Shear, torsional, and tensile bond strengths of

ceramic brackets using three adhesive

filler concentrations
Alan J. Ostertag, DDS, MS, a Virendra B. Dhuru, BDS, MSc, a Donald d. Ferguson, DMD, = and
Ralph A. Meyer, Jr., PhD d
Marshfield and Milwaukee, Wis.

The effect of changes in adhesive filler concentration on the shear, torsional, and tensile bond
strength of a chemical, a mechanical, and a chemical/mechanical retained ceramic bracket was
evaluated. Two hundred ten bovine teeth were bonded with one of three ceramic brackets using a
30%, 55%, or 80% filled adhesive. The brackets were debonded with a shear, torsional, or tensile
force to test the bond strength and the site of bond failure. No significant difference was found in the
shear, torsional, or tensile bond strength of each ceramic bracket type in relation to changes in the
adhesive filler concentration. However, there was a trend toward increased bond strength with
increasing filler concentration. Combining the data according to adhesive type revealed that the 80%
filled adhesive displayed a significantly greater shear bond strength than the 30% or 55% filled
adhesive and a greater torsional bond strength than the 30% filled adhesive. This supports the
hypothesis of increased bond strength with increased adhesive filler concentration. The mechanically
retained ceramic bracket showed greater shear bond strength and maximum shear bond strength in
torsion than the chemical or chemical/mechanical retained ceramic bracket. The tensile bond
strength of the mechanically retained ceramic bracket was similar to that of metal brackets reported
in other studies, and the failure site was at the bracket-adhesive interface. (AM J ORTHOD DENTOFAC
ORTHOP 1991 ;100:251-8.)

C e r a m i c brackets were created when the de-
mand for "esthetic braces" increased. Along with their
improved esthetics, ceramic brackets introduced some
unique clinical characteristics: abrasiveness to enamel,
brittleness leading to fracture, and increased bond
strengths leading to enamel fracture. Once these char-
acteristics were acknowledged, manufacturers intro-
duced elastic "cushions" to cover the bracket occlusal
surface to reduce abrasion, altered bracket design to
help reduce bracket fracture, created new debonding
instruments, and issued more detailed instructions to
assist in bracket removal.
Since ceramic brackets were introduced, there have
been several bonding studies using ceramic brack-
ets. t-5 These studies have relied on commercially avail-
From Marquette University School of Dentistry,
Based on a thesis in partial fulfi]lment of the requirements for the degree of
Master of Science,
=In private practice in Marshfield, Wis.
bActing chairman and associate professor of the Department of Dental Materials
and associate professor of operative dentistry.
~Assistant professor and chairman of the Department of Orthodontics,
dProfessor of physiology.
8/1],24328
able adhesive systems to test bond strength and have
concentrated on differences in bracket base charac-
teristics or major changes in adhesive filler size-
concentration combinations. With the recent concern
expressed about the debonding difficulties encountered
with ceramic brackets, a logical question to be explored
would be whether there is a specific resin-bracket com-
bination that will aid in bracket removal and yet main-
tain adequate bond strength.
The purpose of this study was to evaluate and com-
pare the shear and torsional bond strength and the site
of bond failure of a chemical, a mechanical, and a
chemical/mechanical retained ceramic bracket to spe-
cific changes in adhesive filler concentrations. In ad-
dition, the tensile bond strength and the site of bond
failure of the mechanically retained bracket were
tested.
METHODS
Three commercially available ceramic orthodontic brack-
ets, one single crystalline, the others polycrystalline, were
used in this study (Table I). All the brackets were standard
edgewise 0.022 × 0.028-irtch slot mandibular incisor brack-
ets. The Starfire single :crystalline bracket base contained no

251
252 Ostertag et al. Am, J, Orthod. Dentofac. Orthop.
September 1991

Table I. C e r a m i c brackets used in the study

Bracket base
Name Crystalline form Retention dimensions Manufacturer
Allure IIl Polycrystalline Chemical/mechanical* 2.40 x 3.40 mm GAC International, Central Islip, N.Y.
Starfire Single crystalline Chemical** 3.04 x 3.82 mm "A" Company, San Diego, Calif.
Transcend 2000 Polyerystalline Mechanical*** 2.60 x 3.34 mm Unitek Corp., Monrovia, Calif.

