Multiple peak LED light-curing units may produce a non-homogeneous light output.
Non-homogeneity can result in uneven light activation of dental materials.
Homogeneous beams result in higher immediate bond strength of dental adhesives to dentin.
This study evaluated the effect of beam homogeneity on the microtensile bond strength (μTBS) of two adhesive resins to dentin.
One polywave light-emitting-diode (LED) LCU (Bluephase Style, Ivoclar Vivadent AG) was used with two different light guides: a regular tip (RT, 1010 mW/cm 2 emittance) and a homogenizer tip (HT, 946 mW/cm 2 emittance). The emission spectra and beam profiles were measured from both light guides. Extracted third molars were prepared for μTBS evaluation using two adhesive systems: Excite F (EXF) and Adhese Universal (ADU). Bond strength was calculated for each specimen (n = 10) at locations that correlated with the output of the two LED chips emitting blue (455 nm) and the one chip that emitted violet light (409 nm) after 24-hs and after one-year water-storage. The μTBS was analyzed using a four-way analysis of variance (factors: adhesive system, light guide, LED wavelength, and storage time) and post-hoc Tukey test ( α = 0.05).
EXF always delivered a higher μTBS than ADU (p < 0.0001), with the μTBS of ADU being about 20% lower than EXF. The light guide (p = 0.0259) and storage time (p = 0.0009) significantly influenced the μTBS. The LED wavelengths had no influence on the μTBS (p > 0.05).
Homogeneity of the emitted light beam was associated with higher 24-h μTBS to dentin, regardless of the adhesive tested. Also, differences in the composition of adhesives can affect their compatibility with restorative composites and their ability to maintain bonding over one year.
Several studies have reported that clinical failure of direct restorations occurs often at the tooth-material interface, as a result of inadequate polymerization [ ]. Previous studies have reported that inadequate polymerization of dental adhesives can cause premature failure, increased solubility [ , ], more leaching of uncured monomers [ ], increased degradation of the hybrid layer [ , , , ], and reduced bond strength [ , , ]. Therefore, the clinician must consider all possible influencing factors to reduce the failure rate and produce predictable, long-lasting direct restorations.
It is known that inadequate light curing can occur when insufficient radiant exposure (energy) is delivered to the resin. This can happen if the incident irradiance received by the resin is too low, or the exposure time is too short [ , ]. Also, if there is limited access of the light-curing unit to the cavity preparation [ ] or if there is an increased distance between the light source and the resin being polymerized [ , ] a high output light used for the correct time may still deliver an inadequate amount of energy.
In addition, aesthetic demands from patients and the incorporation of acidic monomers in self-etching adhesives have driven the manufacturers to introduce alternative photoinitiators to the camphorquinone-amine complex to improve the polymerization reaction [ ]. Camphorquinone (CQ) has an intense, bright, yellow shade, that could produce a yellowish hue in CQ-based restorative materials [ , , ]. Also, CQ can be inactivated in acidic environments because the tertiary amines are neutralized by acidic functional monomers [ , , ] that are added to self-etch adhesives to condition the mineralized dental tissues and improve stability of the bonding interface [ , , ]. The presence of these unreacted acidic monomers may affect the polymerization of self and dual-cure restorative composites [ ], and produce continued demineralization of the dentinal tubules [ ], leading to early degradation of the bonded interface.
Considering those limitations, some manufacturers have incorporated alternative photoinitiators such as Diphenyl(2,4,6-trimethylbenzoyl) phosphine oxide (TPO) and Bis-acylphosphine (BAPO). One manufacturer (Ivoclar Vivadent, Schaan, Liechtenstein) has recently introduced a new initiator, Ivocerin, to dental materials to reduce the yellowing of the restorations, provides enhanced light-curing efficiency [ , , ], and allows the material to reach a higher degree of conversion. However, the use of alternative photoinitiators has also created a necessity to develop broad-spectrum light-emitting diode (LED) light-curing units (LCUs) that emit light in the absorption spectrum of those photoinitiators, for them to be adequately activated [ , ]. These LCUs are known as “multiple-peak” or Polywave® lights and are characterized by the presence of 2 or more different types of LEDs, each producing light at different wavelengths. This results in a wider emission spectrum from the LCU with two or more wavelength peaks [ ]. Usually, one peak is in the blue region of the visible light spectrum, and the other peak is in the violet range [ ]. However, unless the LCU is well designed to blend and homogenize the light emitted from each LED, this can result in problems due to a lack of homogeneity of the emitted light beam [ ].
