The European Journal of Orthodontics Advance Access published online on October 25, 2007
The European Journal of Orthodontics, doi:10.1093/ejo/cjm056
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Evaluation of the debonding strength of orthodontic brackets using three different bonding systems
* Department of Oral and Maxillo-Facial Science, University of Naples Federico II
** Institute of Composite and Biomedical Materials, National Research Council, Naples, Italy
Address for correspondence Dr Rosa Valletta, Dipartimento di Scienze Odontostomatologiche e Maxillo-Facciali, Università degli studi di Napoli "Federico II", Via S. Pansini, 580131 Napoli, Italy, E-mail: valletta{at}unina.it
 |
Summary |
|---|
 Top Summary IntroductionMaterials and methodsResultsDiscussionConclusionsReferences |
|---|
The aim of this work was to investigate the stability of the bracket–adhesive–enamel interface, as a function of adhesive material and of debonding procedure, in order to assess which debonding technique is the least detrimental to the enamel. Ninety lower adult bovine incisors were selected and metallic orthodontic brackets were bonded using three adhesive systems: Concise, Transbond, and Fuji Ortho. Three different debonding procedures were used based on tensile, shear, and torsional stresses. One-way analysis of variance statistical analysis was employed to compare mechanical properties, while the adhesive remnant index was used to evaluate fracture properties.
Each adhesive material used showed a statistical difference in tensile failure. The difference between shear and torsion failure loads was statistically significant only for the Fuji GC sample (P < 0.01). The shear test was the most damaging to the enamel surface.
Transbond luting resulted in greater adhesion than the Concise or Fuji Ortho systems. Fuji Ortho was more prone to accidental debonding, while Transbond tended to cause enamel lesions, since high loads were required to debond the bracket. Of the three modes examined, torsional debonding stress resulted in the least enamel damage.
|
Introduction |
|---|
|
TopSummaryIntroductionMaterials and methodsResultsDiscussionConclusionsReferences |
|---|
The evolution of multibracket therapy has resulted in a series of advantages, both to the patient and orthodontist. However, accidental debonding of the brackets still remains one of the most frequently encountered problems. Research has mainly concentrated on the reliability of the adhesive systems, on the possibility of reducing enamel damage during bracket removal, and on simplifying adhesive techniques (Bishara et al., 1999
). As highlighted by Fox et al. (1994) and confirmed by Eliades and Brantley (2000), the protocols used by researchers differ widely. Consequently, comparison of the heterogeneous data is impossible, and clinical inference becomes unreliable. Eliades and Brantley (2000) emphasized the need for a research protocol which standardized a series of parameters and considered all of the variables linked to the operating methodology. In particular, those authors focussed on the influence of the type of substrate used, on the structures of the coupled surfaces, and on the debonding stress methods. In order to investigate adhesive materials for orthodontic bracket bonding, it is necessary to analyse their performance in relation to the stress involved in the bracket–adhesive–tooth system.
The aim of this research was to determine the mechanical behaviour of the bracket–adhesive–enamel system using three different bonding materials. Three types of mechanical tests were carried out, tensile, shear, and torsion. The debonded surface of each specimen was observed using optical microscopy, and an adhesive remnant index (ARI) analysis was carried out to quantify the amount of adhesive on the enamel surface (Årtun and Bergland, 1984; Cacciafesta et al., 2002; Summers et al., 2004).
|
Materials and methods |
|---|
|
TopSummaryIntroductionMaterials and methodsResultsDiscussionConclusionsReferences |
|---|
Central lower bovine incisors, which are commonly used for in vitro debonding testing (Oesterle et al., 1998
; Saleh and Taymour, 2003), were extracted from freshly sacrificed animals with an average age of 2 years (±2 months). Visual analysis, with x4 magnification, was carried out in order to select 90 sound teeth, undamaged by the extraction procedures (Figure 1a). The selected teeth were stored in 10 per cent thymol solution at 37°C until sample preparation and mechanical testing was completed. Bonding procedures were undertaken within 1 week of extraction of the teeth.
