European Journal of Onhodonria 16 (1994) 27-34 O 1994 European Orthodontic Society

Use of an adhesive resin for bonding orthodontic brackets

A. J. Ireland* and M. Sherriff**
•Department of Orthodontics, Royal United Hospital, Bath, and **Department of Dental Materials
Science, UMDS of Guy's and St Thomas' Hospitals, London

SUMMARY To date, most successful bonding agents used in orthodontics rely on mechanical
retention to both the enamel and bracket base. Chemical adhesion to enamel as seen with
glass ionomer cements, and to the silanated base of ceramic brackets have been tried.
Recent developments in resin formulation have led to the production of adhesive diacrylate
resins capable of forming adhesive bonds to certain metals including stainless steel. The
aim of this experiment was to compare such a resin, Panavia EX, with a more conventional
'no-mix' orthodontic bonding resin. Two different base retention mechanisms were used,
and the effect of rebonding and differing environmental conditions were also investigated.
The results indicated that Panavia EX could produce greater bond strengths than the more
conventional bonding resin. Of the two base retention systems tested, braised mesh bases
gave consistently greater bond strengths than the cast base, although no base/resin specifi-
city could be detected. Re-using the same brackets showed rebond strengths to be signific-
antly lower than initial bond strength although the results indicated the adhesive resin was
still able to bond more effectively to these used brackets than the conventional resin.
Environment had the greatest effect on bond strength, such that following environmental
exposure there was no significant difference between the two resins. This latter factor, and
in particular the more complex bonding technique required for the adhesive resin, means
that Panavia EX cannot be recommended for orthodontic use in its present form.

