European Journal of Orthodontics 15 (1993) 125-135 I 1993 European Orthodontic Society

The tensile bond strength of new and rebonded stainless steel

orthodontic brackets
D. Regan, B. LeMasney, and R. van Noort
School of Clinical Dentistry, University of Sheffield, England

SUMMARY The study investigated the effect on the tensile/peel bond strength of the variables
associated with the bracket base, the enamel surface, and the type of adhesive when both new
and used brackets were rebonded to a previously bonded enamel surface.
The tensile/peel bond strength was firstly evaluated for three different types of stainless steel
orthodontic bracket/base combinations. The cast integral base gave a significantly lower bond
strength than the foil-mesh and photo-etched bases. Following debonding, a group of new
brackets were bonded to the teeth using a chemically-activated or a light-cured adhesive. The old
adhesive had been removed from the enamel by either a hand sealer or a tungsten-carbide bur.
The rebonded new brackets demonstrated a small, but statistically significant fall in bond strength.
No differences were found between the enamel preparations or the adhesives.
A further group of previously debonded brackets were rebonded to the same teeth. The bracket
bases were prepared by either smoothing with a green stone or heating in a bunsen flame followed

Downloaded from by guest on July 2, 2015

by sandblasting and electropolishing. Highly significant falls in bond strength were obtained with
all the bases. No significant differences were found between the two methods of bracket prep-
aration.