|
Further analysis by optical microscopy facilitated the identification and marking of an area with an optimally smooth and flat surface on the vestibular side of each tooth (Figure 1b). The samples were cleaned with pumice (Ruse et al., 1990
) and sectioned (Figure 1c) with an Isomet microtome (Buehler Ltd, Lake Bluff, Illinois, USA), fitted with a diamond saw (0.3 mm thick and 100 mm diameter) and cooled by water at a speed of 125 mm/second (150 rpm). The sections were then placed on an adhesive tape (Figure 1d) and embedded into steel cylinders (Figure 1e,f) filled with a self-curing, methyl methacrylate-based resin (Formatray, Kerr Corporation, Orange, California, USA). Again, visual analysis was used to select specimens without damage (i.e. enamel fractures) due to fixation procedures.
Preparation of the specimens
The 90 teeth were randomly divided into three equal groups. The metal brackets used were OmniArch Ricketts model (GAC International Inc., Bohemia, New York, USA) for upper central incisors (slot dimensions 0.018 x 0.030 inch), having a fairly flat base with a half-dome mechanical retention.
In group A, a glass-ionomer cement modified with light-activated resin (Fuji, GC America Inc., Alsip, Illinois, USA) was used; in group B, the brackets were bonded with a chemically activated paste–paste composite (Concise, 3M Unitek, St Paul, Minnesota USA); and in group C, a light-activated composite was used (Transbond, 3M Unitek). The technical specifications of all materials are reported in Table 1.
|
For group A specimens, the enamel surface was conditioned with 10 per cent acrylic acid solution for 15 seconds and then rinsed for 30 seconds. The surface was kept in a humid environment during the application of the cement, previously stirred for 10 seconds. A 300 g weight was then placed on the bracket base for 15 seconds and excess cement was carefully removed. The adhesive was cured for 40 seconds using a halogen light source with a power output of 800 mW/cm2 (Optilux, Demetron, Kerr Corporation). The output of the curing unit was periodically controlled by a radiometer (Demetron Corporation) with 900 mW/cm2 full scale.
For group B specimens, the enamel surface was conditioned with 37 per cent orthophosphoric acid for 45 seconds, then rinsed for 30 seconds and dried with an oil-free air jet. The adhesive resin and the composite paste were applied to the enamel surface and to the bracket base, respectively. As in group A, each bracket was preloaded with a 300-g weight.
For the Transbond samples (group C), the enamel was etched with 37 per cent orthophosphoric acid for 15 seconds, rinsed for 30 seconds, and carefully dried with an oil-free air jet. The adhesive resin was applied to the enamel surface, and the composite resin to the base of the bracket. Each bracket was then preloaded, as in group A. After removing the composite cement excess, curing was then effected by exposure to the halogen light source for 40 seconds.
Mechanical tests
Each group of specimens was divided into three subgroups of 10 samples for mechanical characterization of tensile, shear, and torsion values.
Tensile tests were carried out using an Instron 4204 screw driven dynamometer (Instron, High Wycombe, Bucks, UK). The crosshead speed was set at 0.5 mm/minute and a 1 kN load cell was used. The samples, gripped into the chuck of the testing dynamometer, were stressed using a suitable stainless steel tool (Figure 2a). Shear and torsion tests were carried out using a servohydraulic dynamometer MTS 858 Bionix (MTS, Minneapolis, Minnesota, USA), using a load cell with load and torque ranges of 2.5 kN and 25 Nm, respectively. The shear tests were undertaken in a displacement control with the activator speed set at 0.5 mm/minute. The load was applied to the longer side of the bracket base using a knife-blade (Figure 2b). The torsion tests were carried out controlling the angular position and with the angular speed set at 0.5 degrees/minute. Torque was applied through a flat (2.7 mm thick) screwdriver placed in the vertical groove between the wings of the bracket (Figure 2c). The data acquisition (load, torque, and axial and angular position) was undertaken at a speed of 10 points/second.
|
The statistical significance of the results was assessed using one-way analysis of variance (ANOVA; Cheremisinoff, 1987
) through the Origin® 6.0 package (Microcal Software, Northampton, Massachussets, USA). Confidence levels of 99 and 95 per cent were set for mechanical strength and ARI, respectively.
