National
Bureau
of
Standards,
nor
does
it
imply
that
the
ma-
terial
or
instrument
identified
is
necessarily
the
best
available
for
this
purpose.
This
work
was
supported
by
an
Interagency
Agreement
with
the
National
Institute
of
Dental
Research
(NIDR),
IA
YOI-DE-30001.
1326
that
of
the
matrix
polymer
of
the
composite.
The
values
of
the
solubility
parameters
of
the
solvents
are
in
the
Polymer
Hand-
book
(Burrell,
1975).
The
extent
of
softening
of
the
compos-
ites
may
be
defined
from
the
decline
of
measured
surface
hardness
resulting
from
the
pre-conditioning
in
the
FSL's.
The
purpose
of
this
work
was
to
evaluate
the
role
of
food-
simulating
liquids
(FSL's)
on
the
in
vitro
wear
of
dental
com-
posite
restorative
materials.
The
liquids
selected
embraced
the
complete
variety
of
foods
in
terms
of
their
solubility
parame-
ters.
Materials
and
methods.
Specimens.
Four
different
kinds
of
commercial
compos-
ites
were
investigated.
These
included
a
conventional
quartz-
filled
(CQ),
a
strontium-glass-filled
(SG),
a
visible-light-acti-
vated
(VLA),
and
a
microfilled
(M)
composite.
Table
1
gives
the
trade
names
of
these
materials;
their
codes,
which
pertain
to
the
most
salient
feature
of
each
as
identified
above;
batch
numbers;
cure
modes;
types
of
reinforcing
fillers;
and
manu-
facturers.
All
employ
a
dominantly
BIS-GMA-resin
matrix.
The
chemically-cured
composites
(paste-paste)
were
mixed
in
accordance
with
their
manufacturers'
instructions.
Immedi-
ately
afterward,
they
were
injected
by
syringe1
into
molds
made
from
denture
tray
material
to
produce
specimen
discs
for
wear
tests.
The
visible-light-activated
composite
(single
paste)
was
obtained
in
a
screw-driven
syringe
suitable
for
injection
into
the
molds.
Thin
glass
plates
were
forced
against
all
of
the
specimens,
with
heavy
weights
to
smooth
their
top
surfaces
and
to
prevent
air
inhibition
during
curing.
With
composite
VLA,
the
light2
was
applied
to
the
specimen
with
the
radiation
transmitted
through
the
glass
plate.
About
ten
exposures
of
10
sec
each
were
applied
at
various
positions
on
the
top
surface
of
each
specimen.
This
process
was
repeated
immediately
af-
terward
with
the
plate
removed.
All
cures
were
done
at
room
temperature.
Although
37°C
is
the
in
situ
use
temperature,
mixing
and
curing
were
done
at
room
temperature
to
obtain
cures
under
isothermal
ambient
conditions.
The
chemically-
cured
systems
could
not
be
mixed
as
well
at
37°C,
because
the
reactions
were
too
fast
at
that
temperature.
The
top
surfaces
of
the
specimens
were
ground
flat
and
normal
to
the
axis
of
rotation
and
subsequently
polished
in
sequential
stages
with
grit
sizes
diminishing
to
0.05
tIm.
With
the
visible-light-ac-
tivated
systems,
a
minimum
of
material
was
removed
because
of
the
presence
of
degree-of-cure
gradients
with
depth
caused
by
light
absorption.
Sufficient
material
was
abraded
from
all
specimens,
however,
to
remove
the
polymer-rich
regions
(Okazaki
and
Douglas,
1984)
adjacent
to
the
glass
plates.
The
final
dimensions
of
the
specimen
discs
were
approximately
2.5
x
18
mm
(diameter).
The
finished
specimens
were
stored
in
the
various
FSL's
for
one
week
at
37°C,
during
which
the
surface
hardness
was
measured.
The
specimens
were
re-in-
serted
into
new
tray
material
holders
molded
in
the
specimen
cups
suitable
for
wear
tests.
Cyano-acrylate
glue
was
used
to
secure
the
specimens
to
their
holders.
