Which is effective on bond strength of resin-based sealers: incorporation of powdered dentin to primer or adhesive?

The aim of this study was to investigate the effect of adding powdered dentin to primer or adhesive in a self-etch system on the shear bond strength of three resin-based sealers. Seventy-two premolars were sectioned buccolingually, and 144 root halves were divided into three groups according to the sealer used: epoxy resin-based sealer (AH Plus), methacrylate resin-based (RealSeal, Hybrid Root SEAL) n = 48. The surfaces were irrigated with 5% NaOCl, 17% EDTA, distilled water for 5 min. Four subgroups were created (n = 12): control group; Clearfil Liner Bond 2 V treated group; powdered dentin added to the primer of Clearfil Liner Bond 2 V (40 wt.%); and powdered dentin added to the adhesive of Clearfil Liner Bond 2 V (20 wt.%). Dentin powder was prepared. Three mm high buildups with a constant surface area of 3.45 mm² were created using the sealers and allowed to set (37 ºC, 100% humid, 72 h). The samples were tested to failure for shear bond strength (1 mm/min). The data were calculated (MPa) and analyzed using two-way ANOVA, one-way ANOVA, and Tukey HSD tests. Adhesive use decreased the bonding performance of AH Plus (p = 0.00). Mean bond strength of the other sealers was found similar to control. Primer or adhesive resin with powdered dentin did not increase the adhesive performance of the self-etch system used. The shear bond strength of RealSeal was significantly increased when powdered dentin was added to primer or adhesive (p = 0.00). The effect of adding powdered-dentine to primer or adhesive in a self-etch system on the shear bond strength was sealer-dependent .     

Effect of dental silane primer activation time on resin–ceramic bonding

The purpose of this study was to investigate the effect of activation time for the hydrolysis of dental silane primers on resin bonding to ceramic. Two commercial two-part silane primers (Bis-Silane, BS; Tokuso Ceramic Primer, TCP) were tested. Leucite-reinforced glass-ceramic (IPS Empress CAD) surfaces were used as the bonding substrates. The degree of hydrolysis of each mixed silane was observed at five specified times (immediately, 10 min, 30 min, 1 h, and 2 h after mixing) using Fourier transform infrared (FTIR) spectroscopy. Prior to resin (Duo-Link) bonding, the polished ceramic surfaces were pretreated with silanes that were activated in two different ways: an ‘in-solution’ or ‘on-surface’ activation, at the specified times. All bonded specimens were stored in water at 37 °C for 24 h prior to tensile bond strength (TBS) testing (n = 12). A one-bottle silane (Monobond-S, MS) was also tested. The FTIR analysis showed that the hydrolysis proceeded gradually for 2 h in the BS, but the initial hydrolysis halted in the TCP. When the BS was further hydrolyzed up to 2 h, the TBS values significantly increased up to 30 min (p < 0.001) with no further significant increases (p > 0.05) regardless of the activation method used. For TCP, there were no significant differences in TBS as a function of activation time (p > 0.05). The fractographic analysis of the debonded surfaces was consistent with the TBS data. In conclusion, a lengthened activation time was effective for the BS in enhancing resin bonding efficacy to the ceramic.     

Influence of hydrogen peroxide-based bleaching agents on the bond strength of resin–enamel/dentin interfaces

This study evaluated the effect of different bleaching techniques on the bond strength of pre-existing adhesive restorations in enamel and dentin. Hydrogen peroxide-based bleaching gels with different concentrations (7.5% and 35%) were used on composite restorations of Adper Single Bond 2 (3M/ESPE, St. Paul, USA) and Filtek Z250 (3M/ESPE, St. Paul, USA). Twenty human third molars were randomly divided into 8 groups: GE—enamel control; GE7.5—bleaching using 7.5% hydrogen peroxide; GE35—bleaching using 35% hydrogen peroxide; GE 7.5+35—bleaching using 7.5% and 35% hydrogen peroxide; GD—dentin control; GD7.5—7.5% hydrogen peroxide; GD35—35% hydrogen peroxide; and GD 7.5+35—7.5% and 35% hydrogen peroxide. Bleaching was performed using long clinical application-time to low concentration gel, and short clinical application-time to high concentration gel. Unbleached specimens were stored in artificial saliva for 14 days. Specimens subject to micro-shear testing and data were analyzed by Analysis of Variance and Tukey's test (p=0.05). Enamel micro-shear bond strength was reduced after 7.5% hydrogen peroxide and after association of 7.5% and 35% hydrogen peroxide. Bleaching treatment altered dentin bond strength only when using 7.5% hydrogen peroxide. The results suggest that the bond strength of the restorations was influenced by the clinical extent of bleaching-gel application time and was not dependent on bleaching-gel concentration.     

