d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
Composite cements benefit from light-curing Anne-Katrin Lührs a,b,∗ , Jan De Munck a , Werner Geurtsen b , Bart Van Meerbeek a a
KU Leuven – BIOMAT, Department of Oral Health Sciences, KU Leuven (University of Leuven) & Dentistry, University Hospitals Leuven, Belgium b Department of Conservative Dentistry, Periodontology and Preventive Dentistry, Hannover Medical School, Hannover, Germany
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. To investigate the effect of curing of composite cements and a new ceramic
Received 8 August 2013
silanization pre-treatment on the micro-tensile bond strength (TBS).
Received in revised form
Methods. Feldspathic ceramic blocks were luted onto dentin using either Optibond
25 November 2013
XTR/Nexus 3 (XTR/NX3; Kerr), the silane-incorporated ‘universal’ adhesive Scotchbond Uni-
Accepted 28 November 2013
versal/RelyX Ultimate (SBU/RXU; 3M ESPE), or ED Primer II/Panavia F2.0 (ED/PAF; Kuraray Noritake). Besides ‘composite cement’, experimental variables were ‘curing mode’ (‘AA’: complete auto-cure at 21 ◦ C; ‘AA*’: complete auto-cure at 37 ◦ C; ‘LA’: light-curing of adhesive
Keywords:
and auto-cure of cement; ‘LL’: complete light-curing) and ‘ceramic surface pre-treatment’
Composite cement
(‘HF/S/HB’: hydrofluoric acid (‘HF’: IPS Ceramic Etching Gel, Ivoclar-Vivadent), silaniza-
Adhesive
tion (‘S’: Monobond Plus, Ivoclar-Vivadent) and application of an adhesive resin (‘HB’:
Curing
Heliobond, Ivoclar-Vivadent); ‘HF/SBU’: ‘HF’ and application of the ‘universal’ adhesive
Ceramic
Scotchbond Universal (‘SBU’; 3M ESPE, only for SBU/RXU)). After water storage (7 days at
Silane
37 ◦ C), ceramic–dentin sticks were subjected to TBS testing.
Universal adhesive
Results. Regarding the ‘composite cement’, the significantly lowest TBSs were measured for ED/PAF. Regarding ‘curing mode’, the significantly highest TBS was recorded when at least the adhesive was light-cured (‘LA’ and ‘LL’). Complete auto-cure (‘AA’) revealed the significantly lowest TBS. The higher auto-curing temperature (‘AA*’) increased the TBS only for ED/PAF. Regarding ‘ceramic surface pre-treatment’, only for ‘LA’ the TBS was significantly higher for ‘HF/S/HB’ than for ‘HF/SBU’. Significance. Complete auto-cure led to inferior TBS than when either the adhesive (on dentin) or both adhesive and composite cement were light-cured. The use of a silaneincorporated adhesive did not decrease luting effectiveness when also the composite cement was light-cured. © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Indirect ceramic restorations became more popular during the last years because of their superior esthetics,
∗
biocompatibility and long-term stability. So-called ‘etchable’ feldspathic (and glass-) ceramic restorations should be adhesively luted; it improves the fracture resistance and consequently enhances the survival rate [1]. Possible curing modes of composite cements are ‘dual-cure’, involving also
Corresponding author at: KU Leuven – BIOMAT, Department of Oral Health Sciences, KU Leuven (University of Leuven) & Dentistry, University Hospitals Leuven, Belgium/Department of Conservative Dentistry, Periodontology and Preventive Dentistry, Hannover Medical School, OE 7740, Carl-Neuberg-Straße 1, 30625 Hannover, Germany. Tel.: +49 532 4718; fax: +49 532 4811. E-mail address: [email protected] (A.-K. Lührs). 0109-5641/$ – see front matter © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dental.2013.11.012
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
light-curing, and ‘auto-cure’. The beforehand applied adhesive, if any (not in case of self-adhesive composite cements), can be separately light-cured on the condition that its film thickness is sufficiently thin and does not impair the restoration fit. When adhesively luted ceramic restorations are light-cured, both the adhesive and cement may however not cure completely due to light attenuation, caused by the opacity of the restoration, its shade, and/or simply its thickness, the latter obviously related to cavity depth [2,3]. A ceramic thickness of 2–3 mm is considered to be the threshold to still effectively light-cure adhesively luted ceramic restorations [4,5]. Up to a thickness of 2 mm, there was according to Akgungor et al. [6] no effect of both ceramic thickness and polymerization mode on bond strength. For thicker restorations of 4 mm, a lower micro-tensile bond strength (TBS) was detected [2]. Light attenuation already starts at a composite thickness of 1 mm, as the light intensity was reduced about 85%, and restorations of 4 mm thickness blocked the light almost completely [7]. It was also reported that lightcuring the adhesive and the composite cement separately improved the bond strength to dentin [2,8]. Furthermore, autocuring was found to influence the degree of conversion (DC) of composite cements and their mechanical properties [9–11]. Depending on the material, the bond strength of composite cements that were allowed to auto-cure was inferior to when they were light-cured [12–16]. Apart from the curing mode, the TBS of composite cements may also depend on how the ceramic surface is pretreated. Besides the strength of the dentin–cement interface also that of the cement–ceramic interface will contribute to the overall strength of the bond of the indirect restoration to the tooth [17,18]. Well accepted and most reliable is the etching of (feldspathic) ceramic with hydrofluoric acid (HF) followed by silanization. This pre-treatment provided the highest shear bond strength for four different cements and appeared also to remain stable six months after cementation [18]. Silanization after etching with HF appeared also to be the decisive step in a study by Filho et al. [19], because HF-etching only, without separate silanization, resulted in a significantly lower TBS. Ikemura et al. [20] developed different experimental ‘multi-purpose’ adhesives with an incorporated silane; they were claimed to bond to various dental materials, including ceramics and metal alloys. The 30 wt.% silane-incorporated formulations bonded more effectively than those without silane [20]. Likewise, a so-called ‘universal’ adhesive with incorporated silane was recently introduced (Scotchbond Universal, 3M ESPE, Seefeld, Germany; SBU); it is claimed to be effective for bonding to both tooth surfaces and ceramics, the latter without the need of an additional and separate silane primer. Although numerous studies reported on luting effectiveness, interactions of different factors involved in adhesive luting and their relative importance remain unclear. Therefore, the purpose of this study was to determine the effect of curing mode on the TBS of composite cements to dentin and to evaluate a new ceramic surface pre-treatment using a silane-incorporated, so-called ‘universal’ adhesive. Therefore, the first hypothesis tested was that there was no significant difference in TBS among the different experimental groups that varied for ‘composite cement’ and ‘curing mode’ (1). In
293
addition, the second hypothesis tested was that the auto-cure temperature (room vs. body temperature) did not influence the TBS (2). Finally, the third hypothesis tested was that the TBS was not influenced by the ceramic surface pre-treatment (3: HF-etching followed by silanization and application of an unfilled adhesive, vs. HF-etching followed by application of a silane-incorporated adhesive).
2.
Materials and methods
Dentin surfaces of 88 human 3rd molars were prepared as described by De Munck et al. [21] and randomly assigned to one of the 15 experimental groups. For the luting procedure, three different ‘self-etch’ composite cements were used: Nexus 3 combined with Optibond XTR (XTR/NX3; Kerr, Orange, USA), RelyX Ultimate with Scotchbond Universal (SBU/RXU; 3M ESPE) and Panavia F2.0 with ED Primer II (ED/PAF; Kuraray Noritake, Tokyo, Japan) (Table 1).
2.1.
Ceramic surface pre-treatment (Table 2)
Feldspathic ceramic blocks (10.3 mm × 9 mm with a thickness of 3 mm; Vitablocs Mark II for CEREC/inLab, Vita, Bad Säckingen, Germany) were pre-treated with hydrofluoric acid (‘HF’: IPS Ceramic etching gel, Ivoclar-Vivadent, Schaan, Liechtenstein) for 60 s. Next, two different ceramic pre-treatment protocols were applied. For the XTR/NX3 and ED/PAF, a silane primer (‘S’: Monobond Plus, Ivoclar-Vivadent) was applied and left untouched for 60 s, followed by the application of an unfilled bonding agent (‘HB’: Heliobond, Ivoclar-Vivadent) that was not light-cured. This ceramic surface pre-treatment is further being referred to as ‘HF/S/HB’. For SBU/RXU, the ceramic surface of each block was divided into two parts (5.15 mm × 9 mm) by cutting a shallow, approximately 1 mm deep groove using a diamond saw. A razorblade was inserted into the groove in order to separate the two surfaces from each other and to allow different surface pretreatments to be applied to each part. Then, one side of the surface was treated with the silane primer and adhesive resin following the protocol described above and being referred to as ‘HF/S/HB’, while the other side received the silane-incorporated adhesive Scotchbond Universal (3M ESPE) only, thus without the application of a separate silane primer. The latter ceramic surface pre-treatment is further being referred to as ‘HF/SBU’. In order to directly compare the two different pre-treatment methods (‘HF/S/HB’ vs. ‘HF/SBU’) per tooth, each ceramic block of the SBU/RXU group received two treatments (separated by the groove and temporarily by the razorblade) and was then luted to one tooth.
2.2.
Curing modes (Table 3)
The ceramic blocks were luted onto the dentin surfaces following four different curing modes. Following curing mode ‘AA’, the adhesive applied on dentin was not light-cured after application, but air-thinned according to the manufacturer’s instructions. The ceramic blocks were luted using the composite cements under a constant seating force of 1 kg for 1 min for
294
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
Table 1 – List of all materials used and their application instructions (following the respective manufacturer). Adhesive/composite cement
Manufacturer
Optibond XTR (XTR)
Kerr, Orange, USA
Nexus third generation (NX3)
Kerr
ED Primer II (ED)
Manufacturer’s instructions
Lot number
PRIMER: Application to enamel and dentin for 20 s using a scrubbing motion; air-thin for 5 s. ADHESIVE: Shake; apply for 15 s using light brushing motion; air-thin for 5 s before light curing; light-cure for 10 s or air-thin for 15 s (maximum air-pressure) when left to auto-cure. Apply auto-mixed cement and cure following the respective curing mode: auto-cure for minimum 2–3 min; light-cure for 20 s.