*Silane-treated bracket base containing four rectangular retention slots.

**Silane-treated bracket base.
***Particles fused to bracket base.

Table II. A d h e s i v e s used in the study
Compositlon* l Batchnumber"l Manufacturer
Silica-filled BIS-GMA Paste A: 145-49-C All adhesives were
resin (30% by Paste B: 145-47-A specially formu-
weight) luted by Reli-
ance Orthodon-
tic Products,
Inc., Itasca, Ill.
Silica-filled BIS-GMA Paste A: 145-47-B
resin (55% by Paste B: 145-49-D
weight)
Silica-filled BIS-GMA Paste A: 145-47-C
resin (80% by Paste B: 145-49-E
weight)
*All fillers were 5 to 15 Ixm in size.
mechanical retention but was treated with silane by the man-
ufacturer to enhance retention.
The polycrystalline brackets consisted of two types,
Transcend 2000 and Allure UI. Their bracket bases differed
in that the Transcend 2000 bracket base had fused particles
to provide retention, whereas the Allure III bracket base had
four rectangular retention slots and had been treated with
silane.
Three specially formulated BIS-GMA type, autopoly-
merizing composite orthodontic adhesives were used in this
study (Table II). The size of the fillers in the adhesives ranged
from 5 v m to 15 ~m. The filler concentration was the only
variable that changed among the adhesives. Filler concentra-
tions selected were 30%, 55%, and 80% by weight. The
sealant used was Phase II sealant (Reliance Orthodontic Prod-
ucts, Inc., ltasca, I11.), an unfilled BIS-GMA adhesive.
Two hundred ten extracted bovine teeth were obtained
and stored at room temperature in distilled water. 6 The teeth
were mandibular incisors taken from approximately 15
month-old Holstein steers, raised in the same locale in central
Wisconsin. The teeth were randomly assigned to one of 21
treatment groups containing 10 teeth per group. The bracket,
adhesive, and force mode combinations are listed in Ta-
ble III.
All the teeth were prepared and the brackets were bonded
by a similar technique in a random order. The facial surface
of each tooth was cleaned with a slurry of pumice for 30
seconds with a rubber cup driven by a low-speed handpiece.
The surface was rinsed with water to remove any pumice or
debris and dried for 15 seconds with an air stream. A 37%
orthophosphoric acid etching solution was applied with a
sponge pellet to the facial surface of each tooth for 20 seconds.
The tooth was then rinsed with water for 30 seconds and air
dried for 15 seconds.
The unfilled sealant was mixed in equal parts, and a thin
coat was applied to the enamel surface. The adhesive base
and catalyst pastes were mixed on a paper mixing pad ac-
cording to the manufacturer's instructions. The pastes were
mixed until an even color was obtained and then placed on
the bracket base. The bracket was then positioned on the facial
surface of the tooth with a dental explorer. A level area of
the enamel surface was chosen to provide an area of best fit.
Pressure was applied to the bracket, simulating clinical chair-
side procedures, to express any excess adhesive from between
the bracket and the enamel surface. The excess was then
removed from the periphery of the bracket base with an ex-
plorer. To prevent any reduction in bond strength, the tooth
was left undisturbed for 20 minutes at room temperature. After
this time, the bonded tooth was embedded in a cylinder of
acrylic and stored in 37 ° C distilled water for 48 hours.
Each specimen was tested in one of three force modes:
shear, torsion, or tensile. The shear bond strengths were mea-
sured with an Instron testing machine (Instron universal test-
ing instrument, Model TTC, Instron Corp., Canton, Mass.).
Each shear test specimen was placed into a Vise that positioned
the tooth surface parallel to the direction of load application.
A shear load was applied at a crosshead speed of 0.50 nun
per minute, and the maximum load necessary to debond the
bracket was recorded.
The torsional bond strengths were measured with a torque
meter (Tohnichi torque gauge, Model BTG120Z, Tohnichi
America Corp., Skokie, Ill.). Each specimen was fixed in a
vise, and the torque meter containing a custom-fabricated
torquing wrench was placed over the bracket. The load was
applied manually in such a manner as to simulate clinical
chairside conditions. The maximum torque necessary to de-
bond the bracket was recorded.
The tensile bond strengths were measured with the Instron
universal testing machine. Only the Transcend 2000 brackets
were tested under a tensile force because no other manufac-
turer recommended that their brackets be debonded under
tensile force conditions. It was believed that intrabracket frac-
tures might occur if the other brackets were tested in tension.
Each specimen was suspended from the upper member of the
Volume tO0 Shear, torsional, and tensile bond strengths of ceramic brackets 253
Number 3