The influence of hydroxyl group concentration on epoxy–aluminium bond durability

The influence of hydroxyl group (OH) concentration on the durability of adhesive bonds formed between an epoxy resin and aluminium adherend has been examined. Initially, surface analysis in combination with chemical derivatisation was employed to characterise the OH and epoxy functional groups present in FM-73, a structural epoxy adhesive. Bulk FM-73 indicated a higher degree of cure than the surface of FM-73 present at the interface of an epoxy–aluminium adhesive joint. Plasma and water treatment of the aluminium adherend was employed to alter the metal oxide's surface OH concentration. Despite a several-fold difference of aluminium surface OH concentrations for the different metal pre-treatments, there was no significant variation in the adhesive joint fracture toughness in a humid environment, G _Iscc. In contrast, grit-blasting the aluminium prior to bonding increased G _Iscc almost 15-fold. Simple calculations indicate that the aluminium surfaces used in the bonding experiments would have a large excess of OH groups available to react with a standard epoxy resin and that the influence of surface roughness on adhesion durability is not insignificant.     

An overview of solvents in resin–dentin bonding

Resin bonding to dentin is less durable than bonding to enamel. Bonding to dentin had been a great challenge till date. Though the current dental adhesive systems have tremendously improved the complex dentin bonding, there are still challenges that need to be addressed. Improper handling of the dental adhesive systems is one of the several reasons for the decrease in durability of resin–dentin bonds over time. Improper handling includes inappropriate storage and inaccurate method(s) of application. Solvents incorporated in the dental adhesive systems are known to dissolve methacrylate monomer/co-monomer blends and the initiator/co-initiator systems. These solvents are volatile and they play several important roles in achieving successful and durable bond to dentin. Lack of knowledge about the role of solvents used in dental adhesives could be one of the reasons for improper handling of the dental adhesive systems. This overview discusses the solvent aspects of current dental adhesive systems and their applications in bonding. The overview consists of published literature retrieved from “Pubmed”, by using specific key words “solvents”, “resin adhesives”, “dentin”, “bonding”. The relevant literature was retrieved after screening the titles and the abstracts. Further relevant articles that were included for the overview were retrieved from the reference lists of the initially retrieved articles. The literature seems to indicate that solvent type, solvent content and other factors, like storage of the adhesives and methods of application play a major role in dentin bonding. The knowledge acquired on the solvents and their roles in dentin bonding would enable the users of the dental adhesives to handle them appropriately and consequently achieve strong and durable resin–dentin bonds.     

Effect of N–Ni coordination bond on the electrical and thermal conductivity of epoxy resin/nickel-coated graphite

The agglomeration of nickel-coated graphite (NCG) in epoxy resin (EP) composites leads to low electrical conductivity of EP composites, which limits their development in electronic devices and multilayer circuits. In order to improve the electrical and thermal conductivity of NCG/EP composites, ethylenediamine (EDA) was used to modify NCG and compared with pure NCG-filled EP composites. It was found that the conductive effect of modified composites with 20 wt% filler is better than that of unmodified composites with 40 wt% filler. The results of Fourier transform infrared spectroscopy and thermogravimetric analysis of EDA-modified NCG (ENCG) showed that a coordination adsorption reaction occurred between EDA and NCG, forming N–Ni coordination bonds. When the filling amount of ENCG was 40 wt%, the conductivity and thermal conductivity of the composite are improved most significantly. The volume resistivity was reduced from 2.636 to 0.109 Ω cm, a decrease of 95.85%, and the thermal conductivity was improved from 0.517 to 0.968 W/(m K), an increase of 87.23%, respectively. Meanwhile, ENCG has better dispersion in the EP matrix than NCG.     

Effect of spent sulfite liquor on urea–formaldehyde resin performance

Combination of urea–formaldehyde (UF) resins with technical lignins has been often reported in the literature. However, the actual implications of this approach have not been effectively addressed yet. In this work, unmodified thick spent sulfite liquor (TSSL) and hydroxymethylated TSSL (TSSLH) were incorporated in a standard UF resin in different amounts (10 and 20%) and at different stages. When 10% of TSSLH was incorporated after the synthesis, the produced particleboards performed equivalently to when 90% of UF resin was used. In all other cases tested, combining UF resin with TSSL/TSSLH actually led to lower internal bond strengths. The results evidence that addition of TSSL or TSSLH does not have a beneficial effect on UF bonding performance. © 2018 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2019 , 136, 47389.     

Influence of storage time on bond strength of self-etching adhesive systems to artificially demineralized dentin after a papain gel chemical–mechanical agent application

ObjectivesThe aim of this study was to evaluate the influence of storage time on the bond strength of self-etching adhesive systems to artificially demineralized dentin submitted to application of a papain-based chemical mechanical agent for carious tissue removal.Materials and MethodsTwenty-four blocks of human coronal dentin were randomly divided into 2 groups: (1) restored with a two-step self-etching adhesive system (Clearfil SE Bond); (2) restored with a one-step self-etching adhesive system (One-Up Bond F Plus). After artificial caries induction, the specimens were treated with papain-based gel, received an application of the adhesive system according to the respective group, and blocks of microhybrid resin composite measuring 5.0 mm high and 5.0 mm wide were fabricated incrementally on the tooth, which would later be cut to obtain sticks with a bond area of around 1 mm², for use in the microtensile tests. After this, half of the sticks obtained from each tooth were submitted to the microtensile test 24 h later, while the other half were stored in water, in an oven at 37° C for a period of 180 d. The specimens were submitted to the microtensile test in a Universal Test Machine at a speed of 0.5 mm/min. The surfaces of the fractured test specimens were examined visually under a stereoscopic loupe in order to classify the fracture type. After exploratory analysis of the data, the two-way Analysis of Variance (ANOVA) and the Tukey test were applied. The data on the fracture types were analyzed by the Exact Fisher test.ResultsNo statistically relevant differences were verified among the means of the microtensile bond strength values of the adhesive systems evaluated in the different storage times. Mixed and adhesive fractures types were observed for both groups.ConclusionStorage time did not influence the bond strength of the two-step or one-step self-etching adhesive systems to artificially demineralized dentin submitted to application of a papain-based chemical mechanical agent for carious tissue removal. Both adhesive systems provided similar bond strength at different storage times.     

Effect of surface pre-treatment on the adhesive strength of epoxy–aluminium joints

A detailed study of the effect of pre-treatment applied on the surface characteristics of aluminium substrates and on the adhesive strength of epoxy–aluminium joints is reported. The variation of the density, composition and aspect of the adherends were analysed as a function of the applied pre-treatment. In order to determine the influence of alloying elements, two different aluminium alloys were used, A1050 and A2024. The adhesive strength was measured by the lap shear test, using several epoxy resins to analyse the influence of the adhesive nature.A chromate-free treatment based on the sulphuric acid-ferric sulphate etch provided an improved joint strength compared to dichromate-sulphuric acid etching, alkaline etching or mechanical abrasion. This increase is associated to the porous oxide layer formed, but it depends on the adhesive nature used. The joints with Al–Cu–Mg alloy substrates generally presented higher adhesive strength values than those with pure aluminium adherends, due to the selective etching of some allowing elements and intermetallic compounds, which have different electrochemical potential.     

Adhesion of substrate–adherent combinations for early composite repairs: Effect of intermediate adhesive resin application

This study evaluated the effect of intermediate adhesive resin application (IAR) on tensile bond strength (TBS) for early composite repairs in situations where substrate and repair composite bonded together were once of the same kind with the substrate (similar) and once other than the substrate material (dissimilar). Specimens from three types of composites (TPH Spectrum (TPH), Charisma (CHA) and Filtek Z250 (Z250)) were fabricated. The specimens in each composite group (n=72) were randomly divided into six subgroups (n=12). In each composite group, the similar and two dissimilar composites were bonded onto the substrates once using an IAR (Adper Single Bond Plus) and once without. After water storage for 1 week at 37 °C, substrate–adherent combinations were submitted to tensile test. Data were analyzed with three-way ANOVA and Tukey's tests (α=0.05). The substrate–adherent combination (p=0.0001), adherent (repair) composite (p=0.0001), and application of IAR (p=0.0001) significantly affected the results. Utilization of IAR improved the repair bond strength for all composite combinations.     