'Centrix,
Inc.,
Stratford,
CT
06497
2Command,
Kerr
Division
of
Sybron
Corp.,
Romulus,
MI
48174
by guest on July 10, 2011 For personal use only. No other uses without permission.jdr.sagepub.comDownloaded from

CHEMICAL
SOFTENING
AND
WEAR OF
DENTAL
COMPOSITES
TABLE
1
COMPOSITE
SYSTEMS
Brand
Name
Code
Batch
Nos.
Cure
Filler
Manufacturer
Adaptic
CQ
031081
Chemical
Conventional
Johnson
&
Johnson
Quartz
East
Windsor,
NJ
08520
Profile
SG
040681
Chemical
Hybrid
S.S.
White
091781
Strontium
Glass
Philadelphia,
PA
19102
103081
Radiopaque
111881
Prisma-Fil
VLA
040681
Visible
Light
Conventional
L.D.
Caulk
072081
Activated
Barium
Glass
Milford,
DE
19963
0316821
Radiopaque
Silar
M
012280
Chemical
Microfilled
3M
Company
091880
Silica
St.
Paul,
MN
55144
030982
100682
CQ
=
Conventional
Quartz,
SG
=
Strontium
Glass,
VLA
=
Visible
Light
Activated,
M
=
Microfilled.
Food-simulating
liquids/FSL's.
The
FSL's
used
for
pre-
were
dead-weight-loaded
to
achieve
an
average
normal
stress
conditioning
in
this
investigation
are
among
those
recom-
of
10
MPa
over
the
pin
cross-section.
The
pin-and-wear
track
mended
in
"FDA
Guidelines"
(FDA,
1976)
to
be
used
as
food
diameters
were
2.00
and
12.0
mm,
respectively.
The
discs
simulators.
The
liquids
are
listed
in
Table
2
with
their
solu-
were
rotated
at
27
rpm.
The
specimen
surfaces
were
constantly
bility-parameter
values
and
examples
of
foods
they
are
in-
flushed
with
distilled
water
at
37°C.
Sets
of
wear-track
mea-
tended
to
simulate.
surements
were
taken
periodically
using
linear
variable
differ-
The
period
of
pre-conditioning
in
the
FSL's
before
the
pin-
ential
transformers
with
ruby-tipped
probes.
Each
set
consisted
and-disc
wear
measurement
was
one
week
without
interrup-
of
a
measurement
at
each
of
ten
equidistant
positions
around
tion.
This
length
of
pre-conditioning
time
is
rather
extensive
the
track
circumference.
The
average
of
these
depths
was
used
in
light
of
the
fact
that
composite
restorations
come
into
contact
as
a
measure
of
volume
of
material
removed.
with
foods
only
briefly
in
a
sporadic
way.
Accordingly,
the
Initially,
for
each
run,
track-depth
measurements
were
taken
test
results
to
be
reported
may
exaggerate
the
softening
effects
over
the
first
five
disc
revolutions.
The
value
at
five
revolu-
on
the
chemical
resistance
of
the
dental
composites
tested.
The
tions
was
selected
arbitrarily
as
the
initial
wear.
At
the
end
of
wear
rate
of
the
pre-conditioned
composites
measured
herein
this
period,
a
large
portion
of
the
chemically-damaged
layer
will
surely
exceed
that
which
occurs
in
vivo;
however,
the
was
removed.
Subsequently,
data
were
taken
every
hour,
for
accelerated
in
vitro
rate
provides
an
important
clue
to
the
wear
which
each
hour
corresponded
to
about
1500
revolutions.
The
performance
of
these
composites
in
vivo.
ten
circumferential
track-depth
measurements
at
the
end
of
Hardness
measurements.
Surface-hardness
values
of
the
each
hour
required
about
three
minutes,
during
which
the
wear
composites
were
obtained
with
a
Knoop
Hardness
Tester
was
only
over
one
revolution.
(ASTM,
C-730-1984).
These
measurements
were
used
to
mon-
itor
the
softening
of
the
composites
during
their
one-week
pre-
conditioning
period
in
the
solvents.
The
Knoop
Hardness
Results.
Number
(KHN)
of
each
specimen
was
obtained
from
the
av-
Surface
hardness.
Knoop
hardness
data
obtained
on
the
erage
of
five
indentation
lengths.
Approximately
45-second
four
composites
pre-conditioned
in
the
various
FSL's
are
shown
contact
times
were
used
in
all
cases.
In
order
to
keep
the
in
Fig.
1.
The
composites
are
identified
in
the
upper
left-hand
indentation
lengths
reasonably
constant,
the
load
was
adjusted
comers
by
the
codes
defined
in
Table
1.
The
data
are
expressed
appropriately
between
100
and
300
g.
in
terms
of
a
hardness
ratio,
pi/pf,
where
Pi
is
the
initial
KHN
Wear
tests.
Wear
tests
were
made
on
a
fully
automatic,
before
pre-conditioning,
and
pf
is
the
final
value
taken
seven
three-station,
pin-and-disc
apparatus
(McKinney,
1982),
where
days
later,
at
which
time
pre-conditioning
is
taken
arbitrarily
the
disc
was
the
specimen
and
the
pin,
the
counterface.
Type
to
be
complete.
The
hardness
ratio,
pi/pf,
for
the
four
com-
303
stainless-steel
pins
were
used
in lieu
of
human-enamel
pins
posites
is
plotted
against
the
solubility
parameter
of
the
FSL's
for
the
reasons
given
by
McKinney
and
Wu
(1982).
The
pins
identified
at
the
top
of
Fig.
1.
At
each
experimental
point,
the
TABLE
2
SOLUBILITY
PARAMETERS3
OF
FOOD-SIMULATING
LIQUIDS
Solubility
Parameter
FSL
8
x
10-4
(J
2m-
3/2)
Examples
of
Foods
Simulated
Heptane
1.51
Vegetable
Oils,
Fats,
Meats
Ethanol
in
Water
100%
2.60
Water,
Light
Beverages,
75%
3.15
Alcohol,
Candy,
Syrups,
50%
3.70
Wine,
Beer
25%
4.24
0%
4.79
3Burrell
(1975).
Vol.
64
No.
11
1327
by guest on July 10, 2011 For personal use only. No other uses without permission.jdr.sagepub.comDownloaded from

1328
McKINNEY
&
WU
Composite
CO
Composite
SG
4
U
.-
-
Composite
VLA
Composite
M
1
3
c
4
2
3
4
8xIO-4
(J/2m-3/2)
2
Fig.
1
Hardness
ratio,
pi/pf,
vs.
FSL
solubility
par
commercial
composites.
The
corresponding
positions
of
in
water
solutions
and
heptane
(H)
are
indicated
at
the
Track
Length
or
Time
Fig.
2
Schematic
illustration
showing
distinction
havior
of
an
unsoftened
and
FSL-softened
composite.
the
upper
curve
depicts
the
removal
of
the
chemically-d.
rate
discontinuity,
which
differentiates
Region
I
from
from
a
change
in
the
wear
mode
for
which
wear
is
generation
of
subsurface
damage
resulting
from
fatigue
75%
50%~o
275%o
o0%
length
of
the
vertical
line
is
the
standard
deviation
from
N
indicated
replicate
determinations
on
different
specimens
of
a
given
composite.
Wear.
Previous
work
(McKinney
and
Wu,
1982;
Wu
and
McKinney,
1983)
showed
that
the
pin-and-disc
wear
behavior
may
be
depicted
schematically
as
in
Fig.
2.
For
an
unsoftened
composite,
which
may
result
from
either
no
pre-conditioning
5
or
pre-conditioning
in,
for
example,
heptane,
the
wear
rate
was
essentially
constant
at
a
low
value
approximating
1
[pm/
hr
under
the
conditions
specified
in
the
next
section.
With
a
chemically-softened
composite,
the
wear
rate
was
initially
very
large,
as
shown
in
Fig.
2.
The
large
initial
wear
is
attributed
I6
T
to
the
sudden
removal
of
the
chemically-damaged
surface
layer.
T6
3
After
this
layer
was
penetrated,
the
rate
of
wear
reached
an
apparent
steady
state.