Table III. Mean bond strength and torque values

1--
Bond strength Torque
Force
Group Bracket Adhesive mode MPa ] (SD) N" m _ (SD)

1 Transcend 2000 30 Shear 17,02 (8.78) -- --

2 Transcend 2000 55 Shear 17.79 (11.08) --
3 Transcend 2000 80 Shear 25.92 (6.83) --
4 Starfire 30 Shear 10.60 (3.49) -- --
5 Starfire 55 Shear 12.76 (5.65) ~ --
6 Starfire 80 Shear 18.19 (9.03) --
7 Allure III 30 Shear 8,17 (2.57) --
8 Allure III 55 Shear 10.50 (4.64) --
9 Allure III 80 Shear 11,07 (5.25) ~
10 Transcend 2000 30 Torsion 56.31 (8.69) 0,29 (0.04)
11 Transcend 2000 55 Torsion 57.40 (12.45) 0.29 (0.06)
12 Transcend 2000 80 Torsion 58.69 (9.52) 0.30 (0,05)
13 Starfire 30 Torsion 28.00 (3.24) 0,22 (0.03)
14 Starfire 55 Torsion 36.79 (14.16) 0,29 (0,1 I)
15 Starfire 80 Torsion 38.05 (9.56) 0,30 (0.08)
16 Allure III 30 Torsion 41,65 (6,03) 0.19 (0.03)
17 Allure Ill 55 Torsion 41.43 (7,12) 0.19 (0.03)
18 Allure III 80 Torsion 48,46 (6.24) 0.22 (0,03)
19 Transcend 2000 30 Tensile 5.25 (0.92) --
20 Transcend 2000 55 Tensile 6.04 (1,30) --
21 Transcend 2000 80 Tensile 6,34 (1.14) --

Instmn machine with a custom-made fixture allowing self-
alignment of the tooth surface perpendicular to the line of
force, A custom-fabricated tensile debonding instrument,
fixed to the lower member of the Instron machine, was secured
to the mesial and distal sides of the bracket base. A tensile
load was applied at a crosshead speed of 0.50 mm per
minute, and the maximum tensile force necessary to de-
bond the bracket was recorded. All test specimens were
viewed under a stereomicroscope to determine the site of bond
failure.
The bracket base dimensions were measured to the nearest
0.01 mm with a traveling microscope (Gaertner Scientific
Corp. No. 234P, Chicago, II1.). The mesiodistal curvature of
the bracket base was not considered in the determination of
bracket base area. The radius of curvature of the mandibular
incisor bracket base was minimal; in addition, the adhesive
thickness minimized the influence of the radius on the bracket
base area,
RESULTS
The mean force and the standard deviation (SD)
required to induce bond failure in each treatment group
are shown in Table III. The means of the torque values
are reported as force times distance and also as the
maximum shear stress generated during torsion to dem-
onstrate the influence of bracket base dimensions on
the force levels generated. 7 The bond strength and
torque values shown in Table III are graphically illus-
trated in Figs. 1 through 4. Any brackets that fractured
during testing were omitted from the calculation of the
treatment group means. A summary of the bond failure
pattern is listed in Table IV,
A 3 × 3 factorial analysis of variance (ANOVA)
and a Tukey multiple comparisons test were employed
in the statistical analysis of the shear and torsionaltreat-
ment groups, The tensile treatment groups were sub-
jected to a single-factor analysis of variance. The fac-
torial ANOVA and the Tukey test revealed a significant
difference in the shear force levels among the three
bracket types, with the Transcend 2000 bracket being
significantly greater than the Starfire or Allure III
bracket. The three adhesives also show a significant
difference in shear force values, with the 80% filled
adhesive displaying a significantly greater bond strength
than the 30% or 55% filled adhesive. No statistically
significant interaction between adhesive and bracket
was found.
A factorial ANOVA and a Tukey test for the torsion
data revealed a significant difference in the N . m
torque levels among the bracket types, with the Tran-
scend 2000 and Starfire brackets being significantly
greater than the Allure III bracket. Conversion of the
torque values into maximum shear stress levels revealed
a significantly greater stress for the Transcend 2000
bracket than for the Starfire or Allure III bracket and a
greater stress for the Allure III bracket than for the
Starfire bracket. The adhesives also showed a signifi-
cant difference in torque levels, with the 80% filled
adhesive displaying a significantly greater bond strength
254 Ostertag el al. 4m. J. Orthod. Demofac. Orthop.
September 1991