Effect of CeO2 addition into Pd/Zr–Pr mixed oxide on three-way catalysis and thermal durability

XPS shows that the surface chemical state of Pd in Pd/ZrPrO_x is mainly Pd^4 +, which is likely doped in ZrPrO_x. However, the Pd^4 + is segregated out as PdO when Pd/ZrPrO_x is exposed at 1000 °C in air, resulting in the particle growth of Pd and deterioration. It has been found that the mixing Pd/ZrPrO_x with CeO₂ enhances the catalytic activity and thermal durability. Analysis using XPS and HAADF-STEM shows that the mixing effect of ceria is to stabilize Pd^4 + at the high temperature and ceria also functions as an excellent support for the segregated PdO, which migrates from ZrPrO_x to CeO₂.     

Effect of incorporation of hafnium diboride nanofibers on thermomechanical properties of carbon fiber–phenolic composites

Carbon fiber/phenolic (C/Ph) composites were modified with different weight ratios of hafnium diboride (HfB₂) nanofibers to apperceive thermomechanical properties of C/Ph–Hf nanocomposites. Mechanical properties, thermal stability, and ablation resistance of C/Ph–Hf nanocomposites were found to be optimum when the weight percentage of HfB₂ was equal to one. Maximum flexural strength and modulus were obtained with 118 MPa and 1.9 GPa for C/Ph–1%Hf nanocomposite, respectively. Increasing the proportion of HfB₂, by delaying the temperature of thermal degradation of nanocomposites, enhanced the thermal stability and residual of C/Ph–Hf relative to C/Ph in both nitrogen and air environments. In the oxyacetylene flame test at 2500°C for 160 s, the optimum mass ablation rate of C/Ph–1%Hf nanocomposites was found to be 0.0150 g/s compared to 0.068 g/s for blank C/Ph, along with reducing the back surface temperature by 51%. The ablation mechanism of C/Ph–Hf nanocomposites after the oxyacetylene torch test was concluded from the derivations obtained from X-ray diffraction, energy dispersion spectroscopy, and microstructure analyses. These clarified that the formation of high-temperature species, such as HfO₂, HfC, and B₄C owing to oxidation of HfB₂ and subsequent reaction products with char, resulted in an increased ablation resistance of the nanocomposites.     

Durability of resin bond strength to dental noble metal–ceramic alloys conditioned with novel mercapto silane-based primer systems

The purpose of this study was to investigate the efficacy of novel mercapto silane-based experimental primers on the resin bonding and its durability to dental noble metal–ceramic alloys in comparison with that of commercial primers. Disc-shaped gold–platinum–palladium, gold–palladium–silver, and palladium–silver alloy specimens were used as the adherents after air-abrasion. One of three commercial primers (M.L. Primer, Single Bond Universal, and All-Bond Universal) and two experimental primer systems (2-step application with γ-mercaptopropyltrimethoxysilane and then γ-methacryloxypropyltrimethoxysilane and a blend of the two silanes) was applied to each alloy. Resin cylinders with a diameter of 2.38 mm were bonded to the surfaces and light-cured. All bonded specimens were stored in water at 37 °C for 24 h and then half of them additionally water immersed for 7 days (37 °C) and then thermocycled 10,000 times before the shear bond strength test (n = 10). The surface energy parameters for unprimed and primed alloy surfaces were calculated based on the contact angle measurements. The bond strength data were non-parametrically analyzed at α = 0.05. Regardless of the alloy type, both mercapto silane systems equally and consistently showed superior bonding durability to the commercial primers. Pearson correlation analyses revealed moderate to strong, significant correlations between the surface energy parameters and the bond strength values. The two novel mercapto silane systems are a promising alternative for enhanced durability of resin bonding to dental noble metal–ceramic alloys.     