Introduction enamel without the need for acid etching. Cooke

Bonding of orthodontic brackets has largely
superseded banding as a method of retaining
attachments to all but the posterior teeth.
Conventionally, retention of the bonding resin
to both the bracket base and enamel is mechan-
ical. For enamel the potential for mechanical
interlock with the resin is only realised after
prior acid etching (Buonocore, 1955).
Orthodontic bracket bases are available with
numerous retentive designs in an attempt to
optimize the mechanical bond with the resin. If
an adhesive system were developed then there
may be several theoretical benefits, including:
1. no enamel etching;
2. a less sophisticated bracket base design which
would be cheaper and easier to produce
and recycle;
3. a more coherent joint would be produced
which would be less liable to fail by fatigue
or moisture permeability.
Materials are available which can bond to the
(1990) found glass ionomer cements to be
acceptable for use on anterior teeth in a 2-year
in vivo study. Such cements still rely on mechan-
ical interlock to the bracket base. They have
not found wide acceptance as orthodontic bond-
ing agents due to their greater bond failure and
more critical mixing technique.
In recent years, silane coupling agents have
been used with ceramic brackets enabling a
simpler, smooth base to be produced. The silane
is able to form chemical bonds with both the
ceramic base and the bonding resin. However,
the adhesion is often so good that there is a
high risk of enamel failure at debond (Joseph
and Rossouw, 1990; Winchester, 1991).
Consequently, silanes have now been dropped
by some manufacturers, in favour of a
roughened base that again relies on mechanical
interlock with the resin.
Enamel etching and bracket bases utilizing
mechanical interlock are, therefore, still the
mainstay of present day orthodontic bonding.
An adhesive resin, able to adhere to a metal
28
instead of a ceramic bracket base, may still be
advantageous. Cheaper, less sophisticated bases
could be used, which would be easy and cheap
to recycle. In order to remove the risk of enamel
fracture at debond, the adhesive resin itself
could be modified in order to make it the weak
link in the bonded joint, without necessarily
making the bond strength less than that required
for long-term clinical use. Cohesive failure
within the resin could therefore be a design
feature of such a bonding agent. Residual resin
on the enamel surface after debond may also
be easier to remove. Recent advances in restor-
ative dentistry have led to the development of
a number of resins capable of adhesion to
metals such as stainless steel (Tabata and
Amano, 1985). These materials should, there-
fore, be capable of adhesion to orthodontic
brackets. Retention to enamel is still
mechanical.
The aim of this experiment was to determine
the bonding ability of such a resin, Panavia EX
(Cavex, Haarlem, Holland) and to compare it
with a resin designed specifically for orthodontic
use, Bond Fast (Advanced Orthodontics, New
York, USA). Panavia EX was chosen because
it was also designed to overcome the deficiency
found in many resins for bonding metals to
tooth substance, namely a significant decrease
in bond strength on exposure to moisture.
Increased moisture resistance is achieved by use
of a novel phosphate monomer (Gettleman
et ah, 1986). Two brackets of similar nominal
area, but with different retentive designs were
chosen for testing with one being of a braised
mesh design (Minimesh, Ormco, Glendora,
USA) and the other a cast slot base relying
macroscopically on the near parallel-sided slots,
and microscopically on a rough, cast surface
(Edgeway, Ortho Organisers, San Diego, USA).
Only metal brackets were used because of the
adhesive properties of Panavia EX. The effect
of rebond and environment were also studied.
Materials and methods
In this experiment shear forces were applied to
brackets bonded to human enamel using the
resins under test. To standardize bond testing,
human premolars extracted for orthodontic pur-
poses were mounted in cold cure acrylic with
the flattest expanse of buccal enamel just below
the acrylic surface. After the resin had cured,
A. J. IRELAND AND M. SHERRIFF
this surface was polished on a Struers polishing
machine to produce a reproducible, flat enamel
surface of sufficient size to accommodate a
bracket base. A small relief channel was then
cut in the acrylic around the enamel to ensure
later bonding was confined to the prepared
enamel surface. Prior to bonding, the enamel
was etched with the etchant supplied for 1
minute, followed by rinsing with water for 30
seconds and drying with oil-free compressed air.
The resins were mixed in accordance with
the manufacturers' (Cavex and Advanced
Orthodontics) instructions and the brackets
bonded to the prepared enamel surfaces as
follows.
Bondfast
This bonding agent is classed as 'no-mix', and