30-

25-
T
•I" 20-
I--

Z
ILl
F- 15-
t,O
a
Z i"

0
m
Iz: lO-
<
LU

8-
Legend
f-/71 TRANSCEND
I ~ STARFIRE
ALLURE
i

~0% 55% 8O%

W E I G H T PERCENT OF FILLERS

Fig. 1. Shear bond strength of ceramic brackets and adhesives with varying filler contents.

70

T
a.
60

Z
//
¢~ so /-j
//
//
//
//
// //
,, ~ \\
~ 4o

w
~ k"N
r~
I--

<

O'~ 20-
:E .',.x.
=E ,%
\\
X
,< 1o- "..\ Legend
r-z2 TRANSCEND
STARFIRE
IEE] ALLURE
I

30% 55% 80%

W E I G H T PERCENT OF FILLERS

Fig. 2. Maximum shear stress in torsional debonding of ceramic brackets and adhesives with varying
filler contents,
Volume 100
Number 3 Shear, torsional, and tensile bond strengths of ceramic brackets 255

0,38 -

0.~-

A 0.25-
E
z
-i
UJ

o 0.2-
O
I--

z
Q 0.18 -

Z
O
tU

0.1-

0,08= Legend
EZ2 TRANSCEND
STARFIRE
ALLURE
i
30% 55% 80%
W E I G H T PERCENT OF FILLERS
Fig, 3. Debonding torque for ceramic brackets and adhesives with varying filler contents.

than the 30% filled adhesive. No significant interaction 7-

between adhesive and bracket was found. - T
Four 2 × 4 contingency "tables were designed and
subjected to X2 analysis to determine whether the major e-
sites of bond failure were independent of bracket and
adhesive type. The site of bond failure under a shear
and torsional force was dependent on bracket type. Un-
der both shear and torsional forces, Transcend 2000 and I-
Allure III brackets predominantly underwent combi-
Z
nation (COMB) bond failures, whereas Starfire brackets LIJ 4 -
predominantly underwent bond failures at the bracket- E
p.
adhesive (BA) interface (Table IV). Chi-square analysis CO
c~
o f adhesive type versus failure site in both shear and ~' 8 -
torsion revealed a nonsignificant difference. O
m
A single-factor ANOVA applied to the tensile data U.I
revealed no significant difference in the tensile bond ,..I
2-
strength of Transcend 2000 brackets in relation to Z
changes in the adhesive filler concentration, even LU , / / / / / / ~ ,
I-- //././///11
though there was a trend toward increased bond strength
with increased filler concentration. All bond failures 1-
occurred at the BA interface.

