All C opyrig

eR

ht
ech
te

by
vo
rbe

Qu
ha
n lte
n
Bond Strength Durability of Direct and Indirect t es

i
se nz
Composite Systems Following Surface Conditioning
for Repair
Sheila Pestana Passosa/Mutlu Özcanb/Aleska Dias Vanderleia/Fabiola Pessoa Pereira Leitea/
Estevão Tomomitsu Kimparaa/Marco Antonio Bottinoc

Purpose: This study evaluated the effect of surface conditioning methods and thermocycling on the bond strength be-
tween a resin composite and an indirect composite system in order to test the repair bond strength.

Materials and Methods: Eighteen blocks (5 x 5 x 4 mm) of indirect resin composite (Sinfony) were fabricated according
to the manufacturer’s instructions. The specimens were randomly assigned to one of the following two treatment condi-
tions (9 blocks per treatment): (1) 10% hydrofluoric acid (HF) for 90 s (Dentsply) + silanization, (2) silica coating with 30-
Ìm SiOx particles (CoJet) + silanization. After surface conditioning, the bonding agent was applied (Adper Single Bond)
and light polymerized. The composite resin (W3D Master) was condensed and polymerized incrementally to form a block.
Following storage in distilled water at 37°C for 24 h, the indirect composite/resin blocks were sectioned in two axes
(x and y) with a diamond disk under coolant irrigation to obtain nontrimmed specimens (sticks) with approximately
0.6 mm2 of bonding area. Twelve specimens were obtained per block (N = 216, n = 108 sticks). The specimens from each
repaired block were again randomly divided into 2 groups and tested either after storage in water for 24 h or thermocy-
cling (6000 cycles, 5°C to 55°C). The microtensile bond strength test was performed in a universal testing machine
(crosshead speed: 1 mm/min). The mean bond strengths of the specimens of each block were statistically analyzed us-
ing two-way ANOVA (α = 0.05).

Results: Both surface conditioning (p = 0.0001) and storage conditions (p = 0.0001) had a significant effect on the re-
sults. After 24 h water storage, silica coating and silanization (method 2) showed significantly higher bond strength re-
sults (46.4 ± 13.8 MPa) than that of hydrofluoric acid etching and silanization (method 1) (35.8 ± 9.7 MPa) (p < 0.001).
After thermocycling, no significant difference was found between the mean bond strengths obtained with method 1
(34.1 ± 8.9 MPa) and method 2 (31.9 ± 7.9 MPa) (p > 0.05).

Conclusion: Although after 24 h of testing, silica coating and silanization performed significantly better in resin-resin re-
pair bond strength, both HF acid gel and silica coating followed by silanization revealed comparable bond strength re-
sults after thermocycling for 6000 times.

Keywords: hydrofluoric acid, indirect composite, repair, silica coating, microtensile test.

J Adhes Dent 2007; 9: 443-447. Submitted for publication: 08.09.06; accepted for publication: 11.01.07.

T he advancement in filler and polymer technology allowed

for the development of resin composites for indirect ap-
plications. With the use of indirect resin composites, better
a PhD student, Department of Dental Materials and Prosthodontics, São Paulo
proximal contacts, excellent occlusal morphology, good mar-
State University at São José dos Campos, Brazil.
ginal adaptation with lower polymerization shrinkage of resin
b Professor and Research Associate, University Medical Center Groningen, Uni-
luting agents, good flexural strength, and easy manipulation
versity of Groningen, Department of Dentistry and Dental Hygiene, Clinical and polishing could be achieved.5,16 Although marked im-
Dental Biomaterials, Groningen, The Netherlands. provements have been noted with such materials in terms
c Professor and Chair, Department of Dental Materials and Prosthodontics, São of physical and mechanical properties during the last 10 to
Paulo State University at São José dos Campos, Brazil. 20 years, and good clinical performance was reported,15 the
Reprint requests: Dr. Mutlu Özcan, University Medical Center Groningen, Uni- major problems still seemed to be related to chipping,
versity of Groningen, Department of Dentistry and Dental Hygiene, Clinical Den-
tal Materials, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands. Tel:
delamination, or fractures.6 Chipping or delamination of
+31-50-363 85 28, Fax: +31-50-363 26 96. e-mail: mutluozcan@[Link] these restorations does not necessarily indicate restoration

Vol 9, No 5, 2007 443

 All C opyrig
eR

ht
ech
te

by
vo
rbe

Qu
Passos et al
ha
n lte
n
te

i
failure, since these fractures may not lead to the need for re- vice (Micro-Etcher, Danville; San Ramon, CA, USA) filled with
placement of the restorations. Repair, as an alternative to CoJet-Sand (30-Ìm Al2O3 particles coated with silica) e nz
ss(CoJet-
complete removal or replacement, would preserve the tooth, Sand, CoJet system, 3M ESPE; Seefeld, Germany; batch
as it is often difficult to remove an adhesive restoration with- #142820) perpendicular to the surface from a distance of
out removing an integral part of the tooth.6,8 approximately 10 mm for a period of 4 s at 2.8 bar pressure.
Several techniques are available in order to improve the The conditioned substrates were then coated with a
bond strength of repair resins to indirect resin composites, 3-methacryloxypropyltrimethoxysilane coupling agent, γ-
such as surface etching with hydrofluoric (HF) acid gel, air MPS (ESPE-Sil, 3M ESPE; Seefeld, Germany) and allowed to
abrasion with alumina or silica particles, or utilization of ad- react for 5 min.
hesive systems.9 Despite the hazardous effects of HF acid
gel, etching the surface of a composite restoration with this Bonding Procedure
acid followed by application of silane coupling agent is a well- The conditioned resin blocks were coated with a thin layer of
known and recommended method to increase the bond adhesive resin and air thinned following the manufacturer’s
strength. An alternative method to HF acid etching for repair instructions (Adper Single Bond, 3M ESPE; Seefeld, Ger-
actions is based on airborne particle abrasion using silica many; batch #5EK). Then resin composite was packed
particles. In this technique, the surfaces are abraded with (W3D-Master, Wilcos, Petrópolis, RJ, Brazil; batch #006/03)
silica-coated aluminum oxide particles.13 The pressure of in approximately 2 mm increments and each layer was light
the air abrasion leads to coverage of the restoration surface polymerized for 40 s (XL 3000, 3M ESPE; Seefeld, Germany)
with silica and alumina particles, making it chemically more (light output: 500 mW/cm2) to obtain blocks with a height of
reactive to silane coupling agent and the adhesive resin.11 ca 5 mm.
Although repair methods seem to deliver satisfactory re- After completion of the indirect resin composite/repair
sults, their long-term clinical durability is not widely docu- resin composite assemblies, the specimens were stored in
mented.12 There is also a need for repair methods to be test- distilled water at 37°C for 1 week until preparation of spec-
ed under simulated conditions in in vitro settings. Therefore, imens for the microtensile test.
the objective of this study was to evaluate the effect of sur- The specimens from each repaired block were tested ei-
face conditioning and storage conditions on the bond ther after storage in water for 24 h or thermocycling (6000
strength between a direct resin composite and an indirect cycles 5°C to 55°C, dwell time: 30 s, transfer time from one
resin composite system for repair purposes. bath to the other: 2 s) (Nova Etica; São Paolo, Brazil).