The
steady-state
wear
rate
for
the
soft-
ened
composite
was
usually
found
to
be
larger
than
that
for
3
the
unsoftened
one,
which
indicates
that
some
damage,
or
change,
in
the
former
existed
well
below
the
visibly-damaged
surface
layer.
3
4
5
The
wear
behavior
was
sometimes
further
complicated
by
the
existence
of
rate
discontinuities
in
the
apparent
steady-state
regions
as
indicated
in
Fig.
2.
These
sudden
increases,
often
ameter,
v
,
for
four
approximating
a
factor
of
two
in
the
wear
rate,
were
occa-
the
various
ethanol
sionally
observed
for
both
the
softened
and
unsoftened
com-
top
of
the
Fig.
posites.
According
to
Atkinson
et
al.
(1978),
this
behavior
is
somewhat
general
and
results
from
wear-mode
conversions
at
which
the
wear
is
enhanced
by
fatigue.
Their
interpretation
is
confirmed
by
our
microdefect
analysis,
which
revealed
sub-
surface
damage
generated
in
Region
II,
but
essentially
none
in
Region
I,
as
depicted
in
Fig.
2.
As
expected
from
mechanical
damage
theories,
the
conversion
occurs
at
shorter
times
with
increasing
stress
(McKinney
and
Wu,
1982),
which
is
indic-
ative
of
a
mechanical
damage
mechanism.
In
Fig.
3,
the
initial
wear-track
depth
(at
five
revolutions)
for
three
composites
is
plotted
against
the
solubility
parameters
of
the
FSL's
identified
at
the
top
of
the
Fig.
The
vertical
bars
and
numbers
at
the
data
points
have
the
same
meaning
as
those
for
the
hardness
ratio.
The
data
for
the
microfilled
composite
(M)
are
not
included
in
Fig.
3.
Fig.
4
shows
the
wear-track
depth
vs.
disc
revolutions
over
the entire
wear
runs,
but
exclusive
of
the
initial
wear,
for
two
specimens
of
the
microfilled
composite
pre-conditioned
in
a
tened
75%
ethanol
25%
water
solution.
The
upper
curve
depicts
nposite
the
only
run
for
this
material
for
which
no
changes
in
the
wear
mode
were
observed.
The
lower
curve
is
an
example
of
the
other
20
specimens,
which
illustrates
the
discontinuity
that
occurred
at
about
10,000
disc
revolutions.
The
time
interval
Wear
Pin
between
adjacent
points
is
one
hour,
which
corresponds
to
ear
n
about
1500
revolutions.
The
wear
rates
shown
were
obtained
from
linear
regressions
over
the
regions
enclosed
by
parenthe-
--
ses.
The
average
value
of
the
number
of
revolutions
and
stan-
a
ged
/
dard
deviation
at
which
the
rate
discontinuity
occurred
was
rr
/9800
+
2900
over
the
20
wear
runs
taken
after
different
pre-
I
/
-^
conditioning
treatments,
including
storing
in distilled
water.
/
There
was
no
significant
correlation
between
these
abscissa
values
and
the
solubility
parameters
of
the
solvents.
The
av-
oftened
erage
increase
in
rate
and
its
standard
deviation
after
passing
nposite
through
the
discontinuity
was
47
+
25%.
The
smallest
change
observed
was
12%,
and
the
largest,
116%.
Fig.
5
shows
the
influence
of
pre-conditioning
in
water
on
the
strontium-glass-filled
composite
(SG).
The
wear-track
depths
between
wear
be-
are
plotted
against
disc
revolutions.
The
lower
curve
depicts
The
steep
rise
in
the
wear
for
a
specimen
pre-conditioned
in
air
at
37°
C
for
one
amaged
layer.
The
week.
The
upper
curve
is
that
for
a
specimen
pre-conditioned
Region
II,
results
in
distilled
water
at
the
same
temperature.
This
composite
was
enhanced
by
the
the
only
one
tested
here
for
which
pre-conditioning
in
water
(over
that
in
air)
had
any
significant
effect
on
the
wear.