DISCUSSION 0-
Most bonding studies use commercially available 30% 55% 80%
adhesive systems that have different filler particle WEIGHT PERCENT OF FILLERS
sizes and concentrations. This makes comparisons Fig. 4. Tensile bond strength of Transcend 2000 ceramic brack-
between the groups difficult because of the increased ets and adhesives with varying filler contents,
 Am. J. Orthod. Dentofac. Orthop.
256 Ostertag et al. September 1991

Table 17. The location of bond failures in each treatment group

Transcend 2000
l 30
I Force
mode

Shear
I
BA

1
COMB

9
[ El

0
ENAM

0
BRF

0
Transcend 2000 55 Shear 3 5 0 0 2
Transcend 2000 80 Shear 1 9 0 0 0
Starl~re 30 Shear 6 3 0 0 1
Starfire 55 Shear 8 1 0 1 0
Starflre 80 Shear 9 0 0 0 1
Allure IH 30 Shear 2 8 0 0 0
Allure III 55 Shear 0 10 0 0 0
Allure III 80 Shear 1 9 0 0 0
Transcend 2000 30 Torsion 0 6 3 0 1
Transcend 2000 55 Torsion 3 6 0 0 1
Transcend 2000 80 Torsion 6 4 0 0 0
Starflre 30 Torsion 9 1 0 0 0
Starfire 55 Torsion 10 0 0 0 0
Starfire 80 Torsion 10 0 0 0 0
Allure III 30 Torsion 0 10 0 0 0
Allure III 55 Torsion 1 9 0 0 0
Allure IlI 80 Torsion 1 9 0 0 0
Transcend 2000 30 Tensile 10 0 0 0 0
Transcend 2000 55 Tensile 10 0 0 0 0
Transcend 2000 80 Tensile 10 0 0 0 0

BA, Bracket-adhesive interface. Adhesive may remain within the retention grooves or particles; however, a continuous layer of adhesive remains
on the enamel surface,
COMB, Combination failure, Failure is noted within the adhesive and at the enamel-adhesive interface and/or the bracket-adhesive interface,
El, Enamel interface failure. No adhesive is on the enamel surface. All the adhesive is retained on the bracket base.
ENAM, Enamel failure. Failure is noted within the enamel surface. Adhesive and enamel are present on the bracket base,
BRF, Intrabracket failure. Failure is within the bracket itself.