MATERIALS AND METHODS Production of Specimens for Microtensile Test

Composite blocks were sectioned using a diamond disk (Mi-
An acrylic resin template was machined with dimensions of crodont; São Paulo, Brazil, n. 34570) at low speed under wa-
5 x 5 x 4 mm and impressions were made from this acrylic ter cooling in a sectioning machine.1 Initially, the blocks
block with addition silicone putty (Express, 3M ESPE; St were fixated with cyanoacrylate adhesive gel (Super Bonder
Paul, MN, USA; batch #7312) in order to achieve standard Gel, Loctite; São Paulo, Brazil) on a metallic base that was
molds for the fabrication of 18 indirect resin composite attached to the sectioning machine. The blocks were posi-
blocks with the same dimensions. Indirect resin composite tioned as perpendicularly as possible in relation to the dia-
(Sinfony, 3M ESPE; Seefeld, Germany) was incrementally in- mond disk of the machine. The first section, measuring ap-
jected into the mold and light polymerized initially using a proximately 0.5 mm, was discarded in case of the possibili-
hand light unit (XL 3000, 3M ESPE; Seefeld, Germany) (light ty of excess or absence of adhesive at the interface that
intensity: 500 mW/cm2). The specimens were further poly- might alter the results. Thereafter, four sections measuring
merized in a special polymerization device (Visio Beta Vario, 0.8 ± 0.1 mm in thickness were achieved.
3M ESPE) following the manufacturer’s recommendations. Each section was rotated 90 degrees and once again
The repair surface (5 x 5 mm) was leveled and polished in a fixated to the metallic base. The first section was discarded
machine using silicone carbide papers in sequence (600, (0.5 mm) due to the aforementioned reasons. Subsequently,
800, and 1200 grit) under cooling (KG-Sorensen; Barueri, four other sections were obtained, also measuring 0.8 ± 0.1
Brazil) and cleaned ultrasonically in distilled water (Vitason- mm in thickness. This process was followed for the other
ic, Vita Zanhfabrik; Bad Säckingen, Germany) for 10 min. three sections, and thus only the central specimens were
The blocks were then randomly divided into 2 groups (N used for the experiments. Twelve specimens were obtained
= 18, 9 per group) and conditioned by one of the following from each block. The rectangular beam specimens had a
methods: non-machined (nontrimmed) adhesive zone with a bonded
area measuring approximately 0.6 mm2 and ca 0.8 mm
Surface Conditioning Methods length. The specimen preparation has previously been de-
Method 1: The resin composite substrates were etched with scribed in detail elsewhere.17
10% HF acid gel (Dentsply; Petropolis, Brazil) for 90 s and
washed and rinsed thoroughly in accordance with the man- Microtensile Bond Strength Test
ufacturer’s recommendations. Keeping the adhesive zone free, each specimen was fixated
Method 2: A chairside tribochemical silica coating system with cyanoacrylate gel (Super Bonder Gel, Loctite) to the rods
was employed using an intraoral chairside air-abrasion de- of a device adapted for this test. The specimens were posi-

444 The Journal of Adhesive Dentistry

 All C opyrig
eR

ht
ech
te

by
vo
rbe

Qu
Passos et al
ha
nt lte
n

i
Table 1 Results of two-way analysis of variance for the experimental conditions
e ss e n z
Source of variation df SS MS F P

Surface conditioning 1 3590.20 3590.16 33.99 0.0001

Storage 1 2218.30 2218.35 21.00 0.0001
Interaction 1 978.80 978.65 9.26 0.0026
Error 217 22922.80 105.64
Total 220

70
80
70 60

60 50
50 a 40
a P 30
40
P M
M 30 20
20
10
10
0
0 without with without with
without with without with HF Cojet
HF Cojet

Fig 1 The mean microtensile bond strength values (MPa) for the Fig 2 The mean microtensile bond strength values (MPa) and stan-
two surface conditioning methods and the storage conditions (with dard deviations for the two surface conditioning method with and
and without thermocycling) in dispersion diagram (dot plot). without thermocycling.

tioned parallel to the long axis of the device in order to re- RESULTS
duce the bending stresses. The device was fixated in the uni-
versal testing machine (EMIC DL-1000, EMIC; São José dos No pretest failures were experienced throughout the experi-
Pinhais, Brazil) as parallel as possible in relation to applica- mental procedures.
tion of the tensile load, and testing was performed at a Two-way ANOVA revealed a significant influence of the
crosshead speed of 1 mm/min. storage conditions (p = 0.0001) and the surface condition-
The bond strength was calculated according to the for- ing methods (p = 0.0001) on the test results. Interaction be-
mula R=F/A, where “R” is the strength (MPa), “F” is the load tween factors was also significant (p = 0.0026) (Table 1).
required for rupture of the specimen (N) and “A” is the inter- Means and standard deviations are presented in Fig 1.
face area of the specimen (mm2), measured with a digital After 24-h water storage, silica coating and silanization
caliper before the test. (method 2) showed significantly higher bond strength results
(46.4 ± 13.8 MPa) than that of hydrofluoric acid etching and
Topographic Analysis of Conditioned Indirect Resin Com- silanization (method 1) (35.8 ± 9.7 MPa) (p < 0.001). After
posite Surface thermocycling, however, no significant difference was found
Additional indirect resin composite specimens (n = 4) were between the mean bond strengths obtained with method 1
conditioned using the two surface conditioning methods in (31.9 ± 7.9 MPa) and method 2 (34.1 ± 8.9 MPa) (p > 0.05)
order to observe the topographic surface changes under a (Fig 2).
scanning electron microscope (DSM 962, Zeiss; Jena, Ger- SEM analysis at 5000X magnification, complementary to
many). the microtensile bond strength tests, revealed that HF acid
gel dissolved the fillers of the substrate and produced
Statistical Analysis porous, irregular surfaces (Fig 3a). On the other hand, the sil-
The means of each group were analyzed by two-way analy- ica-coated specimens exhibited a rough surface covered
sis of variance (ANOVA), with microtensile bond strength as with abundant sand particles (Fig 3b).
the dependent variable and the surface conditioning meth-
ods and storage conditions as the independent factors (Sta-
tistix 8.0 for Windows, Analytical Software; Tallahassee, FL, DISCUSSION
USA). P values less than 0.05 were considered to be statis-
tically significant in all tests. According to Crispin,3 indirect resins are more easily re-
paired than ceramics, because of the similarity between