Analysis of fractured surfaces
Analysis of the fractured surfaces, externally illuminated by an optical fibre, was performed using a Zeiss optical microscope (Carl Zeiss IMT Corporation, Minneapolis, Minnesota, USA) at x20 magnification. For each sample, the amount of composite adhering to the enamel substrate was quantified using the ARI (Årtun and Bergland, 1984; Cacciafesta et al. 2002; Summers et al., 2004). A score of 1 was assigned when all the composite remained on the enamel surface and a clear imprint of the bracket was evident; 2, when more than 90 per cent of the composite was present on the surface; 3, when 10–90 per cent of the composite remained on the surface; 4, when less than 10 per cent of the composite remained; and 5, when no composite remained on the enamel surface, i.e. where there was complete transfer of composite to the bracket base.
|
Results |
|---|
|
TopSummaryIntroductionMaterials and methodsResultsDiscussionConclusionsReferences |
|---|
Typical load–displacement curves in tension related to each adhesive system are presented in Figure 3. Groups A, B, and C showed a failure load of 60 N [standard deviation (SD) 17], 114 N (SD 14), and 200 N (SD 12), respectively. For tensile testing, the statistical comparison (Cheremisinoff, 1987
) between the failure loads using the F-test and the level of probability (P) showed that there were significant differences between the three bonding materials. In particular, the difference between the distributions of the maximum loads of groups A and B was represented by F = 17.3 and P = 1.9 x 10–3, while that between groups A and C was F = 75.3 and P = 1.15 x 10–5 and finally the difference between group B and C was F = 12.4 and P = 4.8 x 10–3. Therefore, a significant difference was identified between groups A, B, and C.
|
Typical shear load–displacement is shown in Figure 4. For shear testing, the difference between the failure loads of groups A and B was statistically significant (F = 23.3 and P = 7.0 x 10–4). A similar result was observed when comparing groups A and C (F = 21.9 and P = 1.1 x 10–3), but no statistical difference was found between groups B and C (F = 1.18 and P = 0.29).
|
Typical torque–rotation is shown in Figure 5. In torsion testing, the difference between the failure torque of groups A and B was statistically significant (F = 12.4 and P = 4.8 x 10–3). A significant difference was also observed between groups A and C (F = 15.5 and P = 3.4 x 10–3), while a weaker statistical difference was found between groups B and C (F = 4.9 and P = 0.05).
|
The values of the ARI index for the three groups according to the type of mechanical test are depicted in Table 2. One-way ANOVA testing showed that there was a statistically significant difference in retention between groups A and B (P < 0.05) and between groups A and C (P < 0.05). The ARI values obtained from the shear tests indicated that the statistical difference was limited to groups A and B (P < 0.05), while the torsion test did not show a significant difference between the three materials. These data suggest that the application of a torsional moment resulted in a similar type of fracture, regardless of the adhesive material used. Furthermore, analysis of the fracture surfaces by optical microscopy confirmed the absence of enamel lesions in all of the samples debonded by torsion.