number of variables involved in the adhesive compo-
sition.
This experiment was designed to evaluate the influ-
ence of adhesive filler concentration on bond strength,
keeping the filler particle size constant. The filler par-
ticle size chosen was 5 to 15 p~m, because the most
common two-paste adhesives fall within this range.
• The results of this study indicate that there is an
increase in shear and torsional bond strength with in-
creasing adhesive filler concentrations, A comparison
of the treatment groups using the same bracket and force
mode with different adhesive filler concentrations re-
vealed a trend toward increase d bond strength; however,
n o significant statistical difference was present. It is
believed that a significant difference might have been
achieved with a larger sample size. Therefore, Under
each force mode, the treatment groups were combined
according to adhesive type to test the adhesive's influ-
ence With a greater sample size. The 80% filled adhesive
showed a significantly greater shear bond strength than
the 30% or 55% filled adhesive and a greater torsional
bond strength than the 30% filled adhesive. Therefore
the data support the hypothesis that, with increased ad-
hesive filler concentrations, the bond strength increases.
The influence of bracket base characteristics on
bond strength was determined by combining bracket
types. Transcend 2000 brackets displayed a signifi-
cantly greater shear bond strength and a significantly
greater maximum shear stress in torsion than Starfire
or Allure III brackets. This can be explained by the
many crystals fused to the base of the Transcend 2000
bracket. The line of force in both shear and torsion
force modes is generally perpendicular to the crystal
projections, whereas the Starfire bracket base is com-
pletely smooth and the Allure III bracket is smooth with
four retention grooves. Therefore bracket base sur-
face characteristics appear to greatly influence bond
strength.
An evaluation of fracture sites revealed that bracket
type significantly influences bond failure location. Un-
der shear and torsional forces Transcend 2000 and Al-
lure III brackets underwent predominantly COMB fail-
ures, whereas Starfire brackets underwent predomi-
nantly BA interface failures. All Transcend 2000
brackets failed at the BA interface under tensile forces,
suggesting that the direction of force application influ-
ences bond failure location. The influence of adhesive
type on bond failure location was not statistically sig-
Volume 100 Shear, torsional, and tensile bond strengths of ceramic brackets
Number 3
nificant. Of the 210 brackets tested, only one underwent
an enamel fracture and only six brackets (four Tran-
scend 2000 and two Starfire brackets) fractured. This
is in contrast to a previous study that reported 40%
enamel fractures when an autopolymerizing adhesive
and a ceramic bracket were used. 8 This discrepancy
may be due to differences in the bracket base surface
characteristics and the mechanism of retention.
Clinically, bonded brackets should be able to with-
stand forces generated by the treatment mechanics and
occlusion and yet allow easy debonding without injury
to the tooth. Reynolds 9 has reported that a maximum
tensile bond strength of 60 to 80 kg/cm 2 (5.9 to 7.9
MPa) would be adequate to resist treatment forces but
added that, in vitro, tensile test levels of 50 kg/cm 2
(4.9 MPa) have proved clinically acceptable. Enamel
is weakest under tensile forces (10 MPa) and strongest
under compression (378 MPa). t0.~ The Transcend 2000
brackets debonded with a tensile force range of 3.6 to
8.2 MPa, which is acceptably within tensile force
guidelines. In addition, failure at the BA interface in-
dicates that the bracket-adhesive bond is the weakest
under tensile debonding conditions, thereby protecting
the tooth from damage.
Guidelines for adequate shear and torsional bond
strength have not been reported; therefore, relying on
previous bonding studies using metal brackets as a
guide, the shear bond strength of ceramic brackets may
be analyzed. Shear bonding studies using metal brackets
have reported bond strengths in the 12.1 to 20.7 MPa
range. 3,4 The ceramic brackets in this study generated
shear bond strengths ranging from 8.2 to 25.9 MPa,
which is similar to that of metal brackets.
In a previous study in which torsional forces were
applied to metal and ceramic brackets, bracket base
dimensions were not considered in the analysis (Con-
stant DT, Ogro JE. Evaluation of three adhesives with
ceramic brackets. University of California, San Fran-
cisco, CA, 1988). Hence, comparison of data with this
study is impossible because of the differences in bracket
base dimensions. Another bond strength study using
only metal brackets reported the bracket base dimen-
sions; however, the torsional force was reported in
pound units, t2 The lever arm length was not provided;
therefore no comparisons could be made. No other tor-
sional studies could be found in the literature. Therefore
the torsional values in this study have been reported in
force times distance units, and the reference to the
bracket base dimensions have been given to allow
future comparison. Further, the torque values have
been converted to maximum shear stress so as to
include the influence of bracket base dimensions on
force levels.
257
The clinical application of these specially formu-
lated adhesives must be addressed. As discussed, the
bond strengths achieved with these adhesives have
fallen within clinically acceptable ranges. However, an-
other characteristic, viscosity, is a clinically important
consideration that influences both the penetration of the
adhesive into the retentive mechanism of the bracket
base and the ability of an adhesive to resist bracket drift
during direct bonding. Visual evaluation revealed that
the adhesives in this study adequately penetrated the
retentive mechanism of the bracket bases. However,
bracket drift was a problem as the 30% and 55% filled
adhesives allowed bracket drift after bracket placement.
The influence of filler concentration on the viscosity
remains a clinically important issue. The clinical ac-
ceptability of the 30% and 55% filled adhesives tested
in this study for direct chairside bonding of ceramic
brackets is questionable under their current viscosity
levels.
SUMMARY AND CONCLUSIONS
This study was designed to evaluate the effect of
changes in adhesive filler concentration on the shear,
torsional, and tensile bond strength of a chemical, me-
chanical, and a chemical/mechanical retained bracket.
The site of bond failure was also appraised.
The following conclusions were made:
1. No significant difference was found in the shear,
torsional, or tensile bond strength of each ce-
ramic bracket type to changes in the adhesive
filler concentration. However, there was a trend
toward increased bond strength with increasing
filler concentration.
2. Combining the data according to adhesive type
revealed that the 80% filled adhesive displayed
a significantly greater shear bond strength than
the 30% or 55% filled adhesive and a greater
torsional bond strength than the 30% filled ad-
hesive. Therefore an increase in adhesive filler
concentration resulted in an increase in shear and
torsional bond strength.
3. Transcend 2000 brackets exhibited a greater
shear bond strength and maximum shear stress
in torsion than Allure IU or Starfire brackets.
4. The tensile bond strength of Transcend 2000
brackets was the same as that of metal brackets
reported in other studies. All the failure sites
were at the BA interface,
5. A significant difference existed in the site of
bond failures with different bracket base de-
signs. Under shear and torsional forces, Tran-
scend 2000 and Allure III brackets underwent
predominantly COMB bond failures, whereas
251] Ostertag et al, Am, J, Orthod. Dentofac. Orthop,
September 199t