PRIMER: 3611504 ADHESIVE: 3594445
Panavia F2.0 (PAF)
Kuraray Noritake, Tokyo, Japan Kuraray Noritake
Mix one drop of liquid A and B (use within 5 min); apply and leave for 30 s; gently air-dry. Remove cement from the fridge 15 min before use; use paste within 3 min after mixing; use equal amounts of paste (half turn); mix for 20 s; apply on ceramic; apply Oxyguard II onto the margins and auto-cure for minimum 3 min; light-cure for 20 s.
Liquid A: 00299A Liquid B: 00173A Paste A: 00488A Paste B: 00248A
Oxyguard II
Kuraray Noritake
Auto-cure Panavia F2.0 by applying Oxyguard onto the margins and wait for 3 min.
00643A
Scotchbond Universal (SBU)
3M ESPE, Seefeld, Germany
451145
RelyX Ultimate (RXU)
3M ESPE
SELF-ETCH MODE: Do not overdry dentin (dry with cotton pellet); rub for 20 s; blow with gentle air for 5 s; light-cure for 10 s. Apply auto-mixed cement on pre-treated ceramic; auto-cure for minimum 3 min; light-cure for 20 s or wait for 6 min (auto-cure).
Shade: Yellow 3653758, 3551229
TR447059
Table 2 – Ceramic and materials for ceramic surface pre-treatment. Material
Manufacturer
Vitablocs Mark II for CEREC
VITA, Bad Säckingen, Germany
IPS Ceramic Etching Gel
Ivoclar Vivadent
Monobond Plus
Heliobond (HB)
Ivoclar Vivadent, Schaan, Liechtenstein Ivoclar Vivadent
Scotchbond Universal (SBU)
3M ESPE
Manufacturer’s instructions Clean the inner surfaces with alcohol; etch for 60 s with hydrofluoric acid (5%); rinse for 60 s with water spray; apply silane primer and adhesive (see below). Etch the restoration’s inner surfaces for 60 s; rinse for 60 s. Apply a thin coat of the material with a micro-brush; allow to react for 60 s; remove excess by strong air-blowing. Apply a thin layer onto the ceramic surface; air-thin. Apply the adhesive to the entire surface of the restoration to be cemented and allow it to react for 20 s, air-thin for 5 s.
groups which included light-curing of the composite cement, or for 7 min for auto-curing at room temeprature (21 ◦ C) in full darkness. Afterwards, the teeth were stored for 7 days in water (37 ◦ C) in an incubator, until being further processed and subjected to micro-tensile bond strength testing (TBS). Following curing mode ‘AA*’, the more severe ‘AA’ protocol was slightly
LOT 1M2C I10: 12601
P25429 P20536
N75960 451145
changed in order to resemble better the clinical situation. The teeth were warmed up to 37 ◦ C before cementation and the specimens were left in darkness to auto-cure for 10 min at 37 ◦ C. Then, the teeth were stored in pre-warmed water (37 ◦ C), and further processed as described for ‘AA’. Following curing mode ‘LA’, the respective adhesive was light-cured for 20 s
Table 3 – Overview of the different curing modes employed. Curing mode
Adhesive
AA AA* LA LL ALa
No cure No cure Light-cure Light-cure No cure
a
The curing mode ‘AL’ was only applied for ED/PAF.
Composite cement Auto-cure for 7 min at 21 ◦ C (room temperature) Auto-cure for 10 min at 37 ◦ C, tooth pre-warmed (37 ◦ C) Auto-cure for 7 min at 21 ◦ C (room temperature) Light-cure Light-cure
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
using a polywave LED light-curing unit (Bluephase, IvoclarVivadent) with a light intensity above 1000 mW/cm2 (‘high power’ mode); the latter was checked regularly throughout the procedure. The cement was allowed to auto-cure in full darkness for 7 min at 21 ◦ C. For ED/PAF, this curing mode was not applicable and therefore omitted. Following curing mode ‘LL’, the respective adhesive was light-cured for 20 s, as well as the cement was light-cured from each side and from the top surface for 20 s (total curing time of 100 s). For ED/PAF, also this protocol was not applicable as the primer cannot be light-cured; it was dried only. The cement was nevertheless light-cured (this adapted curing mode is being referred to ‘AL’). For ED/PAF, Oxyguard II (Kuraray Noritake) was applied onto the margin between the ceramic specimen and the dentin surface for both auto-cure modes AA and AA*.
2.3.
TBS testing
After storage in water for 7 days (37 ◦ C), the specimens were sectioned in x and y direction using a semi-automatic high-precision diamond saw (Accutom 50, Struers, Ballerup, Denmark) in order to obtain TBS-specimens with a crosssectional area of approximately 1 mm2 . Only sticks from the central part of the dentin surface were used to avoid regional variability. The TBS was measured with a crosshead speed of 1.0 mm/min, as described by De Munck et al. [22].
2.4.
Statistical analysis
To statistically assess the dentin-bond strength data, oneway ANOVA with the different groups as factors and Tukey multiple comparisons were first performed. Additional twoway ANOVA and Tukey multiple comparisons were used to evaluate the effect of the curing mode ‘AA*’ versus that of ‘AA’, and the effect of the new ceramic pre-treatment involving the application of the silane-incorporated adhesive (‘HF/SBU’) versus the conventional ceramic surface pretreatment ‘HF/S/HB’ (R 2.13.1, R Foundation for Statistical Computing, Vienna, Austria). All tests were performed at a significance level of ˛ = 0.05. Sticks that fractured during sectioning (pre-testing failures, ptf’s) were included as zero MPa into the statistical analyses.
2.5.
Failure analysis and SEM
After testing, the failure mode was determined with a stereomicroscope (Wild M5A, Heerbrugg, Switzerland) at 25/50-fold magnification and recorded as ‘cohesive in dentin’, ‘adhesive at the interface dentin–cement’, ‘cohesive in cement’, ‘adhesive at the interface cement–ceramic, ‘cohesive in ceramic’, or ‘mixed’. Representative specimens with characteristic failure modes were also examined by means of scanning electron microscopy (JEOL JSM-6610/6610LV, Tokyo, Japan) using common standard processing techniques including fixation, dehydration, chemical drying, and gold-sputter coating.
3.
Results
3.1.
Composite cement
295
Regarding the TBS of the different composite cements to dentin following the different curing modes, one-way ANOVA and Tukey multiple comparisons revealed significant differences in TBS between the different groups (p < 0.001; Fig. 1). Overall, the highest TBS was measured for XTR/NX3 and SBU/RXU, which was in general significantly higher than the TBS recorded for ED/PAF.
3.2.
Curing mode
Complete auto-cure (‘AA’) revealed the significantly lowest TBS for all composite cements, approximating a 100% pretesting failure (ptf) occurrence with the exception of ED/PAF (Fig. 1). For the curing mode ‘AA*’, two-way ANOVA showed a significant influence of the factors ‘composite cement’ (p < 0.05) and ‘curing mode’ (p < 0.001). The higher auto-curing temperature (‘AA*’) significantly increased the TBS of only ED/PAF (vs. that measured for ‘AA’; p < 0.001), but nevertheless remained below the TBS of all the other groups that underwent a protocol that included at least light-curing of the adhesive (‘LA’ and ‘LL’) (Fig. 1). The significantly highest TBS was recorded for all adhesive/composite cement combinations when at least the adhesive was light-cured (curing modes ‘LA’ and ‘LL’, Fig. 1), except for ED/PAF for which these curing modes could not be applied, and for SBU/RXU when the ceramic received the ‘HF/SBU’ surface pre-treatment (see below). The latter TBS remained nevertheless significantly higher than that of both auto-cure curing modes ‘AA’ and ‘AA*’ (Fig. 1).
3.3.
Ceramic surface pre-treatment
Comparing the TBS for both SBU/RXU groups that received a different ceramic surface pre-treatment (‘HF/S/HB’ vs. ‘HF/SBU’), two-way ANOVA revealed a significant influence of the factors ‘curing mode’ (p < 0.01) and ‘ceramic surface pre-treatment’ (p < 0.001). Equally low bonding (high amount of ptf’s) was recorded for both SBU/RXU groups for the curing modes ‘AA’ and ‘AA*’ (Fig. 1). When the adhesive was light-cured and the cement was left to auto-cure (‘LA’), a significantly higher TBS was measured for the ceramic surface pre-treatment ‘HF/S/HB’, involving separate application of a silane primer, than for ‘HF/SBU’ that incorporated silane within the adhesive formulation (Fig. 1). Following curing mode ‘LL’, no significant difference in TBS was measured for both ceramic surface pre-treatments (Fig. 1).
3.4.
Failure analysis (Fig. 2)
Light-microscopy failure analysis revealed predominant failures at the interface dentin–cement for the auto-cured groups (‘AA’, ‘AA*’), irrespective of the composite cement, though somewhat less distinct for SBU/RXU + HF/S/HB AA and SBU/RXU + HF/SBU AA* (Figs. 3a and b, 4a and b, 5a and b, 6a). More varying failure patterns were observed for XTR/NX3 and
296
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
Fig. 1 – Boxplots showing the data (MPa) of the TBS test. The horizontal line indicates the median value. Vertically, the mean TBS, the standard deviation (SD), and the number of pre-testing failures (ptfs) per total number of specimens tested (n) are mentioned. Statistically non-significantly different groups (Tukey multiple comparisons, p < 0.05) are labeled with the same lowercase letter.
SBU/RXU (both involving the ‘HF/S/HB’ ceramic surface pretreatment) when applied following the curing modes ‘LA’ and LL’ (Figs. 3c and d, 5c and d). Adhesive failure at the interface dentin–cement was the predominant failure pattern for all ED/PAF curing modes (Fig. 4). In contrast to when the ceramic was pre-treated with ‘HF/S/HB’, SBU/RXU specimens, of which the ceramic was pre-treated with ‘HF/SBU’, failed predominantly at the interface cement–ceramic for the curing modes ‘LA’ (Fig. 6c) and ‘LL’.
4.