|
|
Discussion |
|---|
|
TopSummaryIntroductionMaterials and methodsResultsDiscussionConclusionsReferences |
|---|
In tensile, shear, and torsion tests, the specimens can be considered as a series of bracket–adhesive–enamel: the load is the same for each element of the series, stainless steel, polymer composite, and enamel substrate, while the value of deformation for each element of the series is different. The deformation, i.e. the axial and angular displacement, as shown in Figures 3 and 5, is, in each case, the sum of the contributions of the single elements. For this reason, comparative statistical analysis is appropriate only regarding the failure loads of each sample. In particular, in the tensile and torsion tests (Figures 3 and 5), the samples with the Transbond adhesive (group C) reached loads that deformed the wings of the bracket. Therefore, this deformation quota characterizes the total deformation of the system (group C; Figure 3). The wing deformation of the bracket is clearly evident from the decrease of the slope of the force–displacement curve at about 100 N, where plastic deformation of the wings started to occur (region 2; Figure 3). In group C, the wings converge with increasing load, up to complete closing of the slot (Figure 3c). At this point, an increase of the slope was detected (region 3; Figure 3).
The results indicate a statistically significant difference between the tensile debonding loads of the adhesive systems investigated. Group C showed the highest debonding value, while group A exhibited the weakest tensile debonding load. The tensile load required to debond group A brackets (Figure 3) and the related ARI score (Table 2) are close to the values measured by Algera et al. (2005). It is interesting to observe that the ARI score indicates that acid etching of the substrate (group C) improves the retention of the adhesive material on the enamel. This result is consistent with finding of Abu Alhaija and Al-Wahadni (2004).
There was a statistically significant difference between the shear debonding loads of groups A and B (P < 0.01), as well as between those of groups A and C (P < 0.01). Fuji GC showed the weakest shear debonding load, while no difference was observed between groups B and C. Again, there was a significant statistical difference between the torsional debonding loads observed for groups A and B (P < 0.01), and when comparing values for groups A and C (P < 0.01). The results showed that Fuji GC had the weakest bonding capability. This result is consistent with adhesive shear strength reported by Movahhed et al. (2005). However, if shear stress is extrapolated by dividing the shear force by the bracket base, values of 14 and 26 MPa were obtained for groups A and C, respectively. These values are higher than those found by Movahhed et al. (2005) who used the same adhesive materials. This difference can be ascribed to two main aspects: the duration of the conditioning which in the present protocol was longer, and the application of a load of 300 g on the bracket during bonding.
It is suggested that mechanical properties of light-cured composites undergo a transitory change (Versluis et al., 2004), i.e. the mechanical properties significantly increase with time until a plateau is reached. In addition, the adhesive used to bond brackets to enamel also undergoes transient change in properties. The findings of Wendl and Droschl (2004) clearly show that the strength of directly bonded brackets increases during the first 24 hours. Consequently, the measurements in the present investigation were performed after 24 hours, and the recorded values were higher than those reported by other authors who used shorter conditioning periods (Cacciafesta et al., 2002; Movahhed et al. 2005). Despite the different testing protocols, the ARI values related to the shear tests of groups A and C (Table 2) are consistent with the findings of Cacciafesta et al. (2002) and Movahhed et al. (2005).
With the bracket–adhesive–tooth system, the metal bracket is the least critical variable. Accordingly, in accidental debonding which may occur during orthodontic treatment, the fracture pattern is generally localized at the level of the enamel–composite interface (Lopez, 1980). In clinical practice, the choice of a bonding system is based upon the adhesion capability of the material as well as the handling procedure (Regan and Van Noort, 1989). For example, light-curable materials such as Transbond combine high adhesion values with easy handling, virtually unlimited positioning time, and, therefore, high positioning accuracy. Moreover, the retentive characteristics of bracket bases have evolved to such a level that, usually, a fracture at the composite–bracket interface is uncommon in clinical trials (Smith and Maijer, 1983). Rather, such fractures tend to occur at the enamel–composite interface and are almost always caused by errors in operating procedures or in the manipulation of the adhesive materials (Maijer and Smith, 1981). Indeed, when these materials are used correctly, the adhesion values attained are so high that they may cause enamel lesions during debonding. As noted previously, plastic deformation of the wings of the brackets of groups B and C was commonly observed, indicating a strong mechanical interface.