Starfire brackets underwent p r e d o m i n a n t l y B A
interface bond failures.
6. N o significant difference existed in the site o f
b o n d failure with changes in the a d h e s i v e filler
concentration.
We sincerely thank Mr, Paul Gange, president, Reliance
Orthodontic Products, Inc., Itasca, Ill., for supplying the
materials.
REFERENCES
1. Buzz~tta VA, Hallgren SE, Powers JM. Bond strength of ortho-
dontic direct-bondlng cement-bracket systems as studied in vitro.
AM J ORTHOI) 1982;81:87-92.
2, [wamoto H. Bond strength of new ceramic bracket enhanced by
silane coating. J Jpn Orthod Soc 1987;46:547-57.
3. Owinnett AJ. A comparison of shear bond strengths of metal
and ceramic brackets. AM J ORTHODDENTOFACORTHOP 1988;
93:346-8.
4. Odegaard J, Segner D. Shear bond strength of metal brackets
compared with a new ceramic bracket. AM 1 ORTHODDENTOFAC
ORTHOP 1988;94:20l-6.
5. Guess MB, Watanabe LG, Beck FM, Crall MG. The effect of
silane coupling agents on the bond strength of a polycrystalline
ceramic bracket. J Clin Orthod 1988;22:788-92.
6. Nakamichi I, Iwaku M, Fusayama T. Bovine teeth as possible
substitutes in the adhesion test. 1 Dent Res 1983;62:1076-81.
7. Eshbach OW, Souders M. HandboOk of engineering fundamen-
tals. New York: John Wiley & Sons, 1975:539-40.
8. Joseph VP, Rossouw E. The shear bond strengths of stain-
less steel and ceramic brackets used with chemically and light-
activated composite resins. AM J ORTHOD DENTOFACORTHOP
1990;97i121-5.
9. Reynolds IR. A review of direct orthodontic bonding. Br J Orthod
1975;2:171-8.
10. Phillips RW. Skinner's science of dental materials. Philadelphia:
WB Saunders, 1982:54-5.
!1. Combe EC. Notes on dental materials. New York: Churchill
Livingstone, 1986:113.
12, Thanos CE, MunhoUand T, Caputo AA. Adhesion of mesh-base
direct-bonding brackets. AM J ORTHOD1979;75'421-30.
Reprint requests to:
Dr. Virendra B. Dhuru
Marquette University School of Dentistry
604 N. 16th St,
Milwaukee, WI 53233

AAO MEETING CALENDAR

1992--St. Louis, Mo., May 9 to 13, St. Louis Convention Center
1993--Toronto, Canada, May 15 to 19, Metropolitan Toronto Convention Center
1994--Orlando, Fla., May 1 to 4, Orange County Convention and Civic Center
1995--San Francisco, Calif., May 7 to 10, Moscone Convention Center
(International Orthodontic Congress)
1996--Denver, Colo., May 12 to 16, Colorado Convention Center
1997--Philadelphia, Pa., May 3 to 7, Philadelphia Convention Center
1998--Dallas, Texas, May 16 to 20, Dallas Convention Center
1999~San Diego, Calif., May 15 to 19; San Diego Convention Center