Discussion
In this study, we used a well-established TBS protocol to investigate two relevant clinical issues regarding the adhesive luting of etchable ceramic restorations. We luted only onto dentin as the main tissue medium-to-large sized restorations are typically bonded to. Besides having investigated the effect of different curing modes, basically comparing
Fig. 2 – Results of the light-microscopy failure analysis for the different experimental groups.
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
297
Fig. 3 – Representative photomicrographs of specimens luted with XTR/NX3 (dentin side) when the ceramic surface received the HF/S/HB pre-treatment for the curing mode ‘AA’ in (a) and the curing mode ‘AA*’ in (b), both revealing an ‘adhesive’ fracture at the interface dentin–cement, or more specifically a fracture at the bottom of the hybrid layer with exposure of dentin tubules; the curing mode ‘LA’ in (c), revealing a ‘mixed’ fracture, with partial exposure of dentin (D) and fracture within the composite cement (C); and the curing mode ‘LL’ in (d), also revealing a ‘mixed’ fracture, with partial exposure of dentin (D), fracture within the adhesive (A)/composite cement (C) and above the composite cement (cement–ceramic interface).
a completely ‘auto-cure’ mode versus ‘light-curing’ solely the adhesive or both the adhesive and composite cement, we also evaluated the effect of a silane primer separately applied on the hydrofluoric-acid etched ceramic surface versus the use of a new silane-incorporated adhesive. The latter was recently introduced as a so-called ‘universal’ (thus also for indirect applications) adhesive with a simplified application procedure. As etchable ceramic, feldspathic ceramic CAD-CAM blocks were selected, knowing that they can be manipulated relatively easily due to their for ceramics rather soft mechanical properties. This for instance helped to prepare micro-specimens in a standardized way,
knowing that the TBS specimen processing is relatively intense. Overall, statistical analysis revealed significant differences between the experimental groups, as well as a significant effect of the experimental variables ‘composite cement’, ‘curing mode’ and ‘ceramic surface pre-treatment’ on the luting effectiveness to dentin. Therefore, the first hypothesis tested must be rejected. Regarding ‘composite cement’, the luting effectiveness of both self-etch composite cements, Nexus 3 (‘XTR/NX3’ and RelyX Ultimate (‘SBU/RXU’), was not significantly different for ‘HF/S/HB’ ceramic surface pretreatment. Both cements
Fig. 4 – Representative photomicrographs of specimens luted with ED/PAF (dentin side) when the ceramic surface received the HF/S/HB pre-treatment for the curing mode ‘AA’ in (a), revealing an ‘adhesive fracture at the interface dentin–cement’, or more specifically a fracture at the bottom of the hybrid layer with exposure of dentin tubules and peritubular dentin (arrow); the curing mode ‘AA*’ in (b), revealing an ‘adhesive’ fracture at the interface dentin–adhesive with remnants of the adhesive still attached to the dentin surface (arrows); and the curing mode ‘AL’ in (c), revealing a fracture at the interface dentin–adhesive, similar to that observed for the curing mode ‘AA*’; the tubules were partly obstructed by the adhesive.
298
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
Fig. 5 – Representative photomicrographs of specimens luted with SBU/RXU (dentin side) when the ceramic surface received the HF/S/HB pre-treatment for the curing mode ‘AA’ in (a), revealing a ceramic surface partly covered with composite cement, also revealing voids most likely due to water inflow, while the fracture occurred at the interface dentin–cement; the curing mode ‘AA*’ in (b), revealing a ceramic surface covered with composite cement, but exposing less voids than observed for ‘AA’; the curing mode ‘LA’ in (c), revealing a ‘mixed’ fracture with the dentin surface (D) partly exposed, while the fracture also occurred within the composite cement (C) and on top of it (arrows); and the curing mode ‘LL’ in (d), revealing a ‘mixed’ fracture with dentin (D) partly covered with adhesive (A), while the fracture also occurred within the composite cement (C) and on top of it.
achieved a much and significantly higher TBS to dentin than the comparable luting protocol of the self-etch composite cement Panavia F2.0 (‘ED/PAF’, ‘AL’ curing mode). This might be attributed to the fact that ED Primer cannot not be lightcured separately as it acts as an initiator, while this step appeared essential for both other adhesive/cement combinations ‘XTR/NX3’ and ‘SBU/RXU’. Regarding ‘curing mode’, the significantly lowest TBSs were measured for both the auto-cure modes ‘AA’ and ‘AA*’. From literature, it is known that the degree of conversion (DC) of auto/self-cured dual-curing composite cements was lower than when they were light-cured [9,23]. The real-time polymerization profile during auto-cure differs from that during dual-cure; the maximum rate of polymerization (polymerization per sec, Rpmax ) was significantly lower for the auto-cure mode [9]. Likewise, the curing speed of dual-cured composite cements was measured to be 5–20 times slower during autocure than light-cure [24]. In our study, the curing mode ‘AA’ involved a rather severe protocol, since the luted specimens were after 7 min auto-cure at room temperature immediately immersed in water. The TBSs measured for all ‘AA’ groups were very low and resulted in numerous pre-testing failures (pfts). Most likely, this can be explained by the slower auto-cure polymerization of the composite cements [9–11,25]. As a consequence, water must have hampered polymerization or at least softened the mechanical properties, by
which either pre-testing failures occurred or a relatively low TBS was measured. This was confirmed by the SEM failure analysis, having disclosed interfacial porosities especially for SBU/RXU when applied following the curing mode ‘AA’ (Figs. 5a and 6a). When the composite cements were applied following the adapted auto-cure mode ‘AA*’, the TBS of both SBU/RXU experimental groups (for both ‘HF/S/HB’ and ‘HF/SBU’) was slightly higher than when the cements underwent the ‘AA’ curing mode, and less pre-testing failures occurred. No difference was measured between ‘AA’ and ‘AA*’ for XTR/NX3. Only for ED/PAF, a significant increase in TBS was measured, by which hypothesis 2 was solely rejected for ED/PAF. The curing mode ‘AA*’ involved an adapted, less severe auto-cure protocol with an extended polymerization time (10 min vs. 7 min), and an increased temperature (37 ◦ C vs. 21 ◦ C). Hence, a higher degree of conversion, lower water sorption and higher mechanical properties must have been reached. The significantly higher TBS of ED/PAF measured for ‘AA*’ versus ‘AA’ in this study can be explained by its Rpmax (see above) having been reported to be similar during auto/self-cure and light-cure [10]. Compared to two other dual-cure composite cements, the Rpmax of Panavia F2.0 was significantly higher [10]. Likewise, this finding might also explain the significantly higher TBS of ED/PAF found in our study, when compared to XTR/NX3 and SBU/RXU following the curing mode ‘AA*’.
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
299
Fig. 6 – Representative photomicrographs of specimens bonded with SBU/RXU (dentin side) when the ceramic surface received the HF/SBU pre-treatment for the curing mode ‘AA’ in (a), revealing a ceramic surface covered with composite cement, and showing voids most likely due to water inflow, while the fracture occurred at the interface dentin–cement; the curing mode ‘AA*’ in (b), revealing a ‘mixed’ fracture with exposed dentin (D), partly covered by the adhesive (arrow), while the fracture also occurred on top of the composite cement (C); the curing mode ‘LA’ in (c), revealing the ceramic surface with the fracture having occurred ‘adhesively’ at the interface cement–ceramic; and the curing mode ‘LL’ in (d), revealing a similar ‘mixed’ fracture pattern as in ‘AA*’ with exposed dentin (D), partly covered by the adhesive (A), while the fracture occurred also on top of the composite cement (C).
Overall, auto-cure leads to a lower TBS when compared to dual-cure, as measured in this and other studies [13,14]. In contrast to the adhesives Optibond XTR and Scotchbond Universal, tested in our study, the ‘Liquid B’ of ED Primer II contains T-isoprolic benzenic sodium sulfinate [25], which acts as a co-initiator when it comes in contact with the subsequently applied composite cement Panavia F2.0. This additional coinitiator may also have contributed to the significantly higher TBS measured for ED/PAF, when applied following the curing mode ‘AA*’ and ‘AL’. The latter curing mode, applied solely for ED/PAF, nevertheless resulted in a significantly lower TBS than for all other cements when they were light-cured (curing mode ‘LL’). Faria-e-Silva et al. [25] found that ED Primer II increased Rpmax when Panavia F2.0 was dual-cured. When the adhesive was light-cured and the cement was left to auto-cure following curing mode ‘LA’, no significant differences in TBS were recorded for XTR/NX3 and SBU/RXU, when the ceramic block was pre-treated following the ‘HF/S/HB’ ceramic surface pre-treatment. In our study, light-curing adhesives were used with dual-curing composite cements. Our data clearly showed that while using a lightcuring adhesive in an auto-curing set-up, the adhesive has to be light-cured in order to obtain the most favorable ‘immediate’ bond strength to dentin. By separately light-curing the adhesive at the dentin side, the hybrid and adhesive layer is thought to be stabilized before the composite cement is
subsequently applied, thereby also immediately sealing dentin and thus preventing water uptake from the dentin host through osmosis [26]. Self-evidently, this can only be done when the adhesive has a sufficiently low film thickness (after air-thinning), so that the restoration fit is not impaired. The film thickness of the adhesives Optibond XTR and Scotchbond Universal was previously found to be air-thinned below 10 m (unpublished TEM interfacial analysis). By separately lightcuring the adhesive, no further significant increase in TBS was measured when also the composite cement was cured afterwards following the curing mode ‘LL’. In comparison with activator-initiated self-curing (auto-cure), separate lightcuring of an adhesive without activator was shown to have led to a significantly higher DC [27]. Also, pre-(light-)curing of self-etch adhesives resulted in a higher bond strength than co-(light-)curing the adhesive with the composite cement through a ceramic specimen [28], this especially when the thickness of the ceramic block exceeded 2 mm [2]. Most likely, the improved TBS recorded for the curing modes ‘LA’ and ‘LL’ can be attributed to the higher DC of the adhesive and composite cement reached after light-curing. It must also have been reached more rapidly, thereby having lowered water sorption from dentin and prevented the mechanical properties to decrease, which was confirmed by the fracture analysis. When auto-cured (‘AA’; ‘AA*’), the predominant fracture mode was ‘adhesive’ at the interface dentin–cement. It was associated
300
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
with numerous pre-testing failures (ptf’s). After light-curing of the adhesive (‘LA’), a more diverse fracture pattern, including fractures within the composite cement, at the interface cement–ceramic, and within the ceramic occurred. Overall, light-curing of both adhesive and composite cement (curing mode ‘LL’) led to the highest TBS. Lightcuring of dual-curing composite cements was indeed shown to improve their mechanical properties and DC [9–11], and self-evidently must have contributed to the increased TBS measured in our study. Our results are in agreement with several studies, in which the highest TBS was measured for those composite cements that were light-cured [10,13,29]. As pre-(light-)curing of the adhesive leads to a significant increase of the TBS, this method must be applied in particular for thick restorations, which hamper light transmission. In the latter case, the composite cement is cured less efficiently with inferior bond strength and mechanical properties [3,4,30,31]. Pre-(light-)curing was also shown to have led to an increase in ‘continuous’ margins and bond strength of adhesively luted ceramic inlays [32,33]. Regarding the experimental variable ‘ceramic surface pre-treatment’, the ‘new’ ceramic pre-treatment ‘HF/SBU’, involving the application of a silane-incorporated adhesive (Scotchbond Universal) resulted in a significantly lower TBS when the curing mode ‘LA’ was applied, by which hypothesis 3 was rejected. Nevertheless, no difference in TBS was found when SBU/RXU was applied following the curing mode ‘LL’. The incorporation of silane into the adhesive is a promising approach, because it reduces the amount of working steps during the cementation of indirect restorations and allows the practitioner to use the same adhesive for the dental hard tissues and the ceramic. The fracture analysis indicated that the lower bond strength measured for ‘HF/SBU’ versus ‘HF/S/HB’ for SBU/RXU must be attributed to the ceramic pre-treatment, because the specimens in the ‘HF/SBU’ group mainly failed at the interface cement–ceramic. While following curing mode ‘LL’, no difference in TBS was detected between ‘HF/SBU’ and ‘HF/S/HB’ for SBU/RXU, still differences observed in the fracture pattern of both ceramic surface pre-treatments indicated that the weak link for the ‘new’ silane-incorporated adhesive approach was the interface cement–ceramic; for the ‘established’ separately applied silane primer, the specimens failed more at the interface dentin–cement.
