Clin Oral Invest DOI 10.1007/s00784-015-1494-4
Bonding between CAD/CAM resin and resin composite cements dependent on bonding agents: three different in vitro test methods Simona Gilbert 1 & Christine Keul 1 & Malgorzata Roos 2 & Daniel Edelhoff 1 & Bogna Stawarczyk 1
Received: 3 July 2014 / Accepted: 4 May 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Objectives The aim of this study was to assess the bonding properties between CAD/CAM resin and three resin composite cements combined with different bonding agents using three test methods. Materials and methods Four hundred twenty CAD/CAM resin substrates were fabricated and divided into three test methods (shear bond strength (SBS, n=180), tensile bond strength (TBS, n=180) and work of adhesion (WA, n=60)), further into four pretreatment methods (VP connect (VP), visio.link (VL), Clearfil Ceramic Primer (CP) and no pretreatment (CG)) and three cements (RelyX ARC, Variolink II and Clearfil SA Cement). Each subgroup contained 15 specimens. SBS and TBS were measured after 24 h H2O/37 °C+5000 thermal-cycles (5/55 °C) and failure types were assessed. WA was determined for pretreated CAD/CAM resin and non-polymerized resin composite cements. Data were analysed with Mann–Whitney U, Kruskal–Wallis H, Chi2 and Spearman’s Rho tests. Results Within SBS and TBS tests, CGs and groups pretreated with CP (regardless of resin composite cements), and VP pretreated with Clearfil SA Cement showed no bond. However, CG combined with RelyX ARC showed a TBS of 5.6±1.3 MPa. In general, highest bond strength was observed for groups treated with VL. CG and groups pretreated using VL showed lower WA than the groups treated with VP or CP.
* Bogna Stawarczyk [email protected]
Department of Prosthodontics, Dental School, Ludwig Maximilian University of Munich, Goethestrasse 70, 80336 Munich, Germany
Division of Biostatistics, Institute of Social and Preventive Medicine, University of Zurich, Hirschgraben 84, 8001 Zurich, Switzerland
Conclusions Measured TBS values were higher than SBS ones. In general, SBS and TBS showed similar trends for the ranges of the values for the groups. WA results were not comparable with SBS/TBS results and admitted, therefore, no conclusions on it. Clinical relevance For a clinical use of XHIP-CAD/CAM resin, the bond surface should be additionally pretreated with visio.link as bonding agent. Keywords CAD/CAM resin . Resin composite cement . Bonding agents . Shear bond strength . Tensile bond strength . Work of adhesion
Introduction CAD/CAM resins for provisional fixed dentures reveal a wider range of indications than the conventionally produced ones. The predictability for the definitive restoration is an especially important advantage of the new CAD/CAM methods [1, 2]. The standardized industrial processing of CAD/CAM resins leads to significantly higher mechanical properties [3–7], including wear resistance [6, 7], in comparison to conventionally polymerized resins. Due to the high pressure and temperature during the fabrication of the resin blanks, there is a reduced risk of porosities and inhomogeneities for CAD/CAM manufactured restorations [8–10]. Also, the advanced optical behaviour, such as superior colour stability  and a decrease of monomer release , can be attributed to the advantages of CAD/CAM resins when compared to conventionally polymerized resins. In general, resins consist of a polymeric matrix that is reinforced by inorganic (ceramic or glass ceramic), organic or composite fillers . The first generation of CAD/CAM resins was usually filled or unfilled polymethylmethacrylate
Clin Oral Invest
(PMMA) with modified polymer networks . Further developments led to new and improved classes of resin materials, which can be optimized by variation in view of the matrix and/or the fillers . The standardized industrial polymerization of CAD/CAM resins under high pressure [14, 15] or heat treatment  results in a higher degree of conversion with fewer residual monomer - containing free double bounds - in the material. Due to the low number of reaction units, the bonding to the resin composite cement seems to be difficult . Although creating a durable bond between the restoration and the tooth is crucial for long-term reliability, and therefore, for its success, no certain recommendations are available concerning the intraoral service duration for CAD/CAM manufactured provisionals . Durable bonding between different substrates takes place on the basis of the chemical, physical and mechanical adhesion properties between the substrate surface and the resin cement . Generally, the mechanical adherence after roughening of a solid surface is the most powerful one, but resin–resin interfaces also require physical and chemical bonding, and the latter can be improved by the additional use of adhesive systems . Bonding is related to the surface properties of the resin substrate surfaces [19, 20] and can be quantified Bnon-destructively^ with contact angle measurements [19, 21]. This is the most common test method which gives important information about the work of adhesion (WA) and thus the strength of the contact between two phases, the interfacial tension (IFT) and the spreading coefficient (SC) of the different materials . The IFT describes the tension between the new bonds and, therefore, serves as a factor of long-term effect, while the SC determines the initial wetting. To describe the bond strength between different substrates, different Bdestructive^ test methods, including the well-known shear bond strength and tensile bond strength tests or newer and more accurate test methods such as micro-shear and micro-tensile tests, have been described [22–25]. Prior studies have investigated the bonding properties of resin composites to different CAD/CAM resins [17, 26–28]. PMMA-based CAD/CAM resins without airborne particle abrasion showed no bonds to composite materials . The additional use of bonding agents showed an improvement of bond strength results [26, 27]. An improvement of bond values to resin cement was achieved by treatment of the CAD/CAM composite with silane coupling agents . To the authors’ knowledge, limited information is available about the concurrent investigation of and direct comparison between the established destructive and non-destructive test methods. The aim of this study was to investigate the bond strength of different resin composite cements combined with different bonding agents to CAD/CAM resin. The methods used were the tensile bond strength test, shear bond strength test and determination of work of adhesion. The null hypotheses tested
were that (1) the different bonding agents have no impact on the bond strength and (2) the different test methods lead to the same conclusions about and trends in the bond strengths properties.