consists of two parts, a liquid primer and a
filled paste. The primer is applied to both the
enamel and bracket surfaces, and setting occurs
rapidly when the paste is sandwiched between
them. The paste is therefore placed onto the
base and the bracket is then pushed firmly into
position on the tooth surface using a Mitchell's
trimmer. Excess resin is removed from around
the margins with a probe.
Panavia EX
This material is also supplied as a two-
component bonding agent, but in this case a
liquid and powder are premixed before being
applied to the bracket base. The cure is retarded
in the presence of oxygen, thus working time
was prolonged by spreading the material thinly
over the mixing pad (Fukushima et al., 1985).
Several brackets were bonded with a single mix.
Once the bracket was in position on the tooth,
excess resin was removed with a probe. The
margins were then sealed with Oxyguard (sup-
plied with the kit) to create the necessary anaer-
obic environment to achieve a full cure.
The two-bracket base retention systems under
test were braised gauze mesh (Minimesh.
Ormco, Glendora, USA) and an integral system
of cast slots (Edgeway. Ortho Organisers, San
Diego, USA). Lower incisor brackets were
chosen because they have the flattest bases,
thereby minimizing the effect of variation in
film thickness. The nominal base areas were
10.21 mm2 and 8.64 mm2, respectively.
Samples were tested to failure on an Instron
Universal Testing Machine (Model 1193,
ADHESIVE BRACKET BONDING
Instron, High Wycombe, Bucks) at a cross-head
speed of 2 mm/minute. Each sample was
mounted in the Instron with a custom made
testing jig (Fig. 1), such that the predominant
mode of loading was shear (Ireland and Sherriff,
1988). Testing was performed after 24 hours
bench curing or after water immersion for 1
week at 37°C. The locus of failure in each case
was determined using a low power lens (x 2).
Following these initial tests all the used
brackets were rebonded using the resins and
techniques previously described, and then
retested under the same conditions. Resin
remaining on the enamel surfaces after the
initial tests was removed using a spiral fluted
tungsten carbide bur in a slow speed handpiece
as recommended by Zachrisson and Artun
(1979), and then polished with a slurry of
pumice and water in a rubber cup prior to
re-etching. The original brackets were rebonded.
In each case, a green stone in a slow speed
handpiece was used to remove the old resin
until the metal bracket base was visible. In
addition, with the Edgeway bracket the slots in
the base were also further redefined using a
dental chisel. Of all the resin/bracket combina-
tions, those involving Panavia EX required most
Figure 1 Diagram of testing jig.
29
repreparation. In the case of the Bond Fast
resin and the Minimesh brackets, however, no
repreparation was required. Although this
method of recycling brackets prior to rebond
appears to be somewhat crude, Wright and
Powers (1985) found similar if somewhat more
variable bond strengths could be obtained when
compared with commercially heat- or chemic-
ally-reconditioned brackets. In all cases, recon-
ditioned brackets are stated to have lower bond
strengths in use when compared with new
brackets, whatever recycling method is used. In
any case, in this experiment, if Panavia EX was
able to adhere to metal, then any damage caused
to the bracket bases using this method of recyc-
ling, might not be expected to affect the bond
strength as much as a resin relying solely on
mechanical interlock.
A total of 160 specimens were tested in a 24
factorial experimental design with 10 repeats at
each level tested. This design was chosen
because it is optimal for the evaluation of
interactions between factors. The predetermined
significance level was P<0.05. The power of
the analysis was calculated, retrospectively.
Bonferroni's test was used to evaluate multiple
comparison of means between the main effects.
Data were analysed using SAS/PC Version 6.04
(SAS Software Ltd., Medmenham, Marlow,
SL7 2ED, England). Data were tested for nor-
mality using the Shapiro-Wilk test. The system
Bond Fast/Edgeway/Bond/Wet (Table 1) was
found to be non-normal. Examination of the
data indicated an anonymously high bond
strength of 180.5N, which Grubbs' test (Grubbs,
1969) suggested was an outlier. This was con-
firmed by examination of the influence statistics,
the deletion or studentized residual and dffitts,
as suggested by Belsey et al. (1980) and
Atkinson (1985). The data were analysed both
with and without this value. The trends for both
analyses were the same, but exclusion of this
value resulted in the scatter for this data set
becoming comparable with the other 15 sets.
The presented results exclude this value.
Results
Univariate statistics and 95% confidence inter-
vals are given in Table 1, the standard analysis
of variance in Table 2, and the locus of bond
failure data in Table 3.
30 A. J. IRELAND AND M. SHERRIFF