5.
Conclusion
Complete auto-cure of adhesively luted ceramic restorations onto dentin, independent of the curing temperature (21 or 37 ◦ C) resulted in inferior ‘immediate’ TBS to dentin than when solely the adhesive at the dentin side was separately light-cured or both the adhesive and the composite cement were light-cured. When the adhesive is separately light-cured at the dentin side, special caution is needed to not impair the restoration fit by adequately air-thin the adhesive before light-curing. The new ceramic surface pre-treatment with a silaneincorporated adhesive, thereby simplifying the already complex adhesive luting procedure, affected the ‘immediate’ TBS to dentin only when the composite cement was not
light-cured and allowed to auto-cure. Thus, while this second session of light-curing when adhesively luting indirect restorations is highly recommended, today this is clinically done in most cases.
Acknowledgements This investigation was supported in part by the EFCD Scientific Foundation. All the materials used in this study were provided by the respective manufacturers.
references
[1] Burke FJ, Fleming GJ, Nathanson D, Marquis PM. Are adhesive technologies needed to support ceramics? An assessment of the current evidence. J Adhes Dent 2002;4:7–22. [2] Rathke A, Hokenmaier G, Muche R, Haller B. Effectiveness of the bond established between ceramic inlays and dentin using different luting protocols. J Adhes Dent 2012;14:147–54. [3] Soares CJ, da Silva NR, Fonseca RB. Influence of the feldspathic ceramic thickness and shade on the microhardness of dual resin cement. Oper Dent 2006;31:384–9. [4] Kilinc E, Antonson SA, Hardigan PC, Kesercioglu A. The effect of ceramic restoration shade and thickness on the polymerization of light- and dual-cure resin cements. Oper Dent 2011;36:661–9. [5] el-Mowafy OM, Rubo MH, el-Badrawy WA. Hardening of new resin cements cured through a ceramic inlay. Oper Dent 1999;24:38–44. [6] Akgungor G, Akkayan B, Gaucher H. Influence of ceramic thickness and polymerization mode of a resin luting agent on early bond strength and durability with a lithium disilicate-based ceramic system. J Prosthet Dent 2005;94:234–41. [7] Tashiro H, Inai N, Nikaido T, Tagami J. Effects of light intensity through resin inlays on the bond strength of dual-cured resin cement. J Adhes Dent 2004;6:233–8. [8] de Menezes MJ, Arrais CA, Giannini M. Influence of light-activated and auto- and dual-polymerizing adhesive systems on bond strength of indirect composite resin to dentin. J Prosthet Dent 2006;96:115–21. [9] Arrais CA, Giannini M, Rueggeberg FA. Kinetic analysis of monomer conversion in auto- and dual-polymerizing modes of commercial resin luting cements. J Prosthet Dent 2009;101:128–36. [10] Faria-e-Silva A, Boaro L, Braga R, Piva E, Arias V, Martins L. Effect of immediate or delayed light activation on curing kinetics and shrinkage stress of dual-cure resin cements. Oper Dent 2011;36:196–204. [11] Hofmann N, Papsthart G, Hugo B, Klaiber B. Comparison of photo-activation versus chemical or dual-curing of resin-based luting cements regarding flexural strength, modulus and surface hardness. J Oral Rehabil 2001;28:1022–8. [12] Aguiar TR, Di Francescantonio M, Ambrosano GM, Giannini M. Effect of curing mode on bond strength of self-adhesive resin luting cements to dentin. J Biomed Mater Res B: Appl Biomater 2010;93:122–7. [13] Arrais CA, Giannini M, Rueggeberg FA, Pashley DH. Microtensile bond strength of dual-polymerizing cementing systems to dentin using different polymerizing modes. J Prosthet Dent 2007;97:99–106.
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
[14] Faria-e-Silva AL, Fabião MM, Arias VG, Martins LR. Activation mode effects on the shear bond strength of dual-cured resin cements. Oper Dent 2010;35:515–21. [15] Garcia RN, Reis AF, Giannini M. Effect of activation mode of dual-cured resin cements and low-viscosity composite liners on bond strength to dentin. J Dent 2007;35:564–9. [16] Takahashi R, Nikaido T, Ariyoshi M, Foxton RM, Tagami J. Microtensile bond strengths of a dual-cure resin cement to dentin resin-coated with an all-in-one adhesive system using two curing modes. Dent Mater J 2010;29:268–76. [17] Foxton RM, Pereira PN, Nakajima M, Tagami J, Miura H. Durability of the dual-cure resin cement/ceramic bond with different curing strategies. J Adhes Dent 2002;4:49–59. [18] Stewart GP, Jain P, Hodges J. Shear bond strength of resin cements to both ceramic and dentin. J Prosthet Dent 2002;88:277–84. [19] Filho AM, Vieira LC, Araújo E, Monteiro Júnior S. Effect of different ceramic surface treatments on resin microtensile bond strength. J Prosthodont 2004;3:28–35. [20] Ikemura K, Tanaka H, Fujii T, Deguchi M, Negoro N, Endo T, et al. Design of a new, multi-purpose, light-curing adhesive comprising a silane coupling agent, acidic adhesive monomers and dithiooctanoate monomers for bonding to varied metal and dental ceramic materials. Dent Mater J 2011;30:493–500. [21] De Munck J, Vargas M, Iracki J, Van Landuyt K, Poitevin A, Lambrechts P, et al. One-day bonding effectiveness of new self-etch adhesives to bur-cut enamel and dentin. Oper Dent 2005;30:39–49. [22] De Munck J, Van den Steen PE, Mine A, Van Landuyt KL, Poitevin A, Opdenakker G, et al. Inhibition of enzymatic degradation of adhesive–dentin interfaces. J Dent Res 2009;88:1101–6. [23] Arrais CA, Rueggeberg FA, Waller JL, de Goes MF, Giannini M. Effect of curing mode on the polymerization characteristics
[24]
[25] [26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
301
of dual-cured resin cement systems. J Dent 2008;36: 418–26. Lee IB, Um CM. Thermal analysis on the cure speed of dual cured resin cements under porcelain inlays. J Oral Rehabil 2001;28:186–97. Faria-e-Silva AL, Moraes RR, Ogliari FA, Piva E, Martins LR. Panavia F: the role of the primer. J Oral Sci 2009;51:255–9. Van Landuyt KL, Snauwaert J, De Munck J, Coutinho E, Poitevin A, Yoshida Y, et al. Origin of interfacial droplets with one-step adhesives. J Dent Res 2007;86:739–44. Faria-e-Silva AL, Casselli DS, Lima GS, Ogliari FA, Piva E, Martins LR. Kinetics of conversion of two dual-cured adhesive systems. J Endod 2008;34:1115–8. Chapman JL, Burgess JO, Holst S, Sadan A, Blatz MB. Precuring of self-etching bonding agents and its effect on bond strength of resin composite to dentin and enamel. Quintessence Int 2007;38:637–41. Rathke A, Balz U, Muche R, Haller B. Effects of self-curing activator and curing protocol on the bond strength of composite core buildups. J Adhes Dent 2012;14:39–46. Sinhoreti MA, Manetta IP, Tango RN, Iriyama NT, Consani RL, Correr-Sobrinho L. Effect of light-curing methods on resin cement Knoop hardness at different depths. Braz Dent J 2007;18:305–8. de Paula AB, Tango RN, Sinhoreti MA, Alves MC, Puppin-Rontani RM. Effect of thickness of indirect restoration and distance from the light-curing unit tip on the hardness of a dual-cured resin cement. Braz Dent J 2010;21:117–22. Frankenberger R, Sindel J, Krämer N, Petschelt A. Dentin bond strength and marginal adaptation: direct composite resins vs ceramic inlays. Oper Dent 1999;24:147–55. Haller B, Hässner K, Moll K. Marginal adaptation of dentin bonded ceramic inlays: effects of bonding systems and luting resin composites. Oper Dent 2003;28:574–84.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
Composite cements benefit from light-curing Anne-Katrin Lührs a,b,∗ , Jan De Munck a , Werner Geurtsen b , Bart Van Meerbeek a a
KU Leuven – BIOMAT, Department of Oral Health Sciences, KU Leuven (University of Leuven) & Dentistry, University Hospitals Leuven, Belgium b Department of Conservative Dentistry, Periodontology and Preventive Dentistry, Hannover Medical School, Hannover, Germany
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. To investigate the effect of curing of composite cements and a new ceramic
Received 8 August 2013
silanization pre-treatment on the micro-tensile bond strength (TBS).