Materials and methods This study tested the bond strength properties to XHIPCCAD/CAM resin (Xplus3, Echzell, Germany) after following pretreatment methods using different bonding agents: VP connect (Merz Dental, Lütjanburg, Germany), visio.link (bredent, Senden, Germany) or Clearfil Ceramic Primer (Kuraray Med., Sakazu, Japan). They were bonded with two conventional resin cements, RelyX ARC (3M ESPE, Seefeld, Germany) and Variolink II (Ivoclar Vivadent, Schaan, Liechtenstein) and a self-adhesive resin cement, Clearfil SA Cement (Kuraray, Tokyo, Japan). A control group without a bonding agent was also used in combination with all cements. For preparation of the specimens, the CAD/CAM blanks were separated (Secotom-50, Struers, Ballerup, Denmark) into 420 slices (10×10×3 mm) and embedded in self-curing acrylic resin (ScandiQuick, Scan-Dia, Hagen, Germany). All specimens were polished under running water with silicium carbide paper (SiC) from P80 up to P1200 (Struers) for 10 s each (Tegramin-20, Struers). Before pretreatment, the specimens were abraded for 10 s with airborne particles of alumina oxide (Basic quattro 1S, Hilzingen, Germany) with a mean powder size of 50 μm at an angle of 45° and a distance of 10 mm and subsequently cleaned in an ultrasonic bath with distilled water for 5 min. Then specimens were divided into three groups for determination of bond strength for all combinations of bonding agents and resin cements on the basis of the shear bond strength method (SBS, n=180), the tensile bond strength method (TBS, n=180) and the work of adhesion (WA, n=60), which was theoretically calculated. Pretreatments were performed as described in the following paragraph (n=45 per group): 1. VP connect was applied with a micro-brush as a thin layer and air dried for 180 s. 2. Visio.link was applied with a micro-brush as a thin layer and light cured for 90 s with a manufacturerrecommended light unit (bre.Lux Power Unit, bredent, Senden, Germany). 3. Clearfil Ceramic Primer was applied with a micro-brush as a thin layer and allowed to vaporize completely. 4. CAD/CAM resin with no further pretreatment served as the control group. Manufacturers, composition and LOT numbers of all materials used are listed in Table 1. Each pretreatment group was subdivided according to the above-listed resin composite
Clin Oral Invest Table 1
Summary of materials used in the present study, their manufacturer with LOT number, and their composition
CAD/CAM resin Pretreatment method
XHIPC-CAD/CAM blank VP connect
Xplus3, Echzell, Germany
Merz Dental, Lütjenburg, Germany bredent, Senden, Germany Kuraray Med., Sakazu, Okayama, Japan Kuraray Med., Sakazu, Okayama, Japan
50–80 %: PMMA, 10–20 %: UDMA, BDDMA, mutli-methacrylate, 5–15 % filler MMA
visio.link Clearfil Ceramic Primer Clearfil SA Cement
3M ESPE, Seefeld, Germany
Ivoclar Vivadent, Schaan, Liechtenstein
Base: R35481 Catalyst: P84939
MMA, dimethacrylate PETIA, photoinitiators 3-Methacryloxypropyl trimethoxy silane, MDP, ethanol Paste A: MDP, Bis-GMA, TEGDMA, dimethylacrylate, Ba-Al fluorosilicate glass, SiO2, benzoyl peroxide, initiators Paste B: Bis-GMA, dimethacrylate, Ba-Al fluorosilicate glass, SiO2, pigments Bis-GMA, TEGDMA amine, photoinitiator system (CQ), benzoyl peroxide and stabilizers Bis-GMA, TEGDMA, UDMA, benzoylperoxide, inorganic fillers, ytterbium trifluoride, Ba-Al fluorosilicate glass, spheroid mixed oxide, initiator, stabilizers, pigments
TEGDMA triethylenglycol-dimethacrylate, MMA methyl methacrylate, BDDMA buthanediol dimethacrylat, Bis-GMA bisphenol-Adiglycidylmethacrylate, UDMA urethane dimethacrylate, MDP 10-methacryloyloxydecyl dihydrogen phosphate, PETIA pentaerythritol triacrylate, PMMA polymethylmethacrylate
cements (n=15 per group). Polymerization of resin composite cements (SBS and TBS tests) was performed on two sides of the acrylic cylinder for 20 s per side, making 40 s in total (Elipar S 10, 3M ESPE, Seefeld, Germany). Immediately before the polymerization, the intensity of the LED light-curing unit was measured using an analysis device (Marc V3, BlueLight Analysics Inc., Halifax, NS, USA). The LED lamp had a light intensity of 1200 mW/cm2. After the cementation, all specimens were stored in distilled water in an incubator at 37 °C for 24 h (HERA cell 150 Thermo Scientific, Heraeus, Hanau, Germany) and then artificially aged for 5000 cycles of thermal ageing (Thermocycler THE 1100, SD Mechatronik, Feldkirchen-Westerham, Germany) between 5 and 55 °C with a dwell time of 20 s. Before the SBS and TBS tests, specimens were immersed in distilled water for 1 h at room temperature (23 °C). SBS and TBS were calculated according to the following equation: σ [N/mm2] = F/A (where σ: shear or tensile bond strength, F: load at fracture [N] and A: adhesive area [mm2]). Shear bond strength Resin composite cement was inserted in an acrylic cylinder (SD Mechatronik, Feldkirchen-Westerham, Germany) with an inner diameter of 2.9 mm that was placed centrally on the CAD/CAM material surface. To obtain a standardized and homogeneous cemented layer with a height of 0.5 mm, a screw with an outer diameter of 2.8 mm was driven into the