Introduction brackets are purchased in 'one-patient' kits

The prevalence of clinical bond failures should
have been reduced as bracket base design has
progressed. Recent studies have indicated over-
all clinical failure rates of 6 per cent for a
chemically-cured adhesive (O'Brien et ai, 1989)
and 6.5 per cent for a light-cured adhesive
(Read and O'Brien, 1990). However, wide
variations may be expected depending on the
bracket type, the adhesive used, the teeth
bonded, and a range of operator and patient
variables.
Despite these advances, the rebonding of
brackets may still pose a problem due to the
adoption of pre-adjusted appliances which
require inaccurately located brackets to be
repositioned during treatment in order to take
full advantage of the archwire slot values and
sliding mechanics (McLaughlin and Bennett,
1991). When rebonding a bracket during treat-
ment the operator is faced with the choice of
rebonding the same bracket or using a new
one.
Using a new bracket each time increases the
cost of providing treatment, particularly if
since using a new bracket means breaking into
a new kit. Even if a new bracket is used, the
bond strength may be affected by residual
adhesive on the enamel surface. Large aggrega-
tions of adhesive, such as may occur around
the periphery of the bracket base, will need to
be removed from the enamel in order to allow
correct seating of the replacement bracket. Such
adhesive removal may be performed by hand
sealers or mechanical instruments (Oliver,
1991). However, since routine etching has been
found to penetrate to a mean depth of 28 /*m
(Legler et ai, 1990), complete removal of all
adhesive is unlikely.
A further decision to be made is the selection
of an adhesive to be used to rebond the bracket.
Even if the bracket was originally bonded with
a chemically-cured adhesive, rebonding with a
light-cured adhesive would offer the advantage
of a command set with the ability to tie in the
archwire immediately.
When rebonding previously used brackets,
these will have varying amounts of adhesive on
the fitting surface. Since the bond at the
' bracket/adhesive interface is primarily a
126
mechanical one, and is proportional to the
undercuts and surface roughness (Matasa,
1989), then the amount of remaining adhesive
is likely to affect the bond strength. There are
several possible preparation options regarding
the bracket base. The old bracket can be
rebonded without any preparation, or an
attempt made to remove adhesive remnants
either mechanically, or with a simple chairside
reconditioning process. Although clinical and
laboratory studies have been reported on the
bond strength of commercially recycled ortho-
dontic brackets (Buchwald, 1989; Regan el ai,
1990), there appears to be little guidance
regarding the outcome of rebonding brackets
which have become detached during the course
of orthodontic treatment.
The purpose of this in vitro study was to
investigate the effect on the tensile bond strength
of the variables associated with the bracket
base, the enamel surface, and the type of
adhesive when both new and used brackets are
rebonded to a previously bonded enamel
surface.
Materials and methods
Brackets and adhesives
Three different stainless steel bracket/base com-
binations were chosen to represent types com-
monly used in clinical practice. The bases
selected were a conventional foil-mesh base (A-
Company, San Diego, California), a cast integ-
ral base (Edgeway, Ortho Organizers Inc., San
Marcos, California), and a photo-etched base
(Microlok, GAC International Inc., Comnak,
New York). Pre-adjusted Edgewise premolar
brackets with a slot size of 0.022" by 0.028"
were used. Two different adhesives were
employed in the tests. These were a two-paste,
highly-filled orthodontic bonding system (Max-
imum, Ortho Organizers Inc., San Marcos,
California) and a light-cured, highly-filled
orthodontic adhesive (Transbond, Unitek Cor-
poration/3M, Monrovia, California).
Bonding
The bases were bonded to extracted premolar
teeth which had been stored in 70 per cent
ethyl alcohol. All brackets and teeth were
allocated a number so that a bracket could be
identified and rebonded to the original tooth.
The teeth were carefully examined and any
D. REGAN ET AL.
carious, damaged, or malformed specimens
were rejected. Approximately two-thirds of the
root of the tooth was removed using a band
saw in order to allow the teeth to be placed
centrally in the mounts with the labial surface
uppermost. Yellow dental stone (Kaffir-D)
poured into prepared silicone moulds, was used
as the mounting medium. In order to ensure
that the brackets were aligned horizontally to
the testing assembly, they were bonded before
mounting. This allowed the placement of wires
under the tie wings which could then rest on
the top of the moulds whilst the teeth were
mounted. Before bonding, the enamel surface
was polished for 15 seconds using a rubber
cup, and a slurry of pumice and water. Etching
was performed using a 37 per cent orthophosph-
oric acid liquid for 45 seconds. The teeth were
then rinsed and dried with oil-free compressed
air. The brackets were examined under a
magnifying lens following bonding and any
excess adhesive removed using a probe. All
brackets were allowed to bench-cure for 5
Downloaded from by guest on July 2, 2015
minutes and then mounted in the moulds which
were then stored in water at 37°C for 24 hours
before bond testing.
Bond strength testing
Tensile/peel bond strength testing was carried
out using a Lloyd Mk 5 machine (Lloyd
Instruments pic, Fareham, Hampshire, Eng-
land). The tensile load was applied using a
0.020" x 0.025" stainless steel wire assembly
consisting of two 40 x 10 mm rectangles of wire
which were soldered at one end as described
previously (Regan and van Noort, 1989). A
cross-head speed of 2 mm per minute was used
and the force required to debond the brackets
was recorded in Newtons.
Initial bond strength
The first part of the investigation consisted of
bonding 180 new brackets to the teeth using
the two-paste adhesive (Group I). This group
acted as a control and provided the brackets
for rebonding to the same enamel surfaces.
Following debonding, the bond strength was
recorded, and all teeth and bracket bases were
examined under a magnifying lens in order to
assess the amount of adhesive remaining on
the tooth and the mode of bond failure. The
enamel surfaces were classified according to the
adhesive remnant index of Artun and Bergland
(1984). -
REBONDING STAINLESS STEEL BRACKETS
Score 0 = no adhesive left on the tooth.
Score 1 = less than half of the adhesive left on
the tooth.
Score 2 = more than half of the adhesive left
on the tooth.
Score 3= all adhesive left on the tooth, with
a distinct impression of the bracket
base.
Rebonding new brackets
After bond testing, 40 new Microlok brackets
were bonded to the teeth from which brackets
had been debonded (Group 2). For 20 of these
teeth, the adhesive was removed from the
previously bonded enamel surface with a tung-
sten-carbide bur in a slow speed handpiece.
The remainder were prepared by using a hand
sealer (Ash osteo No. 4, Dentsply Ltd., Glouces-
ter). Equal numbers of the new brackets were
bonded with either the same two-paste adhesive
or a light-cured adhesive. This gave four groups
of tooth preparation/adhesive combinations.
These brackets were again debonded on the
Lloyd testing machine and the tensile/peel bond
strength recorded. The main aim of this part
of the investigation was to compare two
common methods of tooth preparation used
clinically. If no significant difference was found
then the simplest method of tooth preparation
could be used for the remaining rebond tests,
thus reducing the number of specimens required.
Rebonding used brackets
The final part of the study was concerned with
the variables associated with the rebonded
bracket bases. Bracket failure could occur at
one of several interfaces, each of which would
leave a different amount of adhesive remaining
on the bracket base and tooth surface.
In order to ensure a similar distribution of
sites of bond failure in each group, both the
brackets and teeth were examined for adhesive
remnants. Brackets were selected for rebonding
in order to ensure that the distribution of the
adhesive remnant index values seen in the group
as a whole was reflected in each of the sub-
groups. Group 3 consisted of 40 each of the
A-Company and Edgeway brackets together
with 60 Microlok brackets.
In all cases, the enamel was prepared by
removal of adhesive with a hand sealer before
final polishing with a rubber cup and pumice.
127
Two main methods of bracket base preparation
were used before rebonding. Before rebonding,
20 of each type of base were prepared by
removing adhesive from the base using a green
stone in a slow speed handpiece. A further 20
of each type were prepared according to the
method described by Buchman (1980). First,
the bracket was held in a bunsen flame for 3
seconds in order to burn off the acrylic. This
produces a temperature of approximately
1200°C. The bracket was then quenched in
room temperature water and cleaned in a
laboratory sandblaster for 5 seconds in order
to remove inorganic filler particles. Finally, the
bracket was electropolished for 20 seconds by
connecting the alligator clip of a Rocky Moun-
tain 700 electropolisher (Rocky Mountain
Orthodontics, Denver, Colorado) to a wire
ligature holding the bracket. In order to
differentiate the effects of flaming and sandblast-
ing from the electropolishing process, a group
of 20 Microlok brackets were rebonded follow-
Downloaded from by guest on July 2, 2015
ing flaming only.
The groups of brackets were randomly
divided so that half of the brackets were bonded
with a two-paste adhesive and the rest with a
light-cured adhesive. Tensile/peel bond testing
was performed as for the previous groups.
Samples of teeth and bases were selected for
examination under the scanning electron micro-
scope. The area of each base was estimated
from electron micrographs by tracing the photo-
micrograph on a Bit-Pad Two digitizer (Summa-
graphics Corporation, Fairfield, Connecticut)
connected to an AT IBM compatible personal
computer. In addition to tracing the peripheries
of the bases, an estimate was made of the
available cohesive bonding area of each base
by tracing the undercut features.
The data were analysed statistically using an
analysis of variance (ANOVA) and any signi-
ficant differences revealed by this procedure
were further investigated using the Scheffe test
with a 95 per cent confidence limit (/ ) <0.05).
This phased experimental design explored the
effect on the tensile bond strength of the
important variables associated with the original
bracket bases, the preparation of the formerly
bonded enamel surface, and the rebonding of
both new and previously used brackets. At the
same time, it avoided needless duplication of
testing procedures.
128 D. REGAN ET AL.