Received in revised form
Methods. Feldspathic ceramic blocks were luted onto dentin using either Optibond
25 November 2013
XTR/Nexus 3 (XTR/NX3; Kerr), the silane-incorporated ‘universal’ adhesive Scotchbond Uni-
Accepted 28 November 2013
versal/RelyX Ultimate (SBU/RXU; 3M ESPE), or ED Primer II/Panavia F2.0 (ED/PAF; Kuraray Noritake). Besides ‘composite cement’, experimental variables were ‘curing mode’ (‘AA’: complete auto-cure at 21 ◦ C; ‘AA*’: complete auto-cure at 37 ◦ C; ‘LA’: light-curing of adhesive
Keywords:
and auto-cure of cement; ‘LL’: complete light-curing) and ‘ceramic surface pre-treatment’
Composite cement
(‘HF/S/HB’: hydrofluoric acid (‘HF’: IPS Ceramic Etching Gel, Ivoclar-Vivadent), silaniza-
Adhesive
tion (‘S’: Monobond Plus, Ivoclar-Vivadent) and application of an adhesive resin (‘HB’:
Curing
Heliobond, Ivoclar-Vivadent); ‘HF/SBU’: ‘HF’ and application of the ‘universal’ adhesive
Ceramic
Scotchbond Universal (‘SBU’; 3M ESPE, only for SBU/RXU)). After water storage (7 days at
Silane
37 ◦ C), ceramic–dentin sticks were subjected to TBS testing.
Universal adhesive
Results. Regarding the ‘composite cement’, the significantly lowest TBSs were measured for ED/PAF. Regarding ‘curing mode’, the significantly highest TBS was recorded when at least the adhesive was light-cured (‘LA’ and ‘LL’). Complete auto-cure (‘AA’) revealed the significantly lowest TBS. The higher auto-curing temperature (‘AA*’) increased the TBS only for ED/PAF. Regarding ‘ceramic surface pre-treatment’, only for ‘LA’ the TBS was significantly higher for ‘HF/S/HB’ than for ‘HF/SBU’. Significance. Complete auto-cure led to inferior TBS than when either the adhesive (on dentin) or both adhesive and composite cement were light-cured. The use of a silaneincorporated adhesive did not decrease luting effectiveness when also the composite cement was light-cured. © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Indirect ceramic restorations became more popular during the last years because of their superior esthetics,
∗
biocompatibility and long-term stability. So-called ‘etchable’ feldspathic (and glass-) ceramic restorations should be adhesively luted; it improves the fracture resistance and consequently enhances the survival rate [1]. Possible curing modes of composite cements are ‘dual-cure’, involving also
Corresponding author at: KU Leuven – BIOMAT, Department of Oral Health Sciences, KU Leuven (University of Leuven) & Dentistry, University Hospitals Leuven, Belgium/Department of Conservative Dentistry, Periodontology and Preventive Dentistry, Hannover Medical School, OE 7740, Carl-Neuberg-Straße 1, 30625 Hannover, Germany. Tel.: +49 532 4718; fax: +49 532 4811. E-mail address: [email protected] (A.-K. Lührs). 0109-5641/$ – see front matter © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dental.2013.11.012
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
light-curing, and ‘auto-cure’. The beforehand applied adhesive, if any (not in case of self-adhesive composite cements), can be separately light-cured on the condition that its film thickness is sufficiently thin and does not impair the restoration fit. When adhesively luted ceramic restorations are light-cured, both the adhesive and cement may however not cure completely due to light attenuation, caused by the opacity of the restoration, its shade, and/or simply its thickness, the latter obviously related to cavity depth [2,3]. A ceramic thickness of 2–3 mm is considered to be the threshold to still effectively light-cure adhesively luted ceramic restorations [4,5]. Up to a thickness of 2 mm, there was according to Akgungor et al. [6] no effect of both ceramic thickness and polymerization mode on bond strength. For thicker restorations of 4 mm, a lower micro-tensile bond strength (TBS) was detected [2]. Light attenuation already starts at a composite thickness of 1 mm, as the light intensity was reduced about 85%, and restorations of 4 mm thickness blocked the light almost completely [7]. It was also reported that lightcuring the adhesive and the composite cement separately improved the bond strength to dentin [2,8]. Furthermore, autocuring was found to influence the degree of conversion (DC) of composite cements and their mechanical properties [9–11]. Depending on the material, the bond strength of composite cements that were allowed to auto-cure was inferior to when they were light-cured [12–16]. Apart from the curing mode, the TBS of composite cements may also depend on how the ceramic surface is pretreated. Besides the strength of the dentin–cement interface also that of the cement–ceramic interface will contribute to the overall strength of the bond of the indirect restoration to the tooth [17,18]. Well accepted and most reliable is the etching of (feldspathic) ceramic with hydrofluoric acid (HF) followed by silanization. This pre-treatment provided the highest shear bond strength for four different cements and appeared also to remain stable six months after cementation [18]. Silanization after etching with HF appeared also to be the decisive step in a study by Filho et al. [19], because HF-etching only, without separate silanization, resulted in a significantly lower TBS. Ikemura et al. [20] developed different experimental ‘multi-purpose’ adhesives with an incorporated silane; they were claimed to bond to various dental materials, including ceramics and metal alloys. The 30 wt.% silane-incorporated formulations bonded more effectively than those without silane [20]. Likewise, a so-called ‘universal’ adhesive with incorporated silane was recently introduced (Scotchbond Universal, 3M ESPE, Seefeld, Germany; SBU); it is claimed to be effective for bonding to both tooth surfaces and ceramics, the latter without the need of an additional and separate silane primer. Although numerous studies reported on luting effectiveness, interactions of different factors involved in adhesive luting and their relative importance remain unclear. Therefore, the purpose of this study was to determine the effect of curing mode on the TBS of composite cements to dentin and to evaluate a new ceramic surface pre-treatment using a silane-incorporated, so-called ‘universal’ adhesive. Therefore, the first hypothesis tested was that there was no significant difference in TBS among the different experimental groups that varied for ‘composite cement’ and ‘curing mode’ (1). In
293
addition, the second hypothesis tested was that the auto-cure temperature (room vs. body temperature) did not influence the TBS (2). Finally, the third hypothesis tested was that the TBS was not influenced by the ceramic surface pre-treatment (3: HF-etching followed by silanization and application of an unfilled adhesive, vs. HF-etching followed by application of a silane-incorporated adhesive).
2.
Materials and methods
Dentin surfaces of 88 human 3rd molars were prepared as described by De Munck et al. [21] and randomly assigned to one of the 15 experimental groups. For the luting procedure, three different ‘self-etch’ composite cements were used: Nexus 3 combined with Optibond XTR (XTR/NX3; Kerr, Orange, USA), RelyX Ultimate with Scotchbond Universal (SBU/RXU; 3M ESPE) and Panavia F2.0 with ED Primer II (ED/PAF; Kuraray Noritake, Tokyo, Japan) (Table 1).
2.1.
Ceramic surface pre-treatment (Table 2)
Feldspathic ceramic blocks (10.3 mm × 9 mm with a thickness of 3 mm; Vitablocs Mark II for CEREC/inLab, Vita, Bad Säckingen, Germany) were pre-treated with hydrofluoric acid (‘HF’: IPS Ceramic etching gel, Ivoclar-Vivadent, Schaan, Liechtenstein) for 60 s. Next, two different ceramic pre-treatment protocols were applied. For the XTR/NX3 and ED/PAF, a silane primer (‘S’: Monobond Plus, Ivoclar-Vivadent) was applied and left untouched for 60 s, followed by the application of an unfilled bonding agent (‘HB’: Heliobond, Ivoclar-Vivadent) that was not light-cured. This ceramic surface pre-treatment is further being referred to as ‘HF/S/HB’. For SBU/RXU, the ceramic surface of each block was divided into two parts (5.15 mm × 9 mm) by cutting a shallow, approximately 1 mm deep groove using a diamond saw. A razorblade was inserted into the groove in order to separate the two surfaces from each other and to allow different surface pretreatments to be applied to each part. Then, one side of the surface was treated with the silane primer and adhesive resin following the protocol described above and being referred to as ‘HF/S/HB’, while the other side received the silane-incorporated adhesive Scotchbond Universal (3M ESPE) only, thus without the application of a separate silane primer. The latter ceramic surface pre-treatment is further being referred to as ‘HF/SBU’. In order to directly compare the two different pre-treatment methods (‘HF/S/HB’ vs. ‘HF/SBU’) per tooth, each ceramic block of the SBU/RXU group received two treatments (separated by the groove and temporarily by the razorblade) and was then luted to one tooth.
2.2.
Curing modes (Table 3)
The ceramic blocks were luted onto the dentin surfaces following four different curing modes. Following curing mode ‘AA’, the adhesive applied on dentin was not light-cured after application, but air-thinned according to the manufacturer’s instructions. The ceramic blocks were luted using the composite cements under a constant seating force of 1 kg for 1 min for
294
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
Table 1 – List of all materials used and their application instructions (following the respective manufacturer). Adhesive/composite cement
Manufacturer
Optibond XTR (XTR)
Kerr, Orange, USA
Nexus third generation (NX3)
Kerr
ED Primer II (ED)
Manufacturer’s instructions
Lot number
PRIMER: Application to enamel and dentin for 20 s using a scrubbing motion; air-thin for 5 s. ADHESIVE: Shake; apply for 15 s using light brushing motion; air-thin for 5 s before light curing; light-cure for 10 s or air-thin for 15 s (maximum air-pressure) when left to auto-cure. Apply auto-mixed cement and cure following the respective curing mode: auto-cure for minimum 2–3 min; light-cure for 20 s.