With
3
2
c
CD
F-
J
Dent
Res
November
1985
Q.
by guest on July 10, 2011 For personal use only. No other uses without permission.jdr.sagepub.comDownloaded from

CHEMICAL
SOFTENING
AND
WEAR
OF
DENTAL
COMPOSITES
H
100%
75%
50%
25%
0%
H
100%
75%
50%
25%
0%
Cc
)mposite
SG
i4
<
_
14
Fig.
3
Initial
wear,
si
(over
the
first
five
revolutions),
vs.
FSL
solubility
pa-
rameter,
8,
for
composites
CQ,
SG,
and
VLA.
All
Figs.
use
the
same
scale.
8xiO-4
(J'/2m-3/2)
r
Composite
VLA
2
0
I
>3
4
8X10-4
(J'/2m-3/2)
40
Or)
2
3
RX10-4
(Rev)
100
60
-
LO
uo
250
200
150
_()
1oo00
50
Fig.
4
Track
depth,
s,
vs.
disc
revolutions,
R,
for
the
microfilled
composite
(M)
pre-conditioned
in
75%
ethanol.
The
corresponding
time
interval
between
adjacent
points
is
one
hour.
The
lower
curve
depicts
the
observation
of
the
rate
discontinuity
(see
Fig.
2)
which
is
not
seen
in
the
upper
curve.
3.27
nm/rev
i/
0"
*
0
0
0
0
*
*
0
0
0
0
0
0
14
nm/rev
nm/rev
a-
0
1I
2 3
4
5
R
X
10-4
(Rev)
Fig.
5
Track
depth,
s,
vs.
disc
revolutions,
R,
for
the
strontium-
glass-filled
composite.
The
upper
and
lower
curves
show
the
wear
for
this
composite
pre-conditioned
in
water
and
air,
respectively.
the
water-pre-conditioned
specimen,
the
wear
rate
diminished
with
time
or
track
length.
The
indicated
rates
at
the
beginning
and
end
were
obtained
from
linear
regressions
over
the
regions
enclosed
by
parentheses.
All
of
the
SG
specimens
pre-condi-
tioned
in
any
of
the
ethanol
solutions
showed
wear
similar
to
that
shown
in
the
upper
curve
(Fig.
5)
for
water.
All
those
pre-
conditioned
in
heptane
corresponded
to
the
lower
curve
for
air.
In
Fig.
6,
the
steady-state
wear
rates,
plotted
against
the
solubility
parameters
of
the
FSL's,
are
shown
for
Composites
CQ,
VLA,
and
M.
The
rates
for
Composite
SG
are
excluded
here
because
of
their
time-dependent
behavior
(mentioned
in
the
preceding
paragraph).
With
Composites
CQ
and
VLA,
the
rate
discontinuity
(wear
mode
change)
was
observed
in
only
a
few
cases,
as
indicated.
Discussion.
All
of
the
composites
tested
revealed
damage
resulting
from
pre-conditioning
in
chemical
food-simulating
liquids,
as
seen
from
decreases
in
surface
hardness
and
wear
resistance.
The
greatest
overall
effect
was
obtained
from
pre-conditioning
in
a
75%
ethanol
in
water
solution,
for
which
the
solubility
pa-
T
Composite
CQ
15
15
10
E
a
5
C..
orn
U
)
-
Ij
Vol.
64
No.
11
1329
)
L-
by guest on July 10, 2011 For personal use only. No other uses without permission.jdr.sagepub.comDownloaded from

1330
McKINNEY
&
WU
H
!00%
75%
50%
25%
0%
H
Composite
VLA
T
23
4
51
8
x
0-4
(J
l/2m-3/2)
100%
75%
50%
25%
0%
Composite
M
3L
C3
3
3
3
3
3
3
T
3
3
1
Fig.
6
Steady-state
wear
rate,
m,
vs.
FSL
solubility
parameter,
8,
for
three
commercial
composites.
Closed
circles,
Region
I.
Open
circles,
Region
II.
Re-
gion
I
is
the
early
wear
prior
to
formation
of
subsurface
damage
(see
Fig.
2).
1
2
3
4
5
rameter
value
is
about
3
x
104
J12m
-3/2.