Results Table 1 Means and standard deviations of rebonded

brackets (Newtons).
Initial bond strength (Group 1)
The initial mean tensile/peel bond strengths of Bracket Tooth Bracket Adhesive Mean SD
prep prep
the three brackets together with their standard
deviations are illustrated graphically in Fig. 1. A-Co sealer gs lc 57.71 25.20
The highest bond strength was obtained with 2p 77.12 27.11
the foil-mesh A-Company bracket and the f/sb lc 73.25 52.01
lowest with the cast integral Edgeway bracket. 2p 63.25 42.72
The results of the initial analysis of variance, Edgeway sealer lc 33.68 19.10
gs
using the bond strength as the dependant 2p 28.87 16.55
variable, showed a significant difference between f/sb lc 25.23 13.27
the groups (P< 0.0001). The Scheffe multiple 2p 33.41 9.97
range test demonstrated that there was no
Microlok tc bur new lc 66.70 40.81
significant difference between the bond strength 98.34 22.82
2p
of the A-Company and Microlok brackets, but sealer new lc 100.86 34.92
that the bond strength of the Edgeway bracket 2p 87.03 31.54
was significantly poorer than both of the other gs lc 26.67 13.53
two brackets (P<0.05). 2p 49.54 25.29
f lc 21.69 14.30
2p 32.98 25.80
Rebonding new brackets (Group 2) f/sb lc 39.38 9.82
The mean tensile/peel rebond strengths of all 2p 57.99 36.32