Table 1 Univariate statistics.

95% Confidence
Code Resin Bracket Stage Wet/dry Mean force/N SD interval/N

1. Bond Fast Edgeway Bond Dry 52.7 12.94 43.4-61.9

2. Panavia EX Edgeway Bond Dry 92.2 22.81 75.8-108.5
3. Bond Fast Minimesh Bond Dry 93.8 9.4 87.0-100.5
4. Panavia EX Minimesh Bond Dry 123.9 7.82 118.3-129.4
5. Bond Fast Edgeway Rebond Dry 71.5 14.62 61.0-82.0
6. Panavia EX Edgeway Rebond Dry 97.3 29.57 76.2-118.5
7. Bond Fast Minimesh Rebond Dry 81.2 6.76 76.3-86.0
8. Panavia EX Minimesh Rebond Dry 122.6 12.12 114.0-131.3
9. Bond Fast Edgeway Bond Wet 70.4 24.45 50.5-90.4
10. Panavia EX Edgeway Bond Wet 88.4 22.72 71.1-104.6
11. Bond Fast Minimesh Bond Wet 86.8 20.89 71.9-101.7
12. Panavia EX Minimesh Bond Wet 88.2 15.10 77.4-99.0
13. Bond Fast Edgeway Rebond Wet 48.1 8.02 42.3-53.8
14. Panavia EX Edgeway Rebond Wet 59.2 13.00 49.9-68.5
15. Bond Fast Minimesh Rebond Wet 67.8 10.27 60.5-75.1
16. Panavia EX Minimesh Rebond Wet 85.2 14.34 74.9-95.5

Sample size 10, except for the system Bond Fast/Edgeway/Bond/Wet which was 9.

Table 3 Locus of bond failure.

Table 2 Analysis of variance.
Adhesive Cohesive Mixed mode
Sum of Mean
Source DF squares square /"value Pr>F Code ER BR R ER/BR/R

Model 15 67 856.22 452:(.7 13.0 0.001 1 0 8 0 2

Error 144 50 269.28 2 1 1 2 6
Corrected 3 0 10 0 0
total 159 118 125.49 4 0 0 0 10
5 0 10 0 0
Mean 6 0 0 0 10
Source DF square F value Pr>F 7 0 10 0 0
8 1 0 0 9
A 1 18 868.2 54.1 0.001 9 9 0 0 1
B 1 15 731.1 45.1 0.001 10 1 0 5 4
A*B 30.2 0.1 0.769 11 0 10 0 0
C 3464.3 9.9 0.002 12 0 6 0 4
A*C 199.1 0.6 0.451 13 0 9 0 1
B*C 4.7 0.1 0.907 14 0 2 0 8
A*B*C 848.7 2.4 0.121 15 10 0 0 0
D 10 554.4 30.2 0.001 16 0 2 0 8
A*D 6256.3 17.9 0.001
B»D 8012.9 5.8 0.018 Key—ER failure at enamel/resin interface; BR failure at
A*B*D 18.6 0.1 0.820 bracket/resin interface; R failure within resin.
C*D 1 5599.1 16.0 0.001
A«C*D 1 320.7 0.9 0.340
B*C»D 1 3836.7 11.0 0.001 Discussion
A*B*C*D 111.4 0.3 0.573 f_ T