This
result
implies
that
any
oral
or
food-ingredient
component
having
a
solubility
parameter
approximating
this
value
will
produce
damage
in
BIS-GMA-based
composites.
The
extent
of
damage
may
de-
pend
somewhat
on
the
diffusion
rate,
which,
in
turn,
depends
on
the
molecular
weight
of
the
penetrant.
For
the
most
part,
the
damage
mechanism
is
attributed
to
softening
of
the
polymer
matrix,
which
results
in
its
partial
removal
at
the
surface.
From
the
hardness-ratio
data
shown
in
Fig.
1,
one
can
see
that
all
composites
tested
revealed
a
similar
trend
in
this
prop-
erty
with
solubility
parameter.
In
all
cases,
the
maximum
soft-
ening
effect
(maximum
pi/Pf)
occurred
from
pre-conditioning
in
the
75%
ethanol
solution
(8=
3.1
x
104).
The
ranking
in
terms
of
decreasing
pi/pf
was
CQ,
SG,
M,
and
VLA.
In
all
cases
except
VLA,
a
small
but
significant
increase
in
hardness
was
observed
during
pre-conditioning
in
heptane.
Two
possi-
ble
explanations
are
that
heptane
reduces
oxygen
inhibition
during
post-curing
and
eliminates
leaching
out
of
silicon
and
combined
metals
in
fillers,
which
might
occur
from
pre-con-
ditioning
in
aqueous
solutions
(Soderholm,
1983).
The
corresponding
wear
was
complicated,
with
vast
distinc-
tions
observed
between
the
behavior
of
the
different
materials
tested.
As
with
the
hardness
ratio,
the
initial
wear
maximized
at
a
solubility
parameter
value
of
about
3
x
104,
as
seen
in
Fig.
3.
However,
the
extent
of
wear
varied
considerably
over
the
materials.
Since
the
extent
of
damage
generated
in
each
specimen
was
quite
variable
under
the
same
pre-conditioning
conditions,
the
standard
deviations
expanded
rapidly
with
the
extent
of
softening.
With
VLA
the
initial
wear
was
very
low,
but
did
tend
to
maximize
at
8
=
3
x
104.
These
low
values
may
be
attributed
to
one
or
more
of
the
following:
This
restorative
uses
a
polymer
base
slightly
different
from
that
of
the
others
and
therefore
may
be
less
sensitive
to
the
FSL's.
Since
visible-light-activated
composites
are
single-component
systems,
mixing
should
be
more
complete,
thereby
avoiding
possible
domains
of
under-
cured
resin
and
the
introduction
of
air
during
mixing,
which
would
inhibit
curing.
Visible-light-activated
composites
have
very
hard
surfaces
for
which
the
degree
of
cure
diminishes
with
depth.
These
surfaces
appear
to
be
very
solvent-resistant.
The
high
wear
resistance
here
is
in
accord
with
the
microdefect
analyses
on
this
composite,
which
revealed
very
little
surface
damage
on
this
material
after
pre-conditioning.
The
initial
wear
was
extremely
variable
for
the
microfilled
composite.
For
this
reason,
the
data
for
this
material
were
not
included
in
Fig.
3.
Initial
wear-track
depths
for
this
material
varied
from
0
to
200
pim.
This
variable
behavior
is
attributed
to
extensive
softening
of
the
surface
and,
in
some
cases,
to
a
complete
removal
of
the
damaged
region
before
wear
testing
commenced.
The
softening
of
a
microfilled
composite
is
ex-
pected
to
be
more
severe
than
that
of
a
conventional
one
using
the
same
polymer
base,
because
the
former
employs
a
consid-
erably
larger
polymer/filler
ratio.
Unpublished
data
show
that
the
initial
wear
resistance
of
this
material,
pre-conditioned
in
a
75%
ethanol
in
water
solution,
improved
markedly
as
the
degree
of
cure
was
raised
by
increasing
the
cure
temperature
well
above
37°C.
Except
for
SG,
the
steady-state
wear
rates
shown
in
Fig.
6
for
Region
I
(see
Fig.
2)
show
correlations
with
solubility
parameters
similar
to
those
with
the
hardness
ratios
displayed
in
Fig.
1.