Identification of debonding methods which are less detrimental to the enamel surface should, therefore, be a key objective of research in this area. Regan and Van Noort (1989) showed that higher debonding strength values are obtained when brackets are subjected to shear rather than to tensile stresses. Thus, in clinical practice, it would be better to apply tensile loads in the debonding phase. In the present study, torsion stress resulted in less loading being required for debonding with torque debonding values (Figure 5) ranging from 0.2 to 0.4 Nm. Consequently, when using a lever with a length of 1 cm, a load between 20 and 40 N would be required to debond the bracket. In addition, debonding by torsional stress was the least damaging to the enamel.
Groups A and C specimens were cured using an 800 mW/cm2 halogen lamp, which is the most widely used type of light source for this application. Although the degree of conversion of a photo-cured composite is expected to be dependent upon the light source, Cacciafesta et al. (2001) suggested that there was no statistically significant variation in debonding stress with different light sources, at least when quartz-halogen or xenon lamps were used. In the current research, therefore, the hypothesis was that bonding strength is independent of the light unit. Moreover, the light energy dose supplied in 40 seconds is sufficiently high, at 32 J/cm2, to consider the composite material to be adequately cured (Musanje and Darvell, 2003).
This research clearly demonstrates that, as far as debonding procedures are concerned, both material and mechanical stress must be considered. Tensile testing showed that the Transbond adhesive system was the most resistant (Fmax = 68.6 ± 15.7 N). Furthermore, the weakest interface in the sample was between bracket and adhesive, as confirmed by an ARI score of 1.63. This indicates that most of the material remained attached to the enamel surface. In contrast, shear testing resulted in fractures occurring mainly at the enamel–adhesive interface. Using optical microscopy of the dental surface after debonding of the bracket, small cohesive fractures of the enamel were observed. This type of stress, in vitro, therefore seems to be most damaging to the enamel surface.
High strength values were achieved with Concise and Transbond, while Fuji always showed lower adhesion (Figures 3–5
). Torsion testing gave more consistent data for debonding strengths. This emphasizes the importance of investigating a system by a variety of methods, in order to understand to what extent the type of mechanical test used influences the behaviour of the system.
Comparing the effects of the shear and torsion tests (Figure 2b,c), torsion was least damaging to the enamel (Table 2). Currently, the debonding of brackets is carried out using unstandardized procedures, with many variable parameters. Forces that are uncalibrated, in terms of intensity and direction, are applied in an uncontrolled manner. This means that debonding tests that exclusively use only one type of force (i.e. shear, torsion or, tensile) are not clinically relevant.
|
Conclusions |
|---|
|
TopSummaryIntroductionMaterials and methodsResultsDiscussionConclusionsReferences |
|---|
The results of this research have highlighted some clinical considerations. It is necessary to identify the optimum procedure for the preparation of the bonding material and the type of stress which, together, prove to be the least detrimental to the enamel during debonding. In clinical practice, the stresses applied on the brackets during debonding are a combination of tensile, shear, and torsion forces. The tests used in the present study indicate that torsional stresses result in the least enamel damage.
The methods used for mechanical debonding tests are easy to set up and provide reliable results. The Transbond luting system gave higher adhesion values compared with Concise and Fuji Ortho.
|
Acknowledgement |
|---|
The authors acknowledge Dr Federica Taglialatela for specimen preparation and Mr Rodolfo Morra for mechanical testing.
|
References |
|---|
|
TopSummaryIntroductionMaterials and methodsResultsDiscussionConclusionsReferences |
|---|
-
Abu Alhaija ES, Al-Wahadni AMS. Evaluation of shear bond strength with different enamel pre-treatments. European Journal of Orthodontics (2004) 26:179–184.
Algera TJ, Kleverlaan CJ, de Gee AJ, Prahl-Andersen B, Feilzer AJ. The influence of accelerating the setting rate by ultrasound or heat on the bond strength of glass ionomers used as orthodontic bracket cements. European Journal of Orthodontics (2005) 27:472–476.