Key significant main term effects, there are also

significant second and third order interactions
Code Source of variation which complicate the interpretation of the ana-
lysis. Multiple comparison of means showed
A Resin that Panavia EX had a significantly higher bond
B Bracket
C Stage strength than Bond Fast, Minimesh brackets
D Environment had a significantly higher bond strength than
Edgeway, bonding was significantly higher than
ADHESIVE BRACKET BONDING
rebonding and that dry bonds were significantly
higher than wet. Since multiple comparison
of means only considers main effects, the
presence of interactions precludes any useful
interpretation.
The presentation and interpretation of inter-
actions is an ongoing problem in statistics. A
graphical representation of the significant resin-
bracket-stage and resin-bracket-environment
interactions is given in Figs 2 and 3. Although
this demonstrates the interaction it does not
quantify it. The standard method of analysing
a factorial experiment involves the use of con-
trasts (John and Quenouille, 1977), but it is
interesting to note that this is rarely used in the
analysis of dental data.
A factorial experiment is concerned with the
simultaneous evaluation of a series of factors,
in this case the four factors are resin, bracket,
stage, and environment. The manner in which
each factor can occur is known as a level and
a treatment is a particular combination of levels
of factors. It is important to appreciate that the
term level is used only in a statistical sense and
has no other implication. When there are only
two levels it is common to call them low and
high. In the present experiment, factor A is the
resin, with Bond Fast being low level and
Panavia EX the high. These assignments are
purely arbitrary. The effect of a factor is the
difference in response when the factor changes
from low to high with all other factors being
constant. Contrasts estimate the effect of chan-
150 —
125 —
100 — O- -D
c of the Edgeway. However, bracket base size has
(S 75 —
O tion of bond strength by a number of workers
50 —
Bond Rebond
Stage
Figure 2 Representation of the resin-bracket-stage inter-
action. D, Panavia EX; O, Bond Fast; —, Minimesh;
, Edgeway.
31
150 —
125 —
£ 100 —
to
•o
c
"2 75 — O -O
50 — Dry Wet
Environment
Figure 3 Representation of the resin-bracket-environment
interaction. O, Panavia EX; O, Bond Fast; —, Minimesh;
, Edgeway.
ging the level of some factors whilst keeping
others constant, typically they provide an
answer to questions of the type 'What is the
effect of changing both resin and bracket, keep-
ing all other factors constant?' Changing from
Bond Fast to Panavia EX resulted in an increase
of 20.9 N for the Edgeway and 22.6 N for the
Minimesh bracket. This is not a statistically
significant interaction, but the magnitude of the
effect may be of considerable importance. For
instance this increased bond strength may be
due to adhesion via hydrogen bonding to the
oxide layer of the metal bracket base. This lack
of a significant resin/bracket interaction contra-
dicts the work of Ferguson et al. (1984) and
also that of Kinami et al. (1990), both of whom
were investigating conventional orthodontic
bonding agents.
One of the main term effects, namely
brackets, shows Minimesh brackets to give
greater bond strengths than the Edgeway
bracket. The nominal area of the base of the
Minimesh bracket is certainly greater than that
been found not to be important in determina-
(Reynolds and Von Fraunhofer 1976;
Dickinson and Powers, 1980; Lopez, 1980).
This is hardly surprising in view of the complex-
ity of the base retention mechanisms where the
effect of surface topography on stress patterns
is likely to be much more important. In the case
of an adhesive resin, such as Panavia EX, true
32
base surface area will be of greater importance
than nominal area. Since the Minimesh base
has a fine braised mesh and the Edgeway has a
system of much larger cast slots, it is perhaps
safe to assume the former base will have the
greater true surface area. This would be very
difficult to determine experimentally since the
Edgeway base has a microscopically rough sur-
face. Regan and Van Noort (1989) also com-
pared foil mesh with the Edgeway bracket and
found the latter bracket to give greater bond
strengths. Although this also confirms the lack
of a relationship between nominal area and
bond strength, it is in contradiction to the
results observed in this experiment. They also
suggested that the Edgeway bracket had greatly
improved retention due to its roughened sur-
face, compared with the smoother foil mesh
base. Since the Minimesh base gave consistently
higher bond strengths it can be postulated that
the surface roughness could be creating stress
fields more likely to cause bond failure at lower
force levels, rather than improving micromech-
anical retention. The effect of base design on
bond strength is obviously very complex with
most current research still unable to explain
fully the relationship between the two.
Environment had a significant effect, with expo-
sure to moisture reducing bond strength, as
would be expected from environmental effects
on structural adhesives (Kinloch, 1987). This
effect was found to be greater for Panavia EX.
In fact exposure to moisture reduces the initially
higher bond strength of Panavia EX to a value
comparable with that of Bond Fast. Thus, the
improved retention at the etched enamel surface
and the adhesion at the bracket base are appar-
ently lost with environmental exposure. Of fur-
ther practical importance in orthodontics is the
need to rebond a bracket during treatment. As
discussed by Wright and Powers (1985), a
rebonded system has a lower bond strength
than the original bond, and this was confirmed
in this experiment, even though the recycling
method is only used clinically to rebond single
brackets rather than recycling complete sets of
brackets.
Consideration of Table 3 shows no obvious
pattern in the locus of failure. From an ortho-
dontic point of view the preferred site of failure
would be at the enamel/resin interface in order
to reduce clean up time at debond. Retief (1974)
A. J. IRELAND AND M. SHERRIFF
has, however, shown that when such failure
does occur at this interface there is also
unwanted cohesive enamel failure. It is some-
what surprising in view of the recently published
reports of enamel fracture with the use of silane-
coated ceramic brackets, that the adhesive resin
in this experiment did not lead to more
enamel/resin interfacial failure. This might not
only be related to the inherent characteristics
of the adhesive resin, but also to the fact that
metal rather than ceramic brackets were being
used. Certainly, the stress field beneath the
metal brackets is likely to be more complex
than beneath the smooth-based silane-coated
ceramic brackets described in Joseph and