Downloaded from by guest on July 2, 2015

the rebonded brackets following the different Key: new = new, unused bracket; gs = green stone; f/sb =
tooth preparations, bracket preparations, and flame and then sandblasted; f= flamed only; lc = light-cured
adhesives are given in Table 1 together with adhesive; 2p = two-paste adhesive.
their standard deviations.
The new Microlok brackets rebonded to a between the two groups prepared by hand
previously bonded enamel surface showed a scaling or a tungsten-carbide bur. In view of
reduction in bond strength (Fig. 2). This this finding, the other types of brackets were
amounted to a mean 19.6 per cent reduction only rebonded to teeth prepared with a sealer
in tensile/peel bond strength and was statistic- to remove excess adhesive from the enamel
ally significant at the P<0.05 level. surface.
The teeth to be rebonded with these new
Microlok brackets had received two types of Rebonding used brackets (Group 3)
preparation. No significant difference was found The results in Table 1 also showed that the
differences between the light-cured and the two-
paste adhesives were not consistent, and that
"» 160-,
c each performed better than the other with
o
I 140 certain bracket and tooth preparation combina-
o tions. No statistical difference was found
S- 120^ 113 .6 109 .7 between the two adhesives in any case.
O)
c 100 Because the brackets and teeth were num-
£ bered, it was possible to compare the perform-
« 80
•o ance of the same bracket/tooth combinations
c 60. 1
o 60 when bonded and then rebonded. The fall in
n
«
bond strength of the different groups is shown
a 40 graphically in Fig. 3. Since there was no signi-
Q.
ficant difference between them, the results have
20
C been combined for the two adhesives. This
o 0 shows that all brackets gave a lower mean
A-Company Edgeway Microlok tensile/peel bond strength on rebonding. The
Figure 1 Mean tensile/peel bond strengths of the original percentage reduction was 41.4 per cent for the
new brackets (in Newtons). A-Company bases, 47.6 per cent for the Edge-
REBONDING STAINLESS STEEL BRACKETS 129

120
109.7

rength (1Newtoi
98.3 100.9
100
87 2-paste
80 light-cured
66.7
w 60
•a
ensi le/peel bor

40

20

0
Original Tungsten-carbide Sealer
Figure 2 Mean tensile/peel bond strength (in Newtons) of the new Microlok brackets rebonded to a debonded enamel
surface. The original (i.e the control) bond strength is shown for comparison.

Downloaded from by guest on July 2, 2015

Original
Rebond

200 150 100 50 0 50 100 150 200

Tensile/peel bond strength (Newtons)

Figure 3 Mean tensile/peel bond strengths (in Newtons) of brackets debonded and rebonded to the same teeth. Key: gs,
prepared with green stone; f/sb, flamed and then sandblasted; f, flamed only.