PRIMER: 3611504 ADHESIVE: 3594445
Panavia F2.0 (PAF)
Kuraray Noritake, Tokyo, Japan Kuraray Noritake
Mix one drop of liquid A and B (use within 5 min); apply and leave for 30 s; gently air-dry. Remove cement from the fridge 15 min before use; use paste within 3 min after mixing; use equal amounts of paste (half turn); mix for 20 s; apply on ceramic; apply Oxyguard II onto the margins and auto-cure for minimum 3 min; light-cure for 20 s.
Liquid A: 00299A Liquid B: 00173A Paste A: 00488A Paste B: 00248A
Oxyguard II
Kuraray Noritake
Auto-cure Panavia F2.0 by applying Oxyguard onto the margins and wait for 3 min.
00643A
Scotchbond Universal (SBU)
3M ESPE, Seefeld, Germany
451145
RelyX Ultimate (RXU)
3M ESPE
SELF-ETCH MODE: Do not overdry dentin (dry with cotton pellet); rub for 20 s; blow with gentle air for 5 s; light-cure for 10 s. Apply auto-mixed cement on pre-treated ceramic; auto-cure for minimum 3 min; light-cure for 20 s or wait for 6 min (auto-cure).
Shade: Yellow 3653758, 3551229
TR447059
Table 2 – Ceramic and materials for ceramic surface pre-treatment. Material
Manufacturer
Vitablocs Mark II for CEREC
VITA, Bad Säckingen, Germany
IPS Ceramic Etching Gel
Ivoclar Vivadent
Monobond Plus
Heliobond (HB)
Ivoclar Vivadent, Schaan, Liechtenstein Ivoclar Vivadent
Scotchbond Universal (SBU)
3M ESPE
Manufacturer’s instructions Clean the inner surfaces with alcohol; etch for 60 s with hydrofluoric acid (5%); rinse for 60 s with water spray; apply silane primer and adhesive (see below). Etch the restoration’s inner surfaces for 60 s; rinse for 60 s. Apply a thin coat of the material with a micro-brush; allow to react for 60 s; remove excess by strong air-blowing. Apply a thin layer onto the ceramic surface; air-thin. Apply the adhesive to the entire surface of the restoration to be cemented and allow it to react for 20 s, air-thin for 5 s.
groups which included light-curing of the composite cement, or for 7 min for auto-curing at room temeprature (21 ◦ C) in full darkness. Afterwards, the teeth were stored for 7 days in water (37 ◦ C) in an incubator, until being further processed and subjected to micro-tensile bond strength testing (TBS). Following curing mode ‘AA*’, the more severe ‘AA’ protocol was slightly
LOT 1M2C I10: 12601
P25429 P20536
N75960 451145
changed in order to resemble better the clinical situation. The teeth were warmed up to 37 ◦ C before cementation and the specimens were left in darkness to auto-cure for 10 min at 37 ◦ C. Then, the teeth were stored in pre-warmed water (37 ◦ C), and further processed as described for ‘AA’. Following curing mode ‘LA’, the respective adhesive was light-cured for 20 s
Table 3 – Overview of the different curing modes employed. Curing mode
Adhesive
AA AA* LA LL ALa
No cure No cure Light-cure Light-cure No cure
a
The curing mode ‘AL’ was only applied for ED/PAF.
Composite cement Auto-cure for 7 min at 21 ◦ C (room temperature) Auto-cure for 10 min at 37 ◦ C, tooth pre-warmed (37 ◦ C) Auto-cure for 7 min at 21 ◦ C (room temperature) Light-cure Light-cure
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
using a polywave LED light-curing unit (Bluephase, IvoclarVivadent) with a light intensity above 1000 mW/cm2 (‘high power’ mode); the latter was checked regularly throughout the procedure. The cement was allowed to auto-cure in full darkness for 7 min at 21 ◦ C. For ED/PAF, this curing mode was not applicable and therefore omitted. Following curing mode ‘LL’, the respective adhesive was light-cured for 20 s, as well as the cement was light-cured from each side and from the top surface for 20 s (total curing time of 100 s). For ED/PAF, also this protocol was not applicable as the primer cannot be light-cured; it was dried only. The cement was nevertheless light-cured (this adapted curing mode is being referred to ‘AL’). For ED/PAF, Oxyguard II (Kuraray Noritake) was applied onto the margin between the ceramic specimen and the dentin surface for both auto-cure modes AA and AA*.
2.3.
TBS testing
After storage in water for 7 days (37 ◦ C), the specimens were sectioned in x and y direction using a semi-automatic high-precision diamond saw (Accutom 50, Struers, Ballerup, Denmark) in order to obtain TBS-specimens with a crosssectional area of approximately 1 mm2 . Only sticks from the central part of the dentin surface were used to avoid regional variability. The TBS was measured with a crosshead speed of 1.0 mm/min, as described by De Munck et al. [22].
2.4.
Statistical analysis
To statistically assess the dentin-bond strength data, oneway ANOVA with the different groups as factors and Tukey multiple comparisons were first performed. Additional twoway ANOVA and Tukey multiple comparisons were used to evaluate the effect of the curing mode ‘AA*’ versus that of ‘AA’, and the effect of the new ceramic pre-treatment involving the application of the silane-incorporated adhesive (‘HF/SBU’) versus the conventional ceramic surface pretreatment ‘HF/S/HB’ (R 2.13.1, R Foundation for Statistical Computing, Vienna, Austria). All tests were performed at a significance level of ˛ = 0.05. Sticks that fractured during sectioning (pre-testing failures, ptf’s) were included as zero MPa into the statistical analyses.
2.5.
Failure analysis and SEM
After testing, the failure mode was determined with a stereomicroscope (Wild M5A, Heerbrugg, Switzerland) at 25/50-fold magnification and recorded as ‘cohesive in dentin’, ‘adhesive at the interface dentin–cement’, ‘cohesive in cement’, ‘adhesive at the interface cement–ceramic, ‘cohesive in ceramic’, or ‘mixed’. Representative specimens with characteristic failure modes were also examined by means of scanning electron microscopy (JEOL JSM-6610/6610LV, Tokyo, Japan) using common standard processing techniques including fixation, dehydration, chemical drying, and gold-sputter coating.
3.
Results
3.1.
Composite cement
295
Regarding the TBS of the different composite cements to dentin following the different curing modes, one-way ANOVA and Tukey multiple comparisons revealed significant differences in TBS between the different groups (p < 0.001; Fig. 1). Overall, the highest TBS was measured for XTR/NX3 and SBU/RXU, which was in general significantly higher than the TBS recorded for ED/PAF.
3.2.
Curing mode
Complete auto-cure (‘AA’) revealed the significantly lowest TBS for all composite cements, approximating a 100% pretesting failure (ptf) occurrence with the exception of ED/PAF (Fig. 1). For the curing mode ‘AA*’, two-way ANOVA showed a significant influence of the factors ‘composite cement’ (p < 0.05) and ‘curing mode’ (p < 0.001). The higher auto-curing temperature (‘AA*’) significantly increased the TBS of only ED/PAF (vs. that measured for ‘AA’; p < 0.001), but nevertheless remained below the TBS of all the other groups that underwent a protocol that included at least light-curing of the adhesive (‘LA’ and ‘LL’) (Fig. 1). The significantly highest TBS was recorded for all adhesive/composite cement combinations when at least the adhesive was light-cured (curing modes ‘LA’ and ‘LL’, Fig. 1), except for ED/PAF for which these curing modes could not be applied, and for SBU/RXU when the ceramic received the ‘HF/SBU’ surface pre-treatment (see below). The latter TBS remained nevertheless significantly higher than that of both auto-cure curing modes ‘AA’ and ‘AA*’ (Fig. 1).
3.3.
Ceramic surface pre-treatment
Comparing the TBS for both SBU/RXU groups that received a different ceramic surface pre-treatment (‘HF/S/HB’ vs. ‘HF/SBU’), two-way ANOVA revealed a significant influence of the factors ‘curing mode’ (p < 0.01) and ‘ceramic surface pre-treatment’ (p < 0.001). Equally low bonding (high amount of ptf’s) was recorded for both SBU/RXU groups for the curing modes ‘AA’ and ‘AA*’ (Fig. 1). When the adhesive was light-cured and the cement was left to auto-cure (‘LA’), a significantly higher TBS was measured for the ceramic surface pre-treatment ‘HF/S/HB’, involving separate application of a silane primer, than for ‘HF/SBU’ that incorporated silane within the adhesive formulation (Fig. 1). Following curing mode ‘LL’, no significant difference in TBS was measured for both ceramic surface pre-treatments (Fig. 1).
3.4.
Failure analysis (Fig. 2)
Light-microscopy failure analysis revealed predominant failures at the interface dentin–cement for the auto-cured groups (‘AA’, ‘AA*’), irrespective of the composite cement, though somewhat less distinct for SBU/RXU + HF/S/HB AA and SBU/RXU + HF/SBU AA* (Figs. 3a and b, 4a and b, 5a and b, 6a). More varying failure patterns were observed for XTR/NX3 and
296
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
Fig. 1 – Boxplots showing the data (MPa) of the TBS test. The horizontal line indicates the median value. Vertically, the mean TBS, the standard deviation (SD), and the number of pre-testing failures (ptfs) per total number of specimens tested (n) are mentioned. Statistically non-significantly different groups (Tukey multiple comparisons, p < 0.05) are labeled with the same lowercase letter.
SBU/RXU (both involving the ‘HF/S/HB’ ceramic surface pretreatment) when applied following the curing modes ‘LA’ and LL’ (Figs. 3c and d, 5c and d). Adhesive failure at the interface dentin–cement was the predominant failure pattern for all ED/PAF curing modes (Fig. 4). In contrast to when the ceramic was pre-treated with ‘HF/S/HB’, SBU/RXU specimens, of which the ceramic was pre-treated with ‘HF/SBU’, failed predominantly at the interface cement–ceramic for the curing modes ‘LA’ (Fig. 6c) and ‘LL’.
4.