With
CQ
and
VLA,
discontinuities
in
the
wear
rates
were
occasionally
observed.
With
M,
the
occurrence
of
this
phenomenon
(as
revealed
by
an
example
in
Fig.
4)
was
very
reliable.
Out
of
21
wear
runs
on
this
material,
pre-conditioned
under
different
conditions,
rate
discontinuities
occurred
in
all
but
one.
Although
the
wear
in
Region
I
had
the
usual
corre-
lation
with
solubility
parameter,
there
was
no
significant
cor-
relation
in
Region
II
(see
Fig.
6).
As
stated
earlier,
the
existence
of
these
discontinuities
is
attributed
to
wear-mode
conversion
for
which
the
enhanced
wear
rate
suddenly
results
from
the
gradual
build-up
of
subsurface
damage
at
earlier
times.
Since
the
enhanced
wear
in
Region
II
resulted
from
considerable
mechanically-generated
damage,
the
matrix
softening
from
the
FSL's
seemed
to
have
little
influence
here.
Although
the
in-
terest
in
this
phenomenon
is
more
academic
than
utilitarian
at
this
time,
its
observation
indicates
the
possibility
of
an
accel-
erated
wear
process
with
in
vivo
aging.
The
fact
that
the
steady-state
wear
rates
in
Region
I
pass
through
a
maximum
with
solubility
parameter
for
the
three
cases
shown
in
Fig.
6
is
indicative
of
some
damage
being
inflicted
by
the
FSL's
well
below
the
specimen
surfaces.
This
postulated
damage
is
not
visible
by
microdefect
analyses.
This
50%
25%
0%
ih
100%
75°
Composite
CQ
>
03
ai)
E
C:
Ec
J
Dent
Res
November
1985
by guest on July 10, 2011 For personal use only. No other uses without permission.jdr.sagepub.comDownloaded from

CHEMICAL
SOFTENING
AND
WEAR
OF
DENTAL
COMPOSITES
result
implies
that
some
swelling
of
polymer
below
the
dam-
aged
region
resulted
without
removal
or
destruction
of
mate-
rial.
With
SG,
the
existence
of
an
additional
damage
mode
from
pre-conditioning
in
water
is
apparent.
As
shown
in
Fig.
5,
the
steady-state
wear
of
SG
was
enhanced
considerably
by
pre-
conditioning
in
water.
The
ethanol
solutions
also
enhanced
subsequent
wear
in
a
similar
manner,
but
heptane
did
not.
A
plausible
explanation
for
the
accelerated
wear
is
that
this
par-
ticular
glass
is
damaged
by
exposure
to
water.
Microdefect
analysis
by
Soderholm
(1981)
on
some
fractured
dental
com-
posites
revealed
damage
to
the
filler
interface
under
certain
conditions
after
prolonged
storage
in
water.
The
autocatalytic
stress-corrosion
mechanism
by
which
modified
glasses
may
be
attacked
by
water
has
been
described
by
Charles
(1958).
Quartz
is
relatively
inert
to
this
kind
of
corrosion.
Accordingly,
it
is
suspected
that
radiopaque
composites
employing
certain
mod-
ified
glasses
will
be
more
susceptible
to
this
kind
of
degra-
dation
than
will
those
employing
conventional
quartz
fillers.
This
hypothesis
is
corroborated
by
comparison
of
the
wear
data
on
the
quartz-
and
strontium-glass-filled
composites
pre-con-
ditioned
in
water.
In
a
recent
paper,
Soderholm
et
al.
(1984)
presented
a
more
explicit
and
detailed
explanation
of
the
role
of
stress
corrosion
on
the
degradation
of
dental
composites.
The
results
of
this
work
suggest
certain
possible
improve-
ments
in
composite
restoratives
to
improve
their
durability.
For
example,
increasing
the
degree
of
cure
of
the
matrix
pol-
ymer
will
inhibit
diffusion
of
penetrants,
and
the
additional
cross-linking
will
reduce
swelling
and
damage
by
solvents.
Studies
have
been
made
on
wear
reduction
by
increasing
the
degree
of
cure
through
elevation
of
the
cure
temperature.