Non-homogeneous light beams from “multiple peak” LCUs results from the heterogeneous distribution of the different wavelength lights and radiant power outputs at the emitting end of the LCU. As a result, the surface of the light-curable material does not receive light at the same irradiance or at the same wavelengths [ , , ]. The problems of non-homogeneous light beams have been widely reported for resin-based composites [ , , , , ], however, there is little information regarding the effects on dental adhesives. Since the recent guidelines for evaluating the effectiveness of bonding to dentin and enamel recommend using LCUs emitting homogeneous beams [ , ], the impact of beam homogeneity has become relevant.
The purpose of this study was to evaluate the effects on the microtensile bond strength (μTBS) of when two different adhesive resins were light-cured with the same multiple peak LCU that used a homogenizer light guide to blend the light compared with a regular guide tip. The null hypotheses were that (1) the two adhesives would achieve the same μTBS dentin, (2) the use of the homogenizer light guide type would not influence the μTBS of the adhesives to dentin; (3) within the same light guide, the μTBS would not be affected by the wavelength of the light reaching the adhesive, and (4) the μTBS of the adhesives would not be affected by one-year of storage in water.
Materials and methods
One multiple peak LED curing unit (Bluephase Style, Ivoclar Vivadent) was chosen for this study. This LCU has two blue LED chips (B1 and B2 – emitting light with a peak at 455 nm) and one violet chip (VC – emitting light with a peak at 409 nm). Two commercially available light guides from Ivoclar Vivadent, regular (RT), and homogenizer (HT), both with 10 mm external diameter and 9.3 active internal diameters, were used on the same LCU.
The total radiant power, irradiance, and emission spectrum from both light guides were measured using a six-inch integrating sphere (Labsphere, North Sutton, NH, USA) coupled to a fiber-optic spectrometer USB-4000 (Ocean Insight, Largo, FL, USA). For the baseline measurement, the LCU light guide was placed 1-mm away from the entrance to the sphere, and the total area of the tip was measured. After the baseline measurement, individual measurements of the emission spectrum and spectral irradiance at a 4-mm diameter central area of each LED (B1, B2, and VC) were made through a 4-mm diameter aperture into the integrating sphere. The process was repeated through a 1-mm diameter aperture into the sphere to determine if any light beam homogenization using the HT could be detected.
The beam profile of the LCU with each light guide was obtained using a Laser Beam Profiler (USB-L070, Ophir Spiricon, Logan, UT, USA) with a 50-mm focal length lens, without any imaging target. Two-dimensional (2D) and three-dimensional (3D) beam-profile images of the LCU with each light guide were obtained using BeamGage version 6.6 software (Ophir-Spiricon).
Microtensile bond strength (μTBS)
After approval from the local Ethics Committee (Protocol #77867117.1.0000.5418), forty sound extracted third molars were collected and stored in distilled water with 0.5% Chloramine-T (Sigma Aldrich, St Louis, MO, USA) at 4 °C and used within 1-month from extraction [ ]. The enamel occlusal surface was removed and flattened ( Fig. 1 A) using a low-speed diamond-wafering blade (Isomet 1000 Precision Saw; Buehler Co., Lake Bluff, IL, USA) under copious water-cooling. A standardized smear layer was then created by polishing with 600-grit sandpaper (Norton, Guarulhos, SP, Brazil) for 20 s [ ]. Three wedge-shaped markings were prepared in the outer walls of each tooth at 120° from each other, using the bucco-lingual and mesio-distal center of the tooth as a reference. The teeth were randomly divided into four groups (n = 10), according to two light guides and two adhesives that contained different photoinitiator systems: Excite F (EXF, Ivoclar Vivadent AG, Schaan, Liechtenstein) and Adhese Universal (ADU, Ivoclar Vivadent AG, Schaan, Liechtenstein) and the two light guide tips (RT or HT). The tested adhesives were applied using an etch-and-rinse technique ( Fig. 1 B and C) [ ]. All information regarding classification, composition and application technique of the used materials is presented in Table 1 .