Discussion
In this study, we used a well-established TBS protocol to investigate two relevant clinical issues regarding the adhesive luting of etchable ceramic restorations. We luted only onto dentin as the main tissue medium-to-large sized restorations are typically bonded to. Besides having investigated the effect of different curing modes, basically comparing
Fig. 2 – Results of the light-microscopy failure analysis for the different experimental groups.
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
297
Fig. 3 – Representative photomicrographs of specimens luted with XTR/NX3 (dentin side) when the ceramic surface received the HF/S/HB pre-treatment for the curing mode ‘AA’ in (a) and the curing mode ‘AA*’ in (b), both revealing an ‘adhesive’ fracture at the interface dentin–cement, or more specifically a fracture at the bottom of the hybrid layer with exposure of dentin tubules; the curing mode ‘LA’ in (c), revealing a ‘mixed’ fracture, with partial exposure of dentin (D) and fracture within the composite cement (C); and the curing mode ‘LL’ in (d), also revealing a ‘mixed’ fracture, with partial exposure of dentin (D), fracture within the adhesive (A)/composite cement (C) and above the composite cement (cement–ceramic interface).
a completely ‘auto-cure’ mode versus ‘light-curing’ solely the adhesive or both the adhesive and composite cement, we also evaluated the effect of a silane primer separately applied on the hydrofluoric-acid etched ceramic surface versus the use of a new silane-incorporated adhesive. The latter was recently introduced as a so-called ‘universal’ (thus also for indirect applications) adhesive with a simplified application procedure. As etchable ceramic, feldspathic ceramic CAD-CAM blocks were selected, knowing that they can be manipulated relatively easily due to their for ceramics rather soft mechanical properties. This for instance helped to prepare micro-specimens in a standardized way,
knowing that the TBS specimen processing is relatively intense. Overall, statistical analysis revealed significant differences between the experimental groups, as well as a significant effect of the experimental variables ‘composite cement’, ‘curing mode’ and ‘ceramic surface pre-treatment’ on the luting effectiveness to dentin. Therefore, the first hypothesis tested must be rejected. Regarding ‘composite cement’, the luting effectiveness of both self-etch composite cements, Nexus 3 (‘XTR/NX3’ and RelyX Ultimate (‘SBU/RXU’), was not significantly different for ‘HF/S/HB’ ceramic surface pretreatment. Both cements
Fig. 4 – Representative photomicrographs of specimens luted with ED/PAF (dentin side) when the ceramic surface received the HF/S/HB pre-treatment for the curing mode ‘AA’ in (a), revealing an ‘adhesive fracture at the interface dentin–cement’, or more specifically a fracture at the bottom of the hybrid layer with exposure of dentin tubules and peritubular dentin (arrow); the curing mode ‘AA*’ in (b), revealing an ‘adhesive’ fracture at the interface dentin–adhesive with remnants of the adhesive still attached to the dentin surface (arrows); and the curing mode ‘AL’ in (c), revealing a fracture at the interface dentin–adhesive, similar to that observed for the curing mode ‘AA*’; the tubules were partly obstructed by the adhesive.
298
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
Fig. 5 – Representative photomicrographs of specimens luted with SBU/RXU (dentin side) when the ceramic surface received the HF/S/HB pre-treatment for the curing mode ‘AA’ in (a), revealing a ceramic surface partly covered with composite cement, also revealing voids most likely due to water inflow, while the fracture occurred at the interface dentin–cement; the curing mode ‘AA*’ in (b), revealing a ceramic surface covered with composite cement, but exposing less voids than observed for ‘AA’; the curing mode ‘LA’ in (c), revealing a ‘mixed’ fracture with the dentin surface (D) partly exposed, while the fracture also occurred within the composite cement (C) and on top of it (arrows); and the curing mode ‘LL’ in (d), revealing a ‘mixed’ fracture with dentin (D) partly covered with adhesive (A), while the fracture also occurred within the composite cement (C) and on top of it.
achieved a much and significantly higher TBS to dentin than the comparable luting protocol of the self-etch composite cement Panavia F2.0 (‘ED/PAF’, ‘AL’ curing mode). This might be attributed to the fact that ED Primer cannot not be lightcured separately as it acts as an initiator, while this step appeared essential for both other adhesive/cement combinations ‘XTR/NX3’ and ‘SBU/RXU’. Regarding ‘curing mode’, the significantly lowest TBSs were measured for both the auto-cure modes ‘AA’ and ‘AA*’. From literature, it is known that the degree of conversion (DC) of auto/self-cured dual-curing composite cements was lower than when they were light-cured [9,23]. The real-time polymerization profile during auto-cure differs from that during dual-cure; the maximum rate of polymerization (polymerization per sec, Rpmax ) was significantly lower for the auto-cure mode [9]. Likewise, the curing speed of dual-cured composite cements was measured to be 5–20 times slower during autocure than light-cure [24]. In our study, the curing mode ‘AA’ involved a rather severe protocol, since the luted specimens were after 7 min auto-cure at room temperature immediately immersed in water. The TBSs measured for all ‘AA’ groups were very low and resulted in numerous pre-testing failures (pfts). Most likely, this can be explained by the slower auto-cure polymerization of the composite cements [9–11,25]. As a consequence, water must have hampered polymerization or at least softened the mechanical properties, by
which either pre-testing failures occurred or a relatively low TBS was measured. This was confirmed by the SEM failure analysis, having disclosed interfacial porosities especially for SBU/RXU when applied following the curing mode ‘AA’ (Figs. 5a and 6a). When the composite cements were applied following the adapted auto-cure mode ‘AA*’, the TBS of both SBU/RXU experimental groups (for both ‘HF/S/HB’ and ‘HF/SBU’) was slightly higher than when the cements underwent the ‘AA’ curing mode, and less pre-testing failures occurred. No difference was measured between ‘AA’ and ‘AA*’ for XTR/NX3. Only for ED/PAF, a significant increase in TBS was measured, by which hypothesis 2 was solely rejected for ED/PAF. The curing mode ‘AA*’ involved an adapted, less severe auto-cure protocol with an extended polymerization time (10 min vs. 7 min), and an increased temperature (37 ◦ C vs. 21 ◦ C). Hence, a higher degree of conversion, lower water sorption and higher mechanical properties must have been reached. The significantly higher TBS of ED/PAF measured for ‘AA*’ versus ‘AA’ in this study can be explained by its Rpmax (see above) having been reported to be similar during auto/self-cure and light-cure [10]. Compared to two other dual-cure composite cements, the Rpmax of Panavia F2.0 was significantly higher [10]. Likewise, this finding might also explain the significantly higher TBS of ED/PAF found in our study, when compared to XTR/NX3 and SBU/RXU following the curing mode ‘AA*’.
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
299
Fig. 6 – Representative photomicrographs of specimens bonded with SBU/RXU (dentin side) when the ceramic surface received the HF/SBU pre-treatment for the curing mode ‘AA’ in (a), revealing a ceramic surface covered with composite cement, and showing voids most likely due to water inflow, while the fracture occurred at the interface dentin–cement; the curing mode ‘AA*’ in (b), revealing a ‘mixed’ fracture with exposed dentin (D), partly covered by the adhesive (arrow), while the fracture also occurred on top of the composite cement (C); the curing mode ‘LA’ in (c), revealing the ceramic surface with the fracture having occurred ‘adhesively’ at the interface cement–ceramic; and the curing mode ‘LL’ in (d), revealing a similar ‘mixed’ fracture pattern as in ‘AA*’ with exposed dentin (D), partly covered by the adhesive (A), while the fracture occurred also on top of the composite cement (C).
Overall, auto-cure leads to a lower TBS when compared to dual-cure, as measured in this and other studies [13,14]. In contrast to the adhesives Optibond XTR and Scotchbond Universal, tested in our study, the ‘Liquid B’ of ED Primer II contains T-isoprolic benzenic sodium sulfinate [25], which acts as a co-initiator when it comes in contact with the subsequently applied composite cement Panavia F2.0. This additional coinitiator may also have contributed to the significantly higher TBS measured for ED/PAF, when applied following the curing mode ‘AA*’ and ‘AL’. The latter curing mode, applied solely for ED/PAF, nevertheless resulted in a significantly lower TBS than for all other cements when they were light-cured (curing mode ‘LL’). Faria-e-Silva et al. [25] found that ED Primer II increased Rpmax when Panavia F2.0 was dual-cured. When the adhesive was light-cured and the cement was left to auto-cure following curing mode ‘LA’, no significant differences in TBS were recorded for XTR/NX3 and SBU/RXU, when the ceramic block was pre-treated following the ‘HF/S/HB’ ceramic surface pre-treatment. In our study, light-curing adhesives were used with dual-curing composite cements. Our data clearly showed that while using a lightcuring adhesive in an auto-curing set-up, the adhesive has to be light-cured in order to obtain the most favorable ‘immediate’ bond strength to dentin. By separately light-curing the adhesive at the dentin side, the hybrid and adhesive layer is thought to be stabilized before the composite cement is
subsequently applied, thereby also immediately sealing dentin and thus preventing water uptake from the dentin host through osmosis [26]. Self-evidently, this can only be done when the adhesive has a sufficiently low film thickness (after air-thinning), so that the restoration fit is not impaired. The film thickness of the adhesives Optibond XTR and Scotchbond Universal was previously found to be air-thinned below 10 m (unpublished TEM interfacial analysis). By separately lightcuring the adhesive, no further significant increase in TBS was measured when also the composite cement was cured afterwards following the curing mode ‘LL’. In comparison with activator-initiated self-curing (auto-cure), separate lightcuring of an adhesive without activator was shown to have led to a significantly higher DC [27]. Also, pre-(light-)curing of self-etch adhesives resulted in a higher bond strength than co-(light-)curing the adhesive with the composite cement through a ceramic specimen [28], this especially when the thickness of the ceramic block exceeded 2 mm [2]. Most likely, the improved TBS recorded for the curing modes ‘LA’ and ‘LL’ can be attributed to the higher DC of the adhesive and composite cement reached after light-curing. It must also have been reached more rapidly, thereby having lowered water sorption from dentin and prevented the mechanical properties to decrease, which was confirmed by the fracture analysis. When auto-cured (‘AA’; ‘AA*’), the predominant fracture mode was ‘adhesive’ at the interface dentin–cement. It was associated
300
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
with numerous pre-testing failures (ptf’s). After light-curing of the adhesive (‘LA’), a more diverse fracture pattern, including fractures within the composite cement, at the interface cement–ceramic, and within the ceramic occurred. Overall, light-curing of both adhesive and composite cement (curing mode ‘LL’) led to the highest TBS. Lightcuring of dual-curing composite cements was indeed shown to improve their mechanical properties and DC [9–11], and self-evidently must have contributed to the increased TBS measured in our study. Our results are in agreement with several studies, in which the highest TBS was measured for those composite cements that were light-cured [10,13,29]. As pre-(light-)curing of the adhesive leads to a significant increase of the TBS, this method must be applied in particular for thick restorations, which hamper light transmission. In the latter case, the composite cement is cured less efficiently with inferior bond strength and mechanical properties [3,4,30,31]. Pre-(light-)curing was also shown to have led to an increase in ‘continuous’ margins and bond strength of adhesively luted ceramic inlays [32,33]. Regarding the experimental variable ‘ceramic surface pre-treatment’, the ‘new’ ceramic pre-treatment ‘HF/SBU’, involving the application of a silane-incorporated adhesive (Scotchbond Universal) resulted in a significantly lower TBS when the curing mode ‘LA’ was applied, by which hypothesis 3 was rejected. Nevertheless, no difference in TBS was found when SBU/RXU was applied following the curing mode ‘LL’. The incorporation of silane into the adhesive is a promising approach, because it reduces the amount of working steps during the cementation of indirect restorations and allows the practitioner to use the same adhesive for the dental hard tissues and the ceramic. The fracture analysis indicated that the lower bond strength measured for ‘HF/SBU’ versus ‘HF/S/HB’ for SBU/RXU must be attributed to the ceramic pre-treatment, because the specimens in the ‘HF/SBU’ group mainly failed at the interface cement–ceramic. While following curing mode ‘LL’, no difference in TBS was detected between ‘HF/SBU’ and ‘HF/S/HB’ for SBU/RXU, still differences observed in the fracture pattern of both ceramic surface pre-treatments indicated that the weak link for the ‘new’ silane-incorporated adhesive approach was the interface cement–ceramic; for the ‘established’ separately applied silane primer, the specimens failed more at the interface dentin–cement.