way bases and 64.8 per cent for the Microlok
bases. These differences between original and
rebond tensile/peel bond strength were tested
statistically using a paired /-test and found to
be significant at the P< 0.001 level. It is
interesting to note that despite the reduction
in their bond strength, the used A-Company
bases were still stronger than the original new
Edgeway bases. The differences between the
types of bracket preparation were not statistic-
ally significant, although it can be seen that
just flaming the Microlok brackets results in a
very low rebond strength. The tensile/peel bond
strength of new Microlok brackets bonded to
previously bonded enamel was significantly
higher than any of the used brackets (P<0.05).
Adhesive remnant index (ARI) distribution
The main purpose of recording the adhesive
remnant index following the initial debonding
procedure was to allow the selection of compar-
able sub-groups for the subsequent tests. The
percentage frequency distribution of the ARI
is given in Table 2. This shows a broadly similar
distribution amongst the three types of bracket
base with most of the bases leaving over 50
130
Table 2 Percentage frequency distribution of the
adhesive remnant index (ARI) for each type of new
bracket following initial debonding.
Bracket ARI = 0 ARI=l ARI = 2 ARI = 3
A-Company 2.4 19.0 27.4 51.2
Edgeway 0 18.8 34.4 46.8
Microlok 1.8 15.5 48.2 34.5
per cent of the adhesive on the enamel surface
following removal in the tensile/peel mode.
Since all the teeth and brackets could be
identified at each stage of the experiment, it
was possible to investigate the effect of the
ARI on both the initial and rebond strength
using the analysis of variance procedure. No
significant differences were found between the
ARI value and either the initial bond strength
or the rebond strength for each of the bracket
bases.
Discussion
The original bond strength results demonstrate
a significantly lower tensile/peel bond strength
for the Edgeway cast base when compared with
both the A-Company mesh base and the
Microlok photo-etched base. This contradicts
two previous investigations involving the Edge-
way base (Regan and van Noort, 1989; Regan
et al., 1990). The reasons for this could be
related to the bases, the substrates or the
operators. In the earlier studies, flat upper
central incisor brackets were bonded to a flat
composite substrate. The present research used
curved premolar brackets bonded to extracted
teeth. It is possible that the adaptation to the
substrate and the undercut pattern is more
efficient on the flat brackets. The mean area of
the Edgeway base (9.83 mm2) was found to be
slightly smaller than the Microlok base
(10.36 mm2) and the A-Company base
(10.29 mm2). A more important factor is the
area of the available undercut, since this will
determine the surface area of the adhesive
which may fail in a cohesive mode. Again, the
differences between the bases were minimal.
The Microlok base has a slightly larger undercut
area (5.62 mm2) than the Edgeway (5.50 mm2)
and A-Company (5.12 mm2) bases. However,
when the debonded bases were examined, a
difference was found in the mode of failure
D. REGAN ET AL.
between the various types (Fig. 4). The bases
were classified into two groups; those with less
than 50 per cent of the undercuts filled with
adhesive and those with more than 50 per cent
of the undercuts filled with adhesive. It was
found that 49.2 per cent of the Edgeway bases
had less than 50 per cent of the undercut areas
filled with fractured adhesive whereas no
A-Company and only 4.6 per cent of Microlok
bases failed in such a manner. It appeared that
the mesiodistal horizontal channels provided
the main undercut areas in these premolar
Edgeway bases, thus reducing the effective
undercut to 2.98 mm2 (i.e. the area of these
channels as measured on the electron micro-
graphs). A particular problem noticed with the
Microlok bases was the presence of voids due
to air entrapment. This would also tend to
reduce the efficiency of the undercuts.
The difference in these results compared with
earlier data illustrates the danger of basing
recommendations on the outcome of testing
one type of bracket only. It is difficult to
Downloaded from by guest on July 2, 2015
extrapolate from the laboratory to the clinical
situation on the basis of these results since the
brackets were removed in the tensile/peel mode
rather than a shear/peel mode as would occur
with occlusal forces. Both the A-Company and
Microlok brackets should perform adequately,
but there may be some reservations about the
Edgeway base.
Following removal of a bracket, several
methods have been described for eliminating
the residual adhesive from the enamel surface.
Hand sealers are useful for large areas of
adhesive but may gouge the tooth surface
(Burapavong et al., 1978). Rotary instruments
such as green stones and diamond burs may
also produce serious enamel destruction (Retief
and Denys, 1979). Ultrasonic sealers are efficient
at composite removal, but the amount of severe
tooth surface damage is unacceptable (Walmsley