|Classification||Material, and lot number||Composition||Application Technique|
|Etchant||Ultra-Etch Lot BFDJ8||35% phosphoric acid, cobalt aluminate blue spinel, siloxane||Dentin was etched for 15 s and washed for 30 s using a water jet. Dentin was dried with mild air jet at 5 cm distance for 5 s.|
|Bonding agent||Excite F Lot W05459||Phosphonic acid acrylate, HEMA, Bis-GMA, UDMA, highly dispersed silicone dioxide, initiators, stabilizers and potassium fluoride in an alcohol solution, CQ and TPO photoinitiators||A drop of adhesive resin was applied in the etched dentin according to the manufacturer’s instructions. Excess solvent in the adhesive was volatilized with mild oil-free air jet at 5 cm distance for 10 s. The adhesive layer was light cured for 10 s.|
|Adhese Universal Lot W33260||10-MDP, 2-HEMA, Bis-GMA, MCAP, D3MA, DMAEMA, ethanol, water, highly dispersed silicon dioxide, stabilizers, CQ.|
|Restorative composite||BisFill 2B Lot 1700003960||Base: Bis-GMA, BPAEODMA, Bisphenol A ethoxylated dimethacrylate, amorphous silica Catalyst: TEGDM, silica, glass frit||The composite was injected into the stainless-steel matrix until a 5 mm-thick layer was obtained. The material was allowed to polymerize for 7 min.|
At the light guide, the location of the light output from each LED was identified. For that purpose, three markings using adhesive tape that corresponded to the emitting area of the LEDs (B1, B2, and VC) were fixed to the outer wall of the light guides. After the adhesive was applied and before light-activation, the LCU was fixed in a clamp with the emitting end of the light guide 1-mm away from the occlusal surface. The marks on the tooth were aligned with the marks on the light guide, and then the adhesives were photo-cured for 10 s. After light-activation, a custom stainless steel matrix band (10 mm diameter, 5 mm height; Preven, Guapirama, PR, Brazil) was placed on the occlusal surface and fixed with sticky wax (Cerafix, Pradópolis, SP, Brazil) for the coronal tooth reconstruction, with a self-curing restorative composite (BisFill 2B, Bisco, Chicago, IL, USA) ( Fig. 1 D). This avoided any effects from the curing light. The specimens were then stored in distilled water for 24 h at 37 °C. After storage, the teeth were sectioned serially into 1 × 1 mm sticks using a low-speed diamond-wafering blade (Isomet 1000 Precision Saw; Buehler Co., Lake Bluff, IL, USA) at 200 rpm with 150 g load ( Fig. 1 E). Only the sticks obtained from the central irradiant spot of each of the three LEDs were used for the μTBS measurement. Half of the sticks (2 or 3 sticks) were tested immediately, and the other half was stored in distilled water at 37 °C for one year ( Fig. 1 F) before μTBS testing [ ]. The distilled water used to store the specimens was changed every four weeks.
For the μTBS test, each stick was fixed to a custom micro-tensile testing device [ , ] using cyanoacrylate glue (Super Bonder Power Flex, Loctite, São Paulo, SP Brazil) with an accelerator (Zap Zip Kicker, Pacer Technology, Ontario, CA, USA). The testing device was placed in a Universal testing machine (EZ-test-500N, Shimadzu Co., Kyoto, Japan), and a tensile load (0.5 mm/min) was applied until failure ( Fig. 1 G). The sides of the stick were measured with a digital caliper (Mitutoyo Corp., Kawasaki, Japan) to calculate the bonded area, which was then used to calculate the μTBS strength (MPa) from the load N at failure. All specimen fabrication and testing procedures were performed with the teeth in 100% humidity conditions. All measurements were carried out by an operator blinded to the group being tested.