Årtun J, Bergland S. Clinical trials with crystal growth conditioning as an alternative to acid-etch pretreatment. American Journal of Orthodontics (1984) 85:333–340.[CrossRef][Web of Science][Medline]
Bishara SE, Gordan VV, VonWald L, Jakobsen JR. Shear bond strength of composite, glass ionomer, and acidic primer adhesive systems. American Journal of Orthodontics and Dentofacial Orthopedics (1999) 115:24–28.[CrossRef][Web of Science][Medline]
Cacciafesta V, Sfondrini MF, Klersy C, Sfondrini G. Polymerization with a micro-xenon light of a resin-modified glass ionomer: a shear bond strength study 15 minutes after bonding. European Journal of Orthodontics (2002) 24:689–697.
Cacciafesta V, Sfondrini MF, Sfondrini G. Fotopolimerizzazione ad alta densità di un adesivo ortodontico mediante lampada al micro-xenon. Ortognatodonzia Italiana (2001) 10:301–306.
Cheremisinoff NP. Practical statistics for engineers and scientists (1987) Lancaster: Technomic Publishing Company Inc. 71–112.
Eliades T, Brantley WA. The inappropriateness of conventional orthodontic bond strength assessment protocols. European Journal of Orthodontics (2000) 22:13–23.
Fox NA, McCabe JF, Buckley JG. A critique of bond strength testing in orthodontics. British Journal of Orthodontic (1994) 22:33–43.
Lopez JI. Retentive shear strengths of various bonding attachment bases. American Journal of Orthodontics (1980) 77:669–678.[CrossRef][Web of Science][Medline]
Maijer R, Smith DC. Variables influencing the bond strength of metal orthodontic bracket bases. American Journal of Orthodontics (1981) 79:20–34.[CrossRef][Web of Science][Medline]
Movahhed HZ, Øgaard B, Syverud M. An in vitro comparison of the shear bond strength of a resin-reinforced glass ionomer cement and a composite adhesive for bonding orthodontic brackets. European Journal of Orthodontics (2005) 27:477–483.
Musanje L, Darvell BW. Polymerization of resin composite restorative materials: exposure reciprocity. Dental Materials (2003) 19:531–541.[CrossRef][Web of Science][Medline]
Oesterle LJ, Shellhart WC, Belanger GK. The use of bovine enamel in bonding studies. American Journal of Orthodontics and Dentofacial Orthopedics (1998) 114:514–519.[CrossRef][Web of Science][Medline]
Regan D, Van Noort R. Bond strengths of two integral bracket-base combinations: an in vitro comparison with foil-mesh. European Journal of Orthodontics (1989) 11:144–153.
Ruse ND, Smith DC, Torneck CD, Titley KC. Preliminary surface analysis of etched, bleached and normal bovine enamel. Journal of Dental Research (1990) 69:1610–1613.
Saleh F, Taymour N. Validity of using bovine teeth as a substitute for human counterparts in adhesive tests. East Mediterranean Health Journal (2003) 9:201–217.
Smith DC, Maijer R. Improvements in bracket base design. American Journal of Orthodontics (1983) 83:277–281.[CrossRef][Web of Science][Medline]
Summers A, Kao E, Gilmore J, Gunel E, Ngan P. Comparison of bond strength between a conventional resin adhesive and a resin-modified glass ionomer adhesive: an in vitro and in vivo study. American Journal of Orthodontics and Dentofacial Orthopedics (2004) 126:200–206.[CrossRef][Web of Science][Medline]
Versluis A, Tantbirojn D, Douglas WH. Distribution of transient properties during polymerisation of a light-initiated restorative composite. Dental Materials (2004) 20:543–553.[CrossRef][Web of Science][Medline]
Wendl B, Droschl H. A comparative in vitro study of the strength of directly bonded brackets using different curing techniques. European Journal of Orthodontics (2004) 26:535–544.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||