Vol 9, No 5, 2007 445

 All C opyrig
eR

ht
ech
te

by
vo
rbe

Qu
Passos et al
ha
nt lte
n

i
e ss e n z

Figs 3a Representative SEM image (5000X magnification) of a hy- Fig 3b Representative SEM image (5000X magnification) of a sili-
drofluoric acid etched specimen (Sinfony). Note that the acid treat- ca coated specimen. Note that the surfaces are covered with abun-
ment dissolved the filler components of the substrates (*). dant sand particles even after air drying (←).

substrates. In a previous study by Özcan et al,10 three sur- should be evaluated with caution, because discrepancies
face treatments, namely, hydrofluoric acid application and may simply be due to the differences between test methods.
air abrasion with alumina (50 μm) or silica particles (30 μm), The hazardous and extremely caustic effects to soft tis-
for various indirect resins were employed for repair purpos- sues make the clinical use of HF acid gels a controversial af-
es. It was reported that the bond strength varied according fair, due to its rapid vaporization and the danger of inhala-
to the type of resin and conditioning method investigated tion. Although no clinical incidence report exists in the den-
when the shear bond test was used, with the results being tal literature regarding harmful consequences of HF acid gel
more favorable for silica coating and silanization than for hy- use, caution should be exercised during handling this ma-
drofluoric acid etching and silanization. Although SEM find- terial. Considering the possible hazardous effects and the
ings were similar to those of that study, in this study, neither nonsignificant differences between hydrofluoric acid etching
surface conditioning method showed significant differences and silica coating, clinicians should consider the use of the
after thermocycling for the substrate material tested. How- latter for safer applications. However, after both condition-
ever, interaction terms were significant; therefore, it was not ing methods, silane application is compulsory. Clinically, for
possible to draw a conclusion on the main effect. The mean invisible repair procedures, a bevel should be created by a
bond strength results with both methods, and particularly carbide bur or other ultrasonic oscillating devices. Addition-
with HF acid gel application and silanization (31.9 ± 7.9 MPa), ally, non-contaminated surfaces should be achieved using
were higher than those of Özcan et al10 (5.2 ± 1.2 MPa), al- prophylaxis paste and rubber cups.
though the monomer matrices of the indirect composite sub- While hydrofluoric acid is considered hazardous, silica
strate material (HEMA and octahydro-4,7-methano-1H-in- coating requires additional equipment in the dental practice,
denediyl)bis(methylene)-diacrylate) and the repair resin (bis- compared to simply drilling the surface and applying an in-
GMA) used in this study were similar. termediate adhesive resin. However, this approach has re-
Several factors may have contributed to the differences in cently been shown to create almost exclusively adhesive
the results, one of which could be the testing methodology. failures between the substrate and the adherend, implying
The shear test is not considered an ideal mechanical test, the necessity of silica coating and silanization that almost
since it often leads to nonuniform distribution of the stress- exclusively resulted in cohesive failures in the substrate.2
es at the adhesive area. In this kind of test, maximum tensile
forces occur close to the load application area, thereby af-
fecting the substrate more than the adhesive interface it- CONCLUSIONS
self.4 The microtensile test, suggested by Sano et al,14 as-
sesses the bond strength of specimens with reduced areas When repaired specimens were not thermocycled, silica
of adhesive joint, where fractures occur basically at the ad- coating and silanization showed significantly higher bond
hesive interface. For these reasons, in this study a microten- strength results than that of hydrofluoric acid etching and
sile test was used. Therefore, the results of mechanical tests silanization. After thermocycling (6000 cycles), however, no