The effect of different cementing strategies and adhesive interface aging on microtensile bond strength (μTBS) of lithium disilicate ceramics to dentin

This study evaluated the effect of different cementing strategies and adhesive interface aging on microtensile bond strength (μTBS) of lithium disilicate ceramic (IPS e.max CAD) to dentin. Forty coronal dentin fragments were randomly assigned to four groups according to the cementing strategy used to bond lithium disilicate ceramic to coronal dentin surface (n = 10): U200 (self-adhesive resin cement (RC) RelyX U200®/3 M ESPE), SBU (single-step self-etching adhesive system (AS) Single Bond Universal®/3 M ESPE + RelyX ARC®/3 M ESPE RC), AdperSB (two-step etch-and-rinse AS Single Bond 2®/3 M + RelyX ARC®/3 M ESPE RC) and Scotchbond (three-step etch-and-rinse AS Scotchbond Multi-Purpose®/3 M + RelyX ARC®/3 M RC). After 48 h, the ceramic-tooth blocks were sectioned perpendicular to the adhesive interface in the form of sticks and randomly subdivided into two groups according to when they were to be submitted to μTBS testing: immediately or 6 months after storage in water. Some sticks were kept for analysis of the adhesive interface by scanning electron microscopy (SEM). The μTBS test was performed in a universal testing machine (0.5 mm/min). The data (MPa) were analyzed using split-plot ANOVA and Tukey’s test (α = 0.05). Water storage decreased μTBS in all cementing strategies. The μTBS was greatest in the Scotchbond group and lowest in the U200 group, at both storage times. No signs of interface degradation were detected under SEM after water storage. In conclusion, water storage decreased bond strength, regardless of the adhesive cementation strategy, and that the three-step adhesive system/dual-cure resin cement ultimately performed better in terms of bond strength.     

Effect of phenol formaldehyde resin on the adhesion properties of butyl rubber–polyester proofed fabric

Phenol formaldehyde resin was used as an adhesion promoter additive to two bonding system: resorcinol and m-phenylenediamine. These two systems were used to produce butyl rubber–polyester proofed fabric. The addition of resin led to better adhesion properties. The coated fabrics were exposed to heat aging and ionizing radiation. The permeability, dielectric properties, and the electrical resistivity of the coated fabrics were examined.     

Soy-based adhesive cross-linked by melamine–glyoxal and epoxy resin

To develop a soy-based adhesive with good water resistance, non-toxic melamine–glyoxal resin (MG) prepared in the laboratory was used as a cross-linker of soy-based adhesive. The FT-IR and ESI-MS results showed that there was a reaction between melamine and glyoxal. The resulted –CH–OH– groups could be the possible reactive groups for the cross-linking of soy-based adhesive. The wet shear strength of soy-based plywood indicated that the water resistance of soy adhesive cross-linked by MG improved with respect to that with no cross-linker, although it was not good enough to satisfy the relative standard. With the optimized preparation procedures for plywood, specifically, press temperature 180?°C, press time 3 min and resin loading 280 g m^?2, type I soy-based plywood could be prepared with a hybrid cross-linker, namely 12%MG + 2% epoxy resin (EPR). The DSC results showed that the reaction between soy-based adhesive and the hybrid cross-linker MG + EPR was very complex.     

Studies on the adhesive properties of neoprene–phenolic blends

Polychloroprene (neoprene) rubber in combination with phenolic resins is a versatile adhesive formulation. The phenolic resin used in this case was derived from a mixture of cardanol, a meta-substituted naturally-occurring substance, and phenol. Cardanol is the main ingredient of cashew nut shell liquid (CNSL), a renewable resource. This study aims to investigate the adhesive properties of cardanol-based resin when used in combination with two grades of polychloroprene rubber. The effects of varying the solid content and resin content, choice of resin, fillers, crosslinking agents, adhesion promoters, solvents, etc. in the adhesive formulations were also studied. Moreover, relative proportions of rubber and resin that give optimum adhesion performance were identified. The results show that cardanol-phenol-formaldehyde resin is an effective ingredient in adhesives for bonding aluminium to aluminium and SBR to SBR. The addition of 3-aminopropyltriethoxysilane to the formulation improves the bond strength of metal-to-metal specimens.     