et al., 1989). Zachrisson and Artun (1979)
found that no method resulted in complete
adhesive removal but that a tungsten-carbide
bur used at a low speed produced the best
results. Other methods suggested have been a
mechanical band driver with a chisel (Dragiff,
1979; Gait, 1984) or a reciprocating handpiece
designed for finishing composite restorations
(Oliver, 1991).
Whatever method is employed, it would
appear that complete removal of all adhesive
is unlikely and a significant number of ortho-
REBONDING STAINLESS STEEL BRACKETS
dontic patients require the removal of stain
which has formed on resin still present even
years after orthodontic treatment (Brown,
1988).
131
This retention of adhesive on the enamel
surface was confirmed by this study. Figure 5
reveals an enamel surface following bracket
debonding. Etching had produced the typical
Type I etching pattern where the enamel rod
centres are preferentially etched (Ten Cate,
1989). The centres of the rods are now filled
with resin tags. Figure 6 shows an area next to
one of these islands following the application
of an etching liquid. It appears that the etching
procedure has no effect on the areas with
adhesive still present. This may explain the
reduction in bond strength which occurs even
when new brackets are used since there will be
a reduction in the amount of mechanical
retention available on the enamel surface.
Although this fall in bond strength with new
brackets was statistically significant it is unlikely
to be of clinical importance since the rebond
strength still falls within the value regarded as
necessary to withstand orthodontic and occlusal
forces (Reynolds, 1975). In similar studies with
new brackets, Faust et al. (1978) found that
Downloaded from by guest on July 2, 2015
Figure 5 Enamel surface following debonding. The etched
enamel prism centres are now filled with adhesive. (Original
magnification x 640.)
Figure 4 Scanning electron micrographs of the bracket
bases following initial debonding. (a) A-Company base.
Most of the available undercuts are filled with adhesive and
the fracture interface is at the level of the wire mesh
intersections. (Original magnification x 20, Scale bars =
100 /im.) (b) Edgeway base. The bulk of the adhesive is
restricted to the deeper horizontal channels. (Original mag-
nification x40. Scale bars= 100 jim.) (c) Microlok base. In
the centre of the base, the undercuts have only been partially
filled with adhesive. (Original magnification x 20, Scale
bars = 100 /mi.)
132
Figure 6 Enamel surface following debonding and re-etching. Next to the retained adhesive (A) are resin tags (RT) from
the original bonding. The acid-etching procedure has no effect on composite.
the differences among adhesives or between
brackets were more dramatic than bond and
rebond strengths. Jassem et al. (1981) did not
obtain any significant difference when rebond-
ing new foil-mesh brackets with a chemically-
activated adhesive.
The amount of adhesive remaining on the
enamel surface following debonding was not
found to be significantly related to either the
initial bond strength or the rebond strength.
Although the quantity of adhesive was estim-
ated in this study by using a subjective index
(the ARI), the lack of relationship confirms the
findings of O'Brien et al. (1988) who performed
a more rigorous analysis of the residual debris
following debonding.
When a used bracket is to be rebonded, there
is the additional problem of adhesive on the
bracket base. This can vary from a reduction
in the available undercuts to complete oblitera-
tion if the fracture has occurred at the enamel/
adhesive interface.
The base may simply be prepared by
smoothing any projections with a green stone
without attempting to remove the adhesive
from the undercuts. In some cases this may
leave a composite surface devoid of undercuts
for rebonding to the enamel (Fig. 7). The
strength of old composite restorations repaired
with fresh material has been found to be greatly
D. REGAN ET AL.
Downloaded from by guest on July 2, 2015
reduced when compared to a single polymerized
unit (Reisbeck and Brodsky, 1971). The factors
which may affect the bond between an old and
new composite include the ability to wet the
surface of the material, the amount and chem-
ical activity of available resin for bonding and
the effect of surface treatment prior to repair
(Saunders, 1990). Polishing the surface reduces
the bond strength by reducing chemically
reactive groups (Davidson et al., 1981) and
producing surface debris (Miranda et al., 1984).
The age of the repaired composite also affects
the bond strength, with the greatest reduction
in reactivity occurring during the first 24 hours
(Soderholm and Roberts, 1991). It has also
been found that composite stored in water has
a lower bonding potential than when stored
dry (Causton, 1975). This would indicate that
a bracket which had been removed for reposi-
tioning and then rebonded immediately after
could be expected to have a better bond strength
than one which had been in the oral environ-
ment for some time.
Commercial bracket reconditioning processes
use either heat or chemical solvents. Neither of
these are practical to perform fully at the
chairside, although a limited heating procedure
may be used in an attempt to burn off the
adhesive. Buchman (1980) found evidence of
significant annealing and carbide separation
REBONDING STAINLESS STEEL BRACKETS
when using a bunsen flame to remove adhesive
from stainless steel bracket bases. In addition,
further problems such as clogging of the base
and obstruction of undercuts and slots may be
(a)
133
experienced (Matasa, 1989). A stainless steel
bracket subjected to high temperatures also
takes on an unacceptable brown appearance
and requires repolishing using an electropol-
ishing process. This may result in significant
loss of metal and alteration in the undercuts
(Fig. 8). The present study found no difference
between any of the bracket preparation pro-
cesses and the use of heat cannot be recom-
mended.
Zachrisson (1985) recommended a 3-minute
rebonding procedure involving removal of
adhesive from the enamel surface using a sealer
or tungsten-carbide bur. If a new bracket is
not to be used, the adhesive remaining on the
loose bracket is removed with a bur or merely
roughened. Following a 60-second etch, the
bracket is then bonded with a chemically-cured
heavily-filled adhesive. From the present study,
the suitability of this technique has been
confirmed for the foil-mesh brackets. However,
with both the photo-etched and cast bases, the
Downloaded from by guest on July 2, 2015
resulting bond strength may well be unac-
ceptable. In these cases it would be advisable
to use a new bracket. Since no difference was
found between the types of adhesive used in
this study, there would appear to be some
advantage in employing a light-cured adhesive
for rebonding since it offers a generous working
time and a command set which allows brackets
to be positioned and excess material to be
easily removed (Read, 1984).
Extrapolation of laboratory data to the
clinical situation should always be done with
care. However, the findings of this investigation
do seem to be borne out by a clinical investi-
gation of Kinch et al. (1988). First time bonds
were found to have a failure rate of 12.6 per
cent, but subsequent bonds failed at a greatly
increased rate of 23.1 per cent. In addition,
they tended to fail fairly rapidly within the first
14 weeks following rebonding.
Figure 7 Debonded brackets prepared with' a green stone,
(a) A-Company base. This has left a uniformly flat surface
of adhesive. (Original magnification x20, Scale bars =
100 ftm.) (b) Edgeway base. The adhesive has been ground
flat producing a mechanically non-retentive surface. The
studs have also been flattened leaving sharp edges which
could cause stress concentration. (Original magnification
x20, Scale bars= 100/im.) (c) Microlok base. The sur-
face is deeply scratched and few undercuts remain patent.
(Original magnification x 20, Scale bars= 100 /«n.)
134
Conclusions
1. The integral cast base (Edgeway) gave a
significantly lower tensile/peel bond strength
D. REGAN ET AL.
than A-Company foil-mesh and Microlok
photo-etched stainless steel bases.
2. Rebonding new brackets to an enamel
surface which had been previously etched
and bonded resulted in a statistically signi-
ficant fall in tensile/peel bond strength. This
reduction was not considered sufficient to
be important clinically since the brackets
should still have an adequate bond strength
to withstand normal orthodontic and occlu-
sal forces.
3. There was no difference in tensile/peel bond
strength following removal of the adhesive
from the enamel surface with either a hand
sealer or a tungsten-carbide bur prior to
rebonding.
4. Rebonding previously used brackets resulted
in a significant fall in tensile/bond strength.
There was no difference between the results
obtained following bracket base preparation
with a green stone and a more complicated
Downloaded from by guest on July 2, 2015
process involving heating in a bunsen flame,
sandblasting and electropolishing. The fall
in bond strength was considered to be
important clinically and certain types of
bracket base should not be reused.
5. No difference was found in rebond strength
when using a chemically-cured adhesive and
a light-cured adhesive.
Address for correspondence
Mr D Regan
Department of Child Dental Health
School of Clinical Dentistry
University of Sheffield
Sheffield S10 2SZ
England
Figure 8 Debonded brackets prepared by flaming, sand-
blasting and electropolishing. (a) A-Company base. There
has been an appreciable loss of metal from the foil-mesh.
(Original magnification x20, Scale bars = 100/im.) (b)
Edgeway base. The base has lost the rough cast appearance
and the studs have been abraded so that they have become
wedge-shaped. (Original magnification x40, Scale bars =
100 urn.) (c) Microlok base. The undercut features have
become more rounded and some contain acrylic ash and
silica particles. (Original magnification x 20, Scale bars =
100 /mi.)
REBONDING STAINLESS STEEL BRACKETS
Acknowledgements
The authors would like to thank Mr
M. Debenham of Precision Orthodontics who
very kindly supplied the brackets and adhesives
used in this study.
References
Artun J, Bergland S 1984 Clinical trials with crystal growth
conditioning as an alternative to acid-etch enamel pre-
treatment. American Journal of Orthodontics 85:
333-340
Brown D 1988 Orthodontic materials update: 1. Ortho-
dontic bonding adhesives. British Journal of Orthodontics
15: 277-278
Buchman D J L 1980 Effects of recycling on metallic direct-
bond orthodontic brackets. American Journal of Ortho-
dontics 77: 654-668
Buchwald A 1989 A three-cycle in vivo evaluation of recon-
ditioned direct-bonding brackets. American Journal of
Orthodontics and Dentofacial Orthopedics 95: 352-354
Burapavong V, Marshall C W, Apfel D A, Perry H T 1978
Enamel surface characteristics on removal of bonded
orthodontic brackets. American Journal of Orthodontics
74: 176-187
Causton B E 1975 Repair of abraded composite fillings.
An in vitro study. British Dental Journal 139: 286-288
Davidson C L, Duysters P P E, De Lange C, Bausch J R
1981 Structural changes in composite surface material
after dry polishing. Journal of Oral Rehabilitation 8:
431-439
Dragiff D A 1979 A new debonding procedure. Journal of
Clinical Orthodontics 13: 107-111
Faust J B, Grego G N, Fan P L, Poers J M 1978 Penetration
coefficient, tensile strength and bond strength of thirteen
direct bonding cements. American Journal of Orthodont-
ics 73: 512-525
Gait C 1984 Use of an automatic handpiece for debonding
of composite resin. European Journal of Orthodontics 6:
254-256
Jassem H A, Retief D H, Jamison H C 1981 Tensile and
shear strengths of bonded and rebonded orthodontic
attachments. American Journal of Orthodontics 79:
661-668
Kinch A P, Taylor H, Warltier R, Oliver R G, Newcombe
R G 1988 A clinical trial comparing the failure rates of
directly bonded brackets using etch times of 15 or 60
seconds. American Journal of Orthodontics and Dento-
facial Orthopedics 94: 476-483
Legler L R, Reteif, D H, Bradley, E L 1990 Effects of
phosphoric acid concentration and etch duration on
enamel depth of etch: An in vitro study. American Journal
of Orthodontics and Dentofacial Orthopedics 98:
154-160
Matasa C G 1989 Pros and cons of the reuse of direct-
bonded appliances. American Journal of Orthodontics
and Dentofacial Orthopedics 96: 72-76
135
McLaughlin R P, Bennett J C 1991 Finishing and detailing
with a preadjusted appliance system. Journal of Clinical
Orthodontics 25: 251-264
Miranda F J, Duncanson M G, Dilts W E 1984 Interfacial
bonding strengths of paired composite systems. Journal
of Prosthetic Dentistry 51: 29-32
O'Brien K D, Read M J F, Sandison R J, Roberts C T
1989 A visible light-activated direct-bonding material: an
in vivo comparative study. American Journal of Ortho-
dontics and Dentofacial Orthopedics 95: 348-351
O'Brien K D, Watts D C, Read M J F 1988 Residual debris
and bond strength—Is there a relationship? American
Journal of Orthodontics and Dentofacial Orthopedics 94:
222-230
Oliver R G 1991 A new instrument for debonding clean-
up. Journal of Clinical Orthodontics 25: 407-410
Read M J F 1984 The bonding of orthodontic attachments
using a visible light cured adhesive. British Journal of
Orthodontics 11: 16-20
Read M J F, O'Brien K D 1990 A clinical trial of an
indirect bonding technique with a visible light-cured
adhesive. American Journal of Orthodontics and Dento-
facial Orthopedics 98: 259-262
Regan D, van Noort R 1989 Bond strengths of two integral
Downloaded from by guest on July 2, 2015
bracket-base combinations: an in vitro comparison with
foil-mesh. European Journal of Orthodontics 11: 144-153
Regan D, van Noort R, O'Keeffe C 1990 The effects of
recycling on the tensile bond strength of new and clinically
used stainless steel orthodontic brackets: an in vitro study.
British Journal of Orthodontics 17: 137-145
Reisbick M H, Brodsky J F 1971 Strength parameters of
composite resins. Journal of Prosthetic Dentistry 26:
178-185
Retief D H, Denys F R 1979 Finishing of enamel surfaces
after debonding orthodontic attachments. Angle Ortho-
dontist 49: 1-10
Reynolds I R 1975 A review of direct orthodontic bonding.
British Journal of Orthodontics 2: 171-178
Saunders W P 1990 Effect of fatigue upon the interfacial
bond strength of repaired composite resins. Journal of
Dentistry 18: 158-162
Soderholm K J, Roberts M J 1991 Variables influencing
the repair strength of dental composites. Scandinavian
Journal of Dental Research 99: 173-180
Ten Cate A R 1989 Oral histology: development structure
and function. C V Mosby Company, St Louis
Walmsley A D, Jones P A, Hullah W 1989 Ultrasonic
debonding of composite-retained restorations. British
Dental Journal 166: 290-294
Zachrisson B U 1985 Bonding in orthodontics. In: Graber
T M, Swain B F (eds) Orthodontics, current principles
and techniques. C V Mosby Company, St Louis
Zachrisson B U, Artun J 1979 Enamel surface appearance
after various debonding techniques American Journal of
Orthodontics 75: 121-137
Downloaded from by guest on July 2, 2015