Rossouw's (1990), and Winchester's (1991)
work. Stress raisers on the metal bracket bases
in this experiment might be expected to promote
more mixed mode or base/resin interfacial fail-
ure as, indeed, was seen in Table 3.
It has also been suggested that Panavia EX
can form stronger bonds to etched enamel than
conventional resins (Ferrari et al., 1987). Once
again examination of the locus of failure data
does not suggest this to be the case.
In its present form, Panavia EX has little in
the way of advantages over more conventional
'no-mix' orthodontic bonding agents. Not only
is the bonding method time consuming, but the
apparent adhesive qualities were lost on expo-
sure to water. Mechanical interlock with both
the enamel and metal bracket base was the main
means of attachment after environmental expo-
sure. For this reason a simple smooth bracket
base was not tested.
Conclusions
Under the conditions of this experiment the
following conclusions were reached:
1. The adhesive resin, Panavia EX, gives greater
bond strengths with both Minimesh and
Edgeway brackets than the 'no-mix' resin
Bond Fast, under dry conditions.
2. No specific resin/bracket interaction was
found.
3. Minimesh brackets gave higher bond
strengths than the Edgeway bracket. This
may be related to nominal base area, but
is more likely to be related to true surface
area, certainly in the case of the adhesive
ADHESIVE BRACKET BONDING
resin. Other factors such as surface topo-
graphy and hence stress fields will also be
important.
4. Rebonded brackets demonstrate lower bond
strengths than new brackets. However, the
fall in bond strength was significantly lower
when rebonding was performed with the
adhesive resin.
5. Environment had a significant effect on bond
strength in all cases, but more so with the
adhesive resin. No significant difference was
seen between the adhesive and the 'no-mix'
resin following exposure to a moist environ-
ment. This suggests that the adhesive proper-
ties were lost and bond integrity was
maintained by mechanical interlock as with
the 'no-mix' resin.
6. The locus of failure data did not show any
consistency in the site of failure. It is not
possible to confirm, therefore, that the adhes-
ive resin forms a stronger bond to enamel
than the 'no-mix' bonding agent.
7. Although truly adhesive resins may be
advantageous in orthodontics, this experi-
ment shows Panavia EX to have few advant-
ages over a more conventional resin which
relies on mechanical interlock. The more
exacting and time consuming bonding tech-
nique makes Panavia EX unsuitable for
orthodontic use at present. Further develop-
ment is required to improve its resistance to
environmental exposure. If this can be
achieved, then at the same time, the resin
itself could be made inherently weaker such
that any increase in bond strength at either
the bracket/resin or enamel/resin interfaces
could be negated by a weaker cohesive
strength of the resin. In this way, at debond,
cohesive resin rather than cohesive enamel
failure would occur. Additionally, the resin
remaining on the enamel surface at debond
should then be easier to remove.
Address for correspondence
Mr A. J. Ireland
Orthodontic Department
Royal United Hospital
Combe Park
Bath BA1 3NG
Avon
33
Acknowledgements
The authors would like to thank Professor
N. E. Waters for the use of the Instron, and
Mr F. West and Mr A. Black of the Medical
School Workshop U.M.D.S. of Guys and
St Thomas's Hospitals.
References
Atkinson A C 1985 Plots, Transformations and regression.
An introduction to graphical methods of diagnostic
regression analysis. Oxford Science Publications, Oxford
Belsey D A, Kuh E, Welsch R E 1980 Regression diagnost-
ics. Wiley, New York
Buonocore M G 1955 A simple method of increasing the
adhesion of acrylic filling materials to enamel surfaces.
Journal of Dental Research 34: 849-853
Cooke P A 1990 Direct bonding with glass ionomer cement.
Journal of Clinical Orthodontics 8: 509-511
Dickinson P T, Powers J M 1980 Evaluation of fourteen
direct bonding orthodontic bases. American Journal of
Orthodontics 78: 630-639
Ferguson J W, Read M J F, Watts D C 1984 Bond strengths
of an integral bracket base combination: an in vitro study.
European Journal of Orthodontics 6: 267-276
Ferrari M, Cagidiaco M C, Breshi R 1987 Evaluation of
resin-bonded retainers with the scanning electron micro-
scope. Journal of Prosthetic Dentistry 59: 160-165
Fukushima A M, Okamoto A, Katou Y, Kobyashi T,
Satou Y, Kota K, Iwaku M 1985 Physical properties of
a new resinous adhesive. Part 2. The effect of exposure
to air. Abstracts of the Japanese Journal of Conservative
Dentistry 28: 93
Gettleman L, Vrijhoef D C, Uchiyama 1986 Adhesive
prosthodontics. Proceedings of the International
Symposium on Adhesive Prosthodontics, Nijmegen,
Academy of Dental Materials
Grubbs F E 1969 Procedures for detecting outlying observa-
tions in samples. Technometrics 11: 1-21
Ireland A J, Sherriff M 1988 The use of an adhesive
composite for the bonding of orthodontic attachments.
Journal of Dental Research 67: 661
John J A, Quenouille M H 1977 Experiments: design and
analysis. Charles Griffin and Company Ltd, London
Joseph V P, Rossouw E 1990 The shear bond strength of
stainless steel and ceramic brackets used with chemically
and light-activated composite resins. American Journal
of Orthodontics and Dentofacial Orthopedics 97:121-125
Kinami H, Sugimara M, Sakuda M, Okozaki M, Kimura H
1990 A new type of metal bracket for suppression of
resin remaining in debonding. Dental Materials Journal
9: 25-35
Kinloch A J 1987 Adhesion and adhesives. Science and
technology, Chapman and Hall, London
Lopez J I 1980 Retentive shear strengths of various bonding
attachment bases. American Journal of Orthodontics
77: 669-678
34
Regan D, van Noort R 1989 Bond strengths of two integral
bracket-base combinations: an in vitro comparison with
foil-mesh. European Journal of Orthodontics 11: 144-153
Retief D H 1974 Failure at the dental adhesive-etched
enamel interface. Journal of Oral Rehabilitation 1:
265-284
Reynolds I R, Von Fraunhofer J A 1976 Direct bonding of
orthodontic attachments to teeth: The relation of adhesive
bond strength to gauze mesh size. British Journal of
Orthodontics 3: 91-95
Tabata Y, Amano O 1985 Study on anterior adhesive
bridges and splints using newly developed adhesive resins.
Journal of Gnathology 4: 73-85
A. J. IRELAND AND M. SHERRIFF
Winchester L J 1991 Bond strengths offivedifferent ceramic
brackets: an in vitro study. European Journal of
Orthodontics 13: 293-305
Wright W L, Powers J M 1985 In vitro studies of physical
factors affecting adhesion of fissure sealants to enamel.
Proceedings of the International Symposium on the acid
etch technique, 50-62
Zachrisson B U, Artun J 1979 Enamel surface appearance
after various debonding techniques. American Journal of
Orthodontics 75: 121-137