Failure pattern analysis
For the fracture pattern analysis, the sticks were dried, sputter-coated with gold (Desk ll, Denton Vacuum Inc., NJ, EUA) and examined using scanning electron microscopy (SEM) at ×100 and ×400 magnification (JSM IT 300; Jeol, Tokyo, Japan). The failure patterns were classified as: (1) adhesive failure between the adhesive and the composite resin, (2) adhesive failure between adhesive and dentin, (3) mixed failure, (4) cohesive failure within the adhesive layer, and (5) cohesive failure within the composite resin.
Morphology of composite-dentin interface
Four extracted sound third molars were prepared following the protocol applied for the μTBS test. After storing in water for 24 h, each tooth was mesio-distally sectioned in three parts according to the LED that light-cured the adhesive system. Each piece was sectioned again in two parts. One part of the segment was examined immediately, and the other was stored for 1 year in distilled water at 37 °C. After storage, the sections were embedded in epoxy resin and prepared for SEM observations. The specimens were flattened using 600-grit abrasive paper and polished using 800 and 1200-grit abrasive paper (Norton, Guarulhos, SP, Brazil) under copious water cooling and finally polished with felt disc and 1 μm diamond polishing paste (MetaDi II, Buehler Co., Lake Bluff, IL, USA).
After polishing, specimens were ultrasonicated (SoniClean 2PS, Sanders Medical, Santa Rita de Sapucai, MG, Brazil) twice in distilled water for 3 min. The polished area was treated with 50% phosphoric acid (Electron Microscopy Sciences, Hatfield, PA, USA) for 5 s, washed with 10% sodium hypochlorite (Drogal Manipulação, Piracicaba, SP, Brazil) for 10 min and rinsed with distilled water. After washing, the specimens were dehydrated using ascending ethanol series (30%, 50%, 70%, 90%, and 100%; Merck KGaA, Darmstadt, Germany) and fixed using HDMS (Electron Microscopy Sciences, Hatfield, PA, USA). The specimens were stored in silica gel for 48 h before sputter coating with gold (Desk ll) and examined in SEM at ×1200 magnification.
Data for μTBS were tested for normality and homogeneity. Since μTBS values were not normally distributed, they were transformed using the Box-Cox transformation method to achieve normality and confirmed by the Shapiro-Wilk test (p = 0.6292). The corrected data were analyzed by four-way ANOVA (Adhesive*Light guide*LED wavelength*Storage time) and post hoc Tukey test ( α = 0.05). For the failure pattern analysis, the incidence rate of each fracture pattern was calculated as a percentage for each group. The statistical analysis was performed using SAS 9.3 (SAS Institute, Cary, NC, USA).
Table 2 shows the total power output, and radiant emittance at different wavelength ranges from the LCU using the RT and HT light guides. In general, the power output and radiant emittance of the LCU were higher when the RT was used. Fig. 2 shows the selected emission spectrum of the light beam produced by the RT through the 1 and 4-mm apertures into the integrating sphere. It was apparent that the light beam from the RT light guide was not homogeneous. There was almost no violet light in the areas of the Blue LEDs; in the same way, almost no blue light was present in the region of the Violet LED. Fig. 2 also shows the light output of the HT through the 1 and 4-mm apertures into the integrating sphere. In contrast to the output from the RT, violet light was detected in the areas of B1, and B2 LEDs and blue light was detected in the region corresponding to the VC LED.
|Light Guide||Wavelength range (nm)||Power output (mW)||Radiant emittance (mW/cm 2 )|
|Regular Tip (RT)||350–550||687||1010|
|Homogenizer Tip (HT)||350–550||643||946|