core of the acrylic cylinder and loaded with 1 N. Excess resin cement could exit through the screw thread and was cleaned off carefully. Polymerization and artificial ageing were performed as described. For testing, the specimens were fixed in a Universal Testing Machine (Zwick 1445, Zwick, Ulm, Germany) with the CAD/CAM resin surface parallel to the loading direction and the acrylic cylinder in the horizontal direction and vertically loaded until fracture (1 mm/min) (Fig. 1a) .
Tensile bond strength An acrylic cylinder (SD Mechatronik) with an inner diameter of 2.9 mm was positioned on the pretreated CAD/CAM resin. The resin composite cement was manually filled into the acrylic cylinder ensuring a porosity-free consistency. Excess cement was carefully removed. Polymerization and artificially ageing were performed as described. For testing, the specimens were fixed in a special holding device, ensuring an axial moment-free force application in the Universal Testing Machine (Zwick 1445). The acrylic cylinder was held by a collet while an alignment jig allowed self-alignment of the specimens. The device was installed in the load cell of the Universal Testing Machine and pulled apart by an upper chain, guaranteeing a self-centring of the whole system. The TBS was measured by pulling axially with a constant crosshead speed of 5 mm/min until the specimens disconnected (Fig. 1b) .
Clin Oral Invest Fig. 1 Design of test methods: a SBS, b TBS and c WA
Fracture type analyses After obtaining the SBS and TBS measurements, the failure type analyses were performed. The failure types were analysed for adhesive, cohesive and mixed fracture types (Fig. 2). The adhesive type was defined as fracture in the bonding area (interface), whereas the cohesive fracture was distinguished as fractures of the tested CAD/CAM resin or otherwise of the cement. The mixed failure was used to describe more types of fractures (cohesive and adhesive) in one specimen. All failure types were evaluated by two calibrated examiners, who were unaware of the group allocation and
treatment, under an optical microscope (Axioskop 2 MAT, Carl Zeiss Microscopy, Göttingen, Germany). Work of adhesion measurements To evaluate the theoretical work of adhesion, the surface of the pretreated CAD/CAM resin, and the surface of all three nonpolymerized resin composite cements (n=15 per group) were analysed. The sessile drop technique  was used to perform the contact angle measurement. The measurements were accomplished in a contact angle meter (EasyDrop, Krüss, Hamburg, Germany) using two micro-syringes, one filled with
Fig. 2 Failure types analyses: adhesive failure, cohesive failure among resin composite cement, cohesive failure among substrate and mixed failure (cohesive and adhesive at the same time) left to right
Clin Oral Invest
distilled water and the other with diiodomethane (99 %; Cat: 15.842-9, Sigma-Aldrich, Steinheim, Germany, LOT No.: S65447–448) as polar and disperse fluids at room temperature (23 °C). An attached digital camera registered the applied fluid drops with a known volume (10 μl of water and 5 μl of diiodomethane) after exactly 1 s. With a special computer program (DSA4, Krüss), the height and diameter of each drop were measured, and therefore, the static contact angle was determined using two different computation methods (Fig. 1c). For flat angles of diiodomethane, the circle method was chosen. The contact angle for distilled water was determined with the tangent 1 method. Three drops of each liquid were placed on each specimen to calculate a mean contact angle for each. Based on the formula of Owens, Wendt, Rabel and Kaeble, the computer program determined the surface free energies (SFE) of the CAD/CAM resin in combination with all pretreatments as well as those of the resin cements [30, 31]. All SFE results were divided into their polar and disperse shares. sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ ð1 þ cosθÞ ⋅ S FE L S FE PL þ S FE D qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ ¼ S FE PS S D S FE D L 2 ⋅ S FE L
SFEPL SFEPS SFED L SