Shear bond strength of ceramic orthodontic brackets to enamel Johanna C. Britton, DDS," Pamela Mclnnes, BDS, MSc(Dent), b Roger Weinberg, PhD, c William Ross Ledoux, DDS, d and Daniel Hugo Retief, MSc, BDS, PhD(Dent), DSc (Odont) ° New Orleans, La. and Birmhzgham, Ala.

The recent introduction of ceramic orthodontic brackets has generated interest among orthodontists. The aim of this investigation was to evaluate the in vitro shear bond strengths to enamel of four ceramic orthodontic brackets and one stainless steel bracket in trials with two separate acid-etching times for enamel. Eighty extracted human central incisors were prepared for bonding to Starfire, Allure, Transcend, Quasar, and stainless steel (in the control group) orthodontic bracket systems. Enamel etching times of 15 seconds and 60 seconds were used. There was a total of 10 groups. After acid etching, one coat of low-viscosity bonding agent was applied and the brackets were bonded to etched enamel with Concise orthodontic bonding resin. The bonded test specimens were stored in distilled water at 37 ° C for 14 days, after which they were thermocycled for 500 cycles (5 ° C to 60 ° C). The bonds were stressed to failure in an Instron machine at a crosshead speed of 0.02 inch per minute. The shear bond strengths were calculated and Weibull analysis was used to obtain a shape factor (the slope of the straight line and a measure of predictability) and the characteristic level (the 63.2% bond strength value of median rank on the straight line) for each group. Predictability and high bond strength, along with other factors, are important in the clinical selection of a bracket system. When either predictability or bond strength was considered independently, several bracket systems, coupled with a particular etch time, had either high predictability or high bond strength. The highest predictability and the highest bond strength were both found with the Allure bracket system. (AMJ ORTHOD DENTOFACORTHOP 1990;98:348-53.)

E s t h e t i c s has always been important in the field of orthodontics. Only in recent years, however, has the esthetics of the appliance itself become a subject of major interest. In 1955 Buonocore ~ proposed the acid-etch technique, but it was not until the next decade that this technique was applied to the direct bonding of orthodontic brackets. 2-7 The concept of bandless appliances had exciting possibilities: no band spaces to close after treatment, less gingival irritation, increased ease of plaque removal and a more esthetic appearance for the patient: By the late 1970s the bonding technique and bracket design produced bond strengths adequate to withstand intraoraI forces consistently: -~3 Direct bonding of brackets became an accepted clinical procedure in orthodontics. Patients preferred the appearance of these "bandless braces" but still desired improved esthetics. To meet

Supported in part by BRSG SO7 RR05704, DRR, National Institutes of ilealth. *Graduate student in Orthodontics, Louisiana State University School of Dentistry. ~'Associate Professor of Orthodontics, LSU School of Dentistry. 'Professor of Biornetry, LSU Medical Center. dClinical Associate Professor of Orthodontics, LSU School of Dentistry. =Professor of Biornatedals, University of Alabama School of Dentistry. 811116428


this demand, manufacturers marketed brackets made of various types of plastic. Although these brackets received initial acceptance by many clinicians, they were soon abandoned because of slot-dimension distortion and staining. Manufacturers returned to stainless steel but began designing brackets with increasingly smaller bracket dimensions, thereby reducing the amount of visible metal and improving the appearance of the appliance. The most recent esthetic brackets have been constructed of a ceramic material or of sapphire, a single-crystal ceramic. The manufacturers of both of these bracket types claim that their brackets are esthetic and will not stain or distort. They also report that their brackets have high retentive bond strengths. To achieve clinically useful bond strengths between brackets and enamel, some manufacturers have enhanced the chemical bonding of composite resin to ceramic materials with silane coupling agents. 14.15Several ceramic bracket systems incorporte silane coating with conventional mechanical-lock features. At least one bracket system minimizes mechanical retention and relies on silane coating for chemical bonding of the composite resin. Clinicians have reported that these new ceramic brackets are difficult to debond, although we could not

Volume 98 Number 4

Shear bond strength o f ceramic brackets to enamel


Fig. 2. Details of bracket bases. A, Stainless steel; B, Starfire;

C, Allure;/9, Transcend; E, Quasar.

Fig. 1. Bonding alignment block. A, Tooth specimen cup; B, orthodontic bracket; C, mounting rod.

find this claim documented in the literature. With conventional methods of debonding, instances of enamel fracture have been reported, and there is a concern that these ceramic brackets bond so strongly to enamel that the enamel fails before the brackets at debonding. The role of acid etching on enamel in the production of these bond strengths needs to be investigated. Applying 37% phosphoric acid to the enamel for 60 seconds is an accepted and effective method of etching enamel in preparation for bonding with a resinous material. 4"9.t3,t6 Factors such as acid type, acid concentration, and length of etching time may affect bond strength. 1"5"6"t°'16-~8 The effect of etching time remains controversial. Gorelick, ~° Barkmeier et al., '7 and Beech and Jalaly 19 demonstrated no decrease in bond strength as the result of shortened etching times. However, Mardaga and Shannon 2° and Wickwire and Rentz ~2showed significant bond-strength reductions with decreased etching times. Concern with the strength of ceramic brackets bonded to enamel may necessitate changing enameletching techniques in order to decrease the bond strengths. This concept led us to evaluate the in vitro retentive bond strength of four ceramic bracket types that had been bonded after a 15-second acid-etching time and a 60-second acid-etching time. MATERIALS AND METHODS Eighty freshly extracted noncarious, human maxillary permanent central incisors were used. The teeth were cleaned with a water slurry of flour of pumice and stored in distilled water. The roots were cut off, and retentive notches were

placed in the lingual surfaces of the crowns. Each crown was mounted in epoxy resin in a brass specimen cup with the facial surface projecting above the rim of the cup. The facial enamel was wet-ground with 600-grit silicon carbide (SIC) disks and a polishing block for a planar surface. The enamel surface was then thoroughly cleaned with a slurry of flour of pumice and water. The teeth were rinsed and dried with oil-free compressed air. The enamel surfaces were etched with 37% phosphoric acid gel for either 15 seconds or 60 seconds. The acid was rinsed off for 20 seconds and the etched enamel surfaces were redried with oil-free compressed air. All the etched surfaces were sealed with one coat of Concise Enamel Bond* a chemically cured, unfilled BIS-GMA resin. Each tooth cup was attached to a mounting rod and held in the bonding alignment block to ensure that the base of the orthodontic bracket was parallel to the treated enamel surface (Fig. 1). The orthodontic brackets were attached to the upper mounting rod. Five bracket types were bonded in this study: (1) Mini-Diamondt (stainless steel control, mesh base for mechanical retention); (2) Starfire~ (sapphire, grooved base for mechanical retention, and silane treated); (3) Allure§ (ceramic, indented base for mechanical retention, and silane treated); (4) Transcendll (ceramic, smooth base, and silane treated); and (5) Quasar¶ (ceramic, fibrous base for mechanical retention, and silane treated). The bracket bases are illustrated in Fig. 2. *3M/Dental Products Division, St. Paul, Minn. tOrmeo Company, Glendora, Calif. :~A-Company, San Diego, Calif. §GAC, Central Islip, N.Y. I[Unitek Corporation, Montovia, Calif. ¶Rocky Mountain Orthodontics, Denver, Colo.


Am. d. Orthod. Dentofac. Orthop. October 1990

Britton et al.

croscope at a magnification of × 8, and the failure mode was noted. RESULTS

Fig. 3. Bond-breakage apparatus. A, Holding jig; B, table of the Instron compression cell; C, shearing rod.

Each bracket type was bonded to 16 teeth, of which eight were acid etched for 15 seconds and eight for 60 seconds. Two-paste Concise orthodontic resin* was used to bond all the brackets. A fresh mix of composite bonding resin was placed on the base of each orthodontic bracket, and the mounting rod with the attached bracket was lowered to contact the prepared enamel surface. Excess composite was removed from the periphery of each orthodontic bracket, and the specimens were allowed to cure for 10 minutes in the bonding alignment block under a load of 500 gm. After removal from the bonding alignment block, the bonded units were stored in distilled water at 37 ° C for 14 days. All the bonded units were then thermocycled in distilled water for 500 complete cycles. The temperatures ranged from 5 ° C to 60 ° C, and the dwell time in each bath was 60 seconds. The specimens were then stored in distilled water at 37 ° C for 7 days before mechanical testing. The shear bond strength was tested with a Model 1011 Instront testing machine. Each bonded unit was placed in a jig that allowed a compressive force to be applied to the bond interface, thereby inducing failure of the bond (Fig. 3). A cross-head speed of 0.02 inch per minute was used. The load at failure was recorded, and the stress at failure was calculated. The surface area of the base of each bracket type was calculated with an Olympus Cue-2 image analyzer.~ After failure, the orthodontic bracket bases and enamel surfaces were observed with a dissecting mi*3M/Dental Products Division, St. Paul, Minn. tlnstron Corp., Canton, Mass. :l.Olympus Corp., Overland Park, Kan.

Plotting the data revealed that they were not normally distributed. Descriptions of traditional measures, such as mean and standard deviation, are therefore inappropriate. The abnormal distribution of our data prompted our use of the Weibull analysis. The probability distribution in this method of analysis generates two parameters: shape factor and characteristic level. The shape factor is the slope of the straight line, fitted through the data values. It has no units. (In normally distributed data, this factor would be described by the standard deviation.) The characteristic level is the data value that corresponds to the 63.2% bond strength value of median rank on the straight line. Its units are MN • m -2. (In normally distributed data, this would be described by the mean.) Table I lists, for each group of eight teeth, the range of shear bond strengths and the results of the Weibull analysis of shear bond strength. The predictability of a bonded, ceramic bracket system can be seen on examination of the shape factor. Higher shape factors indicate a more predictable system and, therefore, a more clinically reliable system. For example, Transcend, in combination with a 60-second etch, has a shape factor of 8.5, while Quasar, with a 60-second enamel etch, has a shape factor of 2.0. This difference implies more clinical predictability with Transcend than with Quasar, when the enamel is etched for 60 seconds. Ninety percent confidence limits about shape factor indicate the precision of our estimates. The above example illustrates that the upper limit (4.0) of the Quasar group is lower than the lower limit (6.1) of the Transcend group. These nonoverlapping confidence limits indicate that the difference observed in the shape factors is valid at the 90% level of confidence. The characteristic level is similar to the mean, derived from an analysis that assumes a normal distribution. Therefore, the higher the characteristic level of a group, the higher the shear bond strength of that group. For example, the stainless steel control system, in combination with 15-second enamel etch, has a characteristic level of 23.6 MN • m -s, while the Quasar system, with a 60-second enamel etch, has a characteristic level of 13.6 MN • m -2. This difference indicates that the control stainless steel system with a 15second enamel etch produced higher shear bond strengths. Confidence limits of 90% with respect to characteristic level indicate the precision of our estimates of

Volume 98 Number 4

Shear bond strength of ceramic brackets to enamel


Table h Shear bond strengths: Range and Weibull analysis

Bracket Stainless steel Startire Allure Transcend Quasar

Etch time (sec)

Range (MN'm -2)

Shape factor

15 60 15 60 15 60 15 60 15 60

16.0-31.2 5.0-20.8 7.4-18.7 2.8-26.0 15.0-24.9 14.4-22.1 14.6-24.3 9.1-15.9 12.0-27.2 3.7-23.9

4.6 3.4 4.4 2.1 6.4 7.4 5.3 8.5 4.5 2.0

that parameter. The above example shows that the upper limit (19.8) of the Quasar group is lower than the lower limit (20.0) of the stainless steel group. Thus these nonoverlapping confidence limits indicate that the difference observed in the characteristic levels is valid at the 90% level of confidence. A comparison of the shape factor of the two etch times within each bracket system revealed that all 90% confidence limits overlapped. These confidence limits, therefore, provided no evidence-that varying the etch time affected clinical predictability of the bracket systems studied. A comparison of the characteristic levels of the two etch times within each bracket system showed that the 15-second enamel etch produced higher shear bond strengths than the 60-second enamel etch. However, the only nonoverlapping confidence limits were for stainless steel and T~anscend bracket systems. These nonoverlapping confidence limits suggested that it was only for these bracket systems that enamel-etch time influenced the shear bond strength. Extending our comparisons from intrabracket to interbracket systems, we examined 90% confidence limits for both shape factor and characteristic level. From shape factor, we observed that Allure at 15 seconds, Allure at 60 seconds, and Transcend at 60 seconds produced more clinically predictable bond strengths than Starfire at 60 seconds and Quasar at 60 seconds. Examining the characteristic levels, we observed that the stainless steel brackets bonded for 15 seconds produced higher shear bond strengths as a group than Starfire at 15 seconds, Transcend at 60 seconds, and Quasar at 60 seconds. When we compared just the ceramic bracket systems, Quasar at 15 seconds produced a higher shear bond strength than either Starfire at 15 seconds or Transcend at 60 seconds.

90% confidence limits 3.3 2.5 3.2 1.5 4.6 5.3 3.8 6.1 3.3 1.4

9.3 6.9 8.9 4.2 12.8 14.9 10.7 17.1 9.1 4.0

Characteristic level (MN.m -2) 23.7 15.6 15.1 14.5 19.9 19.8 18.6 14.2 21.5 13.6

90% confidence limits 20.0 12.4 12.6 9.9 17.5 17.8 16.1 12.9 18.0 9.1

27.8 19.5 17.9 20.8 22.3 21.9 21.5 15.5 25.3 19.8

The failure sites observed were (1) adhesive at the bracket/resin interface, (2) adhesive at the enamel/resin interface, or (3) combination adhesive/cohesive failure with resin retained at some sites on the bracket and at other sites on the enamel. The adhesion of the Trariscend and Quasar groups usually failed at the bracket/resin interface for both acid-etch times. By contrast, the Starfire group had failures at all three sites, and the Allure group failed only at the combined adhesive/cohesive site for both acid-etch times. It was only with the stainless steel group that enameletch time appeared to influence the site of failure. The 15-second etch resulted in both adhesive failure at the bracket-resin interface and cohesive/adhesive failure; the 60-second etch resulted in adhesion failure almost exclusively at the enamel/resin interface. There was no clear correlation between predictability and bond strength of a bracket system and the site of bond failure. DISCUSSION

An orthodontist's deeison to select a particular bracket system will depend on both the shape factor and the characteristic level. Selection based on high shape factor alone--that is, clinical predictability-could involve shear bond strengths that are not adequate. A decision based on characteristic level alone-that is, high shear bond strength--could result in an unacceptable percentage of brackets failing. Ideally, we want a high shape factor and a high characteristic level. Assuming that both high predictability and high bond strength are desirable, clinicians may find use for some mathematical tool that summarizes both of these parameters. For example, the investigator could use the square root of the shape factor, multiplied by the characteristic level, as such a tool (Table II). Although this formula


Am. J. Orthod. Dentofac. Orthop. October 1990

Britton et al.

Table II. An arbitrary measure of

clinical effectiveness Bracket

Stainless steel Starfire Allure Transcend Quasar

I 'ch'ime I (seconds)

SF x CL*

15 60 15 60 15 60 15

50.8 28.9 31.6 20.9 50.1 53.9 43.0

60 I5 60

41.2 45.6 19.1

*Shape factor × characteristic level.

is an arbitrary measure of clinical effectiveness, we chose the square root of the shape factor because it is somewhat analogous to variance in measuring dispersion of the data. From Table II it can be seen that the Allure bracket system, regardless of enamel acid-etch time, may be the ceramic bracket system of choice for both predictability and bond strength. However, other factors are involved in the selection of a bracket system for clinical use. Factors such as smooth edges, translucency of the ceramic material, relative slot orientation, and a bracket base large enough to avoid plaque accumulation under the tie wings, all play a role in bracket selection. An interesting observation from this study was the clinical effectiveness scores achieved with the Transcend system. Although all the ceramic brackets have a silane coating for chemical bonding to composite resin, the Transcend bracket is the only one that is not designed with mechanical retention features. The bond strength results achieved with this bracket system were good, and it appears that mechanical retention features in ceramic bracket bases may not be necessary. Although acid etching of enamel for 60 seconds before resin bonding has been advocated, our results indicate that increased bond strengths can be achieved with a reduced acid-etch time. This outcome confirms the findings of Barkmeier et al. ~7 and Legler et al. ~8 With the exception of the Allure system, the clinical effectiveness scores indicate that a 15-second enamel etch increases both clinical predictability and bond strength when compared with a 60-second enamel etch. An unexpected observation was the high percentage of adhesion failure at the enamel/resin interface in the stainless steel group that had been etched for 60 seconds. It is possible that the longer etch time weakened

the etched enamel rods and resulted in this type of failure, but this trend was not observed in the ceramic bracket systems. The results of this study do not confirm anecdotal reports of enamel fracture at debonding of ceramic brackets. The shear bond strengths of ceramic brackets to enamel were similar to those of the control stainless steel brackets to enamel. As enamel fracture on debonding of stainless steel brackets is not frequently reported, it can be concluded that shear bond strength of the ceramic brackets to enamel is not, by itself, the cause of these reported enamel fractures. In this study, all bonded brackets were stressed to failure in a predominantly shear mode. Clinically, torsional forces are also introduced during debonding, and perhaps it is these forces that result in enamel damage. Another possibility is that the stainless steel brackets tend to flex when the debonding pliers are used, and this flexing may dissipate the debonding forces, thereby protecting the enamel. Our clinical experience has revealed a tendency for ceramic bracket tie wings to fracture when a torqued arch wire is engaged into the bracket slot. At clinical debonding, the ceramic brackets also have a tendency to fracture. This occurrence necessitates extensive finishing and polishing to remove the bonded bracket remnants. Although there have been significant advances in the development of high-strength dental ceramics, the tensile strength of the materials remains comparatively low. This low tensile strength, together with the phenomenon of crack propagation, has long been a problem in restorative dentistry; with the introduction of ceramic brackets, it now presents a problem in orthodontics. The dental materials industry continues to search for tougher ceramic materials. More immediately, the orthodontist needs information on appropriate means of dealing clinically with the ceramic bracket systems now available, including information on torquing force limits and on debonding techniques.


1. BuonocoreMG. A simple method of increasing the adhesion of acrylic filling materials to enamel surfaces. J Dent Res 1955;34:849-53. 2. NewmanGV. Adhesion and orthodonticplasticattachments. A.~I J ORanOD1969;56:573-88. 3. NewmanGV, Snyder WH. Wilson CW. Acrylic adhesives for bonding attachments to tooth surfaces. Angle Orthod 1968; 38:12-8. 4. RetiefDH. Dreyer CJ, GavronG. The direct bonding of orthodontic attachmentsto teeth by meansof an epoxyresin adhesive. AM J ORTHOD1970;58:21-40.

Volume 98 Number 4 5. Retief DH. A comparative study of three etching solutions: effects on contact angle, rate of etching and tensile bond strength. J Oral Rehabil 1974;1:381-90. 6. Mulholland RD, DeShazer DO. The effect of acidic pretreatment solutions on the direct bonding of oahodontic brackets. Angle Orthod 1968;38:236-43. 7. Mizrahi E, Smith DC. Direct cementation of orthodontic brackets to dental enamel. Br Dent J 1969;127:371-5. 8. Reynolds IR, Von Fraunhofer JA. Direct bonding of orthodontic attachments of teeth: the relation of adhesive bond strength to gauge mesh size. Br J Orthod 1976;3:91-5. 9. Thanos CE, Munholland T, Caputo AA. Adhesion of mesh-base direct bonding brackets. AM J OR'mOP 1979;75:421-30. 10. Gorelick L. Bonding metal brackets with a self-polymerizing sealant-composite: a 12-month assessment. AM J ORTHOD 1977;71:542-53. 11. Zachrisson BU, Brobakken BO. Clinical comparison of direct versus indirect bonding with different bracket types and adhesives. A~I J ORTHOD 1978;74:62-77. 12. Wickwire NA, Rentz D. Enamel pretreatment: A critical variable in direct bonding systems. AM J OarrtOD 1973;64:499-512. 13. Retief DH. The mechanical bond. Int Dent J 1978;28:18-27. 14. Newman SM, Dressier KB, Grenadier MR. Direct bonding of orthodontic brackets to esthetic restorative materials using a silane. AM J ORrHOD 1984;86:503-6. 15. Johnson RG. A new method for direct-bonding orthodontic at-

Shear bond strength o f ceramic brackets to enamel







taehments to porcelain teeth using a silane coupling agent: an in vitro evaluation. A,,,iJ ORTnOD 1980;78:233-4. Retief DH. The use of a 50 percent phosphoric acid as an etching agent in orlhodomics: a rational approach. AM J OR'moP 1975;68:165-78. Barkmeier WW, Gwinnett A.I, Shaffer SE. Effects of enamel etching time on bond strength and morphology. J Clin Orthod 1985;19:36-8. Legler LR, Retief DH, Bradley EL, Denys FR, Sadowsky PL. Effects of phosphoric acid concentration and etch duration on the shear bond strength of an orthodontic bonding resin to enamel: an in vitro study. AM J ORTHOD DENTOFACORTIIOP 1989;96:485-92. Beech DR, Jalaly T. Bonding of polymers to enamel: influence of deposits formed during etching, etching time, a period of water immersion. J Dent Res 1980;59:1156-62. Mardaga WJ, Shannon IL. Decreasing the depth of etch for direct bonding in orthodontics. J Clin Orthod 1982;16:130-2.

Reprint requests to: Dr. Pamela Mclnnes Department of Orthodontics LSU School of Dentistry 1100 Florida Ave. New Orleans, LA 70119

Shear bond strength of ceramic orthodontic brackets to enamel.

The recent introduction of ceramic orthodontic brackets has generated interest among orthodontists. The aim of this investigation was to evaluate the ...
755KB Sizes 0 Downloads 0 Views