Al-
though
sufficient
elevations
in
cure
temperature
may
not
be
practical
for
restorations
placed
in
situ,
this
method
affords
a
way
to
connect
degree
of
cure
with
durability.
Other
methods
to
increase
the
degree
of
cure
usually
involve
changes
in
the
chemical
structure
of
the
polymer.
An
alternative
to
increasing
the
degree
of
cure
is
to
develop
or
modify
polymers
to
obtain
solubility
parameters
with
values
lower
than
those
encountered
in
foods.
Acknowledgment.
The
authors
appreciate
the
assistance
and
advice
from
Dr.
Nelson
W.
Rupp
of
the
American
Dental
Association
Health
Foundation
Paffenbarger
Research
Center
in
connection
with
specimen
preparation
for
these
tests.
REFERENCES
American
Society
for
Testing
and
Materials
(1984):
Knoop
Indenta-
tion
Hardness
of
Glass,
Vol.
15.02,
Designation
C730,
Philadel-
phia,
PA:
ASTM,
pp.
421-486.
ATKINSON,
J.P.;
BROWN,
K.J.;
and
DOWSON,
D.
(1978):
The
Wear
of
High
Molecular
Weight
Polyethylene,
Trans
Am
Soc
Mech
Eng
100:208-218.
BURRELL,
H.
(1975):
Solubility
Parameter
Values.
In:
Polymer
Handbook,
J.
Brandrup
and
E.H.
Immergut,
Eds.,
New
York,
NY:
John
Wiley
&
Sons,
Inc.,
pp.
iv,
337-359.
CHARLES,
R.J.
(1958):
Static
Fatigue
of
Glass,
I,
J
Appl
Phys
29:1549-
1553.
Food
and
Drug
Administration
(
1976):
FDA
Guidelines
for
Chemistry
and
Technology
Requirements
of
Indirect
Additive
Petitions,
Washington,
DC:
FDA,
March,
1976.
HILDEBRAND,
J.H.
and
SCOTT,
R.L.
(1950):
Solubility
of
Non-
Electrolytes,
3rd
ed.,
New
York,
NY:
Reinhold
Publishing
Co.,
pp.
424-434.
McKINNEY,
J.E.
(1982):
Apparatus
for
Measuring
Wear
of
Dental
Restorative
Materials,
Wear
76:337-347.
McKINNEY,
J.E.
and
WU,
W.
(1982):
Relationship
Between
Sub-
surface
Damage
and
Wear
of
Dental
Restorative
Composites,
J
Dent
Res
61:1083-1088.
OKAZAKI,
M.
and
DOUGLAS,
W.H.
(1984):
Comparison
of
Sur-
face
Layer
Properties
of
Composite
Resins
by
ESCA,
SEM
and
X-ray
Diffractometry,
Biomaterials
5:284-288.
SODERHOLM,
K.J.
(1981):
Degradation
of
Glass
Filler
in
Experi-
mental
Composites,
J
Dent
Res
60:1867-1875.
SODERHOLM,
K.J.M.
(1983):
Leaking
of
Fillers
in
Dental
Com-
posites,
J
Dent
Res
62:126-130.
SODERHOLM,
K.J.
(1984):
Hydrolytic
Degradation
of
Dental
Com-
posites,
J
Dent
Res
63:1248-1254.
WU,
W.
and
COBB,
E.
(1981):
A
Silver
Staining
Technique
for
Investigating
Wear
of
Restorative
Dental
Composites,
J
Biomed
Mater
Res
15:343-348.
WU,
W.
and
McKINNEY,
J.E.
(1982):
Influence
of
Chemicals
on
Wear
of
Dental
Composites,
J
Dent
Res
61:1180-1183.
WU,
W.;
TOTH,
E.E.;
MOFFA,
J.F.;
and
ELLISON,
J.A.
(1984):
Subsurface
Damage
Layer
of
in
vivo
Worn
Dental
Composite
Res-
torations,
J
Dent
Res
65:675-680.
Vol.
64
No.
11
1331
by guest on July 10, 2011 For personal use only. No other uses without permission.jdr.sagepub.comDownloaded from