Bonding of resin adhesives to caries-affected dentin – A systematic review

Minimal invasive dentistry aims at preserving the firm, discolored caries affected dentin (CAD), which is remineralizable. Research studies on resin adhesives are usually performed on sound dentin (SD), though CAD is the substrate routinely encountered for bonding in clinical practice. The aim of this paper was to systematically analyze the published literature on resin-dentin bonding to CAD substrate, in order to answer the question: “Does resin adhesive bonding to CAD produce lower bond strength when compared to SD?”. Three electronic databases (Pubmed, Scopus and ISI web of Science) were searched to identify original laboratory studies that evaluated the bond achieved between resin adhesive and natural CAD by measuring their bond strength. Only articles that met the specific inclusion criteria were included in the review. Among the 29 studies included for this review, majority of the studies had tested the simplified etch-and-rinse or self-etch adhesives. 85% of them showed higher bond strength to SD compared to CAD and the remaining 15% of them showed no difference between these two substrates. Among the studies that used 3-step etch-and-rinse adhesives, 40% showed higher and 60% showed no difference, when bond strength was compared between SD and CAD. Resin adhesives produce lower bond strength to caries-affected dentin than sound dentin. Research studies that reported bond strength of resin adhesives to dentin from sound extracted teeth alone cannot be blindly extrapolated to clinically relevant CAD. Hence, the results from such studies should be dealt with caution.     

Effect of impregnated phenolic resin on the properties of Si–SiC ceramic matrix composites fabricated by SLS-RMI

Selective laser sintering (SLS) combined with reaction melt infiltration was used to fabricate Si–SiC ceramic matrix composites, and the effects of different concentrations of phenolic resin (PF) on the properties of the SLS green body and carbonized and final Si–SiC samples were investigated. The results showed that the impregnation with PF can increase the bulk density, reduce the porosity of the samples at all stages, and improve the mechanical properties of the reactive bonded samples. The degree of densification and mechanical properties of the sample gradually enhanced with an increase in PF concentration. The main phases of the Si–SiC composites were free Si, α-SiC, β-SiC, plus an extremely small amount of Al–Si alloy, and the SiC and the Si phase contents increased and decreased, respectively, as the concentration of PF increased when measured using Rietveld refinement and image analysis software. The macroscopic properties of the samples improved greatly after precursor infiltration pyrolysis (PIP) treatment with 66.7%vol PF-ethanol solution twice. According to the crystal nucleation-growth theory, it was inferred that the infiltrated PF could provide a certain amount of pyrolytic carbon in the carbonized specimen. During the reaction bonded process, the carbon formed by carbonization pyrolysis first dissolves into the molten Si and reaches saturation. With the further dissolution of carbon, [C] and [Si] in the liquid phase contact each other to form β-SiC nuclei, the nuclei that precipitate at the pore wall position and gradually form a continuous interfacial layer of β-SiC. The β-SiC layer prevents the liquid Si from direct contact with C inside the prefabricated body, therefore, further reactants diffuse through the layer. Finally, the fine crystalline β-SiC grains were fabricated inside the specimen.     

Micromorphology of resin–dentin interfaces using self-adhesive and conventional resin cements: A confocal laser and scanning electron microscope analysis

Purpose: The aim of this study was to evaluate the resin–dentin morphology created by four dual-cured resin cements. Materials and Methods: Two self-adhesive resin cements (RelyX Unicem, 3 M ESPE and Clearfil SA Luting, Kuraray Med.) and two conventional resin cementing systems (RelyX ARC, 3 M ESPE and Clearfil Esthetic Cement, Kuraray Med.) were evaluated. Occlusal dentin surfaces of 32 extracted human third molars were flattened to expose coronal dentin. Teeth were assigned to 8 groups (n=4), according to resin cement products and microscope analysis (SEM: scanning electron microscope or CLSM: confocal laser scanning microscopy). For CLSM, two different fluorescent dyes, fluorescein isothiocyanate–dextran and rhodamine B, were incorporated into the adhesive system and resin cement, respectively. The resin cements were applied to indirect composite resin disks, which were cemented to dentin surface according to manufacturer's instructions. After 24 h, all restored teeth were vertically sectioned into 1-mm-thick slabs for SEM or CLSM analyses. Results: Scotchbond Multi-Purpose Plus/RelyX ARC cementing system formed a thin hybrid layer and resin tags penetration into the dentin tubules. Clearfil DC Bond/Clearfil Esthetic Cement showed only short resin tags. Neither hybrid layer nor resin tags were detected for self-adhesive resin cements. Conclusion: Representative SEM and CLSM images provided resin–dentin interfaces variability among resin cements studied.