446 The Journal of Adhesive Dentistry

 All C opyrig
eR

ht
ech
te

by
vo
rbe

Qu
Passos et al
ha
n tPK. Effect of lte
n

i
11. Özcan M, Lassila L, Raadschelders J, Matinlinna JP, Vallittu
significant differences were found between the two repair
methods.
some parameters on silica-deposition on a zirconia ceramic e[abstract
ss e n z
63114]. J Dent Res 2005;84.
12. Özcan M. Longevity of repaired composite and metal-ceramic restora-
tions:3.5 year clinical study [abstract 0076]. J Dent Res 2006;85.
REFERENCES 13. Peutzfeldt A, Asmussen E. Silicoating. Evaluation of a new method of bond-
ing composite resin to metal. Scand J Dent Res 1988;96:171-176.
14. Sano H, Shono T, Sonoda H, Takatsu T, Ciucchi B, Carvalho RM. Relation-
1. Andreatta Filho OD, Galhano G, Bottino MA, Camargo FP, Valandro LF,
ship between surface area for adhesion and tensile bond strength-evalua-
Nishioka RS. Evaluation of the bond strength between an aluminous ce-
tion of a micro-tensile bond test. Dent Mater 1994;10:236-240.
ramic and a resin cement: Effect of thermo-cycling. Braz Dent Sci
2003;6:32-39. 15. Scherrer SS, Wiskott AH, Coto-Hunziker V, Belser UC. Monotonic flexure
and fatigue strength of composites for provisional and definitive restora-
2. Brendeke J, Özcan M. Effect of aging and adhesion mediums on the adhe-
tions. J Prosthet Dent 2003;89:6:579-588.
sion of particulate filler composites to composites. J Adhes Dent
2007;9:399-406. 16. Touati B, Aidan N. Second generation laboratory composite resins for indi-
rect restorations. J Esthetic Dent 1997;9:108-118.
3. Crispin BJ. Indirect composite restorations: alternative of replacement for
ceramic? Compendium 2002;23:611-624. 17. Valandro LF, Özcan M, Bottino MC, Bottino MA, Scotti R, Bona AD. Bond
strength of a resin cement to high-alumina and zirconia-reinforced ceram-
4. Della Bona A, Van Noort R. Shear vs. Tensile bond strength of resin com-
ics: the effect of surface conditioning. J Adhes Dent 2006;8:175-181.
posite bonded to ceramic. J Dent Res 1995;74:1591-1596.
5. Duke ES. The introduction of a new class of composite resin: ceromers.
Compendium 1999;20:246-247.
6. Kukrer D, Gemalmaz D, Kuybulu EO, Bozkurt FO. A prospective clinical
study of ceromer inlays: results up to 53 months. Int J Prosthodont
2004;17:17-23.
7. Mjör IA, Gordan VV. Failure, repair, refurbishing and longevity of restora-
tions. Oper Dent 2002;27:528-534.
8. Özcan M, Niedermeier W. Clinical study on the reasons for and location of
Clinical relevance: Both silica coating plus silanization
failures of metal-ceramic restorations and survival of repairs. Int J Prostho- and hydrofluoric acid etching plus silanization could be ad-
dont 2002;15:299-302. vised for the repair of the indirect resin composite and re-
9. Özcan M. Evaluation of alternative intraoral repair techniques for fractured pair resin combination tested in this study. However, clin-
ceramic-fused-to-metal restorations. J Oral Rehabil 2003;30:194-203.
10. Özcan M, Alander P, Vallitu PK, Huysmans MC, Kalk W. Effect of three sur-
icians may consider choosing silica coating and silaniza-
face conditioning methods to improve bond strength of particulate filler tion, due to the hazardous effects of hydrofluoric acid gel.
resin composites. J Mater Sci Mater Med 2005;16:21-27.

Vol 9, No 5, 2007 447