5.
Conclusion
Complete auto-cure of adhesively luted ceramic restorations onto dentin, independent of the curing temperature (21 or 37 ◦ C) resulted in inferior ‘immediate’ TBS to dentin than when solely the adhesive at the dentin side was separately light-cured or both the adhesive and the composite cement were light-cured. When the adhesive is separately light-cured at the dentin side, special caution is needed to not impair the restoration fit by adequately air-thin the adhesive before light-curing. The new ceramic surface pre-treatment with a silaneincorporated adhesive, thereby simplifying the already complex adhesive luting procedure, affected the ‘immediate’ TBS to dentin only when the composite cement was not
light-cured and allowed to auto-cure. Thus, while this second session of light-curing when adhesively luting indirect restorations is highly recommended, today this is clinically done in most cases.
Acknowledgements This investigation was supported in part by the EFCD Scientific Foundation. All the materials used in this study were provided by the respective manufacturers.
references
[1] Burke FJ, Fleming GJ, Nathanson D, Marquis PM. Are adhesive technologies needed to support ceramics? An assessment of the current evidence. J Adhes Dent 2002;4:7–22. [2] Rathke A, Hokenmaier G, Muche R, Haller B. Effectiveness of the bond established between ceramic inlays and dentin using different luting protocols. J Adhes Dent 2012;14:147–54. [3] Soares CJ, da Silva NR, Fonseca RB. Influence of the feldspathic ceramic thickness and shade on the microhardness of dual resin cement. Oper Dent 2006;31:384–9. [4] Kilinc E, Antonson SA, Hardigan PC, Kesercioglu A. The effect of ceramic restoration shade and thickness on the polymerization of light- and dual-cure resin cements. Oper Dent 2011;36:661–9. [5] el-Mowafy OM, Rubo MH, el-Badrawy WA. Hardening of new resin cements cured through a ceramic inlay. Oper Dent 1999;24:38–44. [6] Akgungor G, Akkayan B, Gaucher H. Influence of ceramic thickness and polymerization mode of a resin luting agent on early bond strength and durability with a lithium disilicate-based ceramic system. J Prosthet Dent 2005;94:234–41. [7] Tashiro H, Inai N, Nikaido T, Tagami J. Effects of light intensity through resin inlays on the bond strength of dual-cured resin cement. J Adhes Dent 2004;6:233–8. [8] de Menezes MJ, Arrais CA, Giannini M. Influence of light-activated and auto- and dual-polymerizing adhesive systems on bond strength of indirect composite resin to dentin. J Prosthet Dent 2006;96:115–21. [9] Arrais CA, Giannini M, Rueggeberg FA. Kinetic analysis of monomer conversion in auto- and dual-polymerizing modes of commercial resin luting cements. J Prosthet Dent 2009;101:128–36. [10] Faria-e-Silva A, Boaro L, Braga R, Piva E, Arias V, Martins L. Effect of immediate or delayed light activation on curing kinetics and shrinkage stress of dual-cure resin cements. Oper Dent 2011;36:196–204. [11] Hofmann N, Papsthart G, Hugo B, Klaiber B. Comparison of photo-activation versus chemical or dual-curing of resin-based luting cements regarding flexural strength, modulus and surface hardness. J Oral Rehabil 2001;28:1022–8. [12] Aguiar TR, Di Francescantonio M, Ambrosano GM, Giannini M. Effect of curing mode on bond strength of self-adhesive resin luting cements to dentin. J Biomed Mater Res B: Appl Biomater 2010;93:122–7. [13] Arrais CA, Giannini M, Rueggeberg FA, Pashley DH. Microtensile bond strength of dual-polymerizing cementing systems to dentin using different polymerizing modes. J Prosthet Dent 2007;97:99–106.
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 292–301
[14] Faria-e-Silva AL, Fabião MM, Arias VG, Martins LR. Activation mode effects on the shear bond strength of dual-cured resin cements. Oper Dent 2010;35:515–21. [15] Garcia RN, Reis AF, Giannini M. Effect of activation mode of dual-cured resin cements and low-viscosity composite liners on bond strength to dentin. J Dent 2007;35:564–9. [16] Takahashi R, Nikaido T, Ariyoshi M, Foxton RM, Tagami J. Microtensile bond strengths of a dual-cure resin cement to dentin resin-coated with an all-in-one adhesive system using two curing modes. Dent Mater J 2010;29:268–76. [17] Foxton RM, Pereira PN, Nakajima M, Tagami J, Miura H. Durability of the dual-cure resin cement/ceramic bond with different curing strategies. J Adhes Dent 2002;4:49–59. [18] Stewart GP, Jain P, Hodges J. Shear bond strength of resin cements to both ceramic and dentin. J Prosthet Dent 2002;88:277–84. [19] Filho AM, Vieira LC, Araújo E, Monteiro Júnior S. Effect of different ceramic surface treatments on resin microtensile bond strength. J Prosthodont 2004;3:28–35. [20] Ikemura K, Tanaka H, Fujii T, Deguchi M, Negoro N, Endo T, et al. Design of a new, multi-purpose, light-curing adhesive comprising a silane coupling agent, acidic adhesive monomers and dithiooctanoate monomers for bonding to varied metal and dental ceramic materials. Dent Mater J 2011;30:493–500. [21] De Munck J, Vargas M, Iracki J, Van Landuyt K, Poitevin A, Lambrechts P, et al. One-day bonding effectiveness of new self-etch adhesives to bur-cut enamel and dentin. Oper Dent 2005;30:39–49. [22] De Munck J, Van den Steen PE, Mine A, Van Landuyt KL, Poitevin A, Opdenakker G, et al. Inhibition of enzymatic degradation of adhesive–dentin interfaces. J Dent Res 2009;88:1101–6. [23] Arrais CA, Rueggeberg FA, Waller JL, de Goes MF, Giannini M. Effect of curing mode on the polymerization characteristics
[24]
[25] [26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
301
of dual-cured resin cement systems. J Dent 2008;36: 418–26. Lee IB, Um CM. Thermal analysis on the cure speed of dual cured resin cements under porcelain inlays. J Oral Rehabil 2001;28:186–97. Faria-e-Silva AL, Moraes RR, Ogliari FA, Piva E, Martins LR. Panavia F: the role of the primer. J Oral Sci 2009;51:255–9. Van Landuyt KL, Snauwaert J, De Munck J, Coutinho E, Poitevin A, Yoshida Y, et al. Origin of interfacial droplets with one-step adhesives. J Dent Res 2007;86:739–44. Faria-e-Silva AL, Casselli DS, Lima GS, Ogliari FA, Piva E, Martins LR. Kinetics of conversion of two dual-cured adhesive systems. J Endod 2008;34:1115–8. Chapman JL, Burgess JO, Holst S, Sadan A, Blatz MB. Precuring of self-etching bonding agents and its effect on bond strength of resin composite to dentin and enamel. Quintessence Int 2007;38:637–41. Rathke A, Balz U, Muche R, Haller B. Effects of self-curing activator and curing protocol on the bond strength of composite core buildups. J Adhes Dent 2012;14:39–46. Sinhoreti MA, Manetta IP, Tango RN, Iriyama NT, Consani RL, Correr-Sobrinho L. Effect of light-curing methods on resin cement Knoop hardness at different depths. Braz Dent J 2007;18:305–8. de Paula AB, Tango RN, Sinhoreti MA, Alves MC, Puppin-Rontani RM. Effect of thickness of indirect restoration and distance from the light-curing unit tip on the hardness of a dual-cured resin cement. Braz Dent J 2010;21:117–22. Frankenberger R, Sindel J, Krämer N, Petschelt A. Dentin bond strength and marginal adaptation: direct composite resins vs ceramic inlays. Oper Dent 1999;24:147–55. Haller B, Hässner K, Moll K. Marginal adaptation of dentin bonded ceramic inlays: effects of bonding systems and luting resin composites. Oper Dent 2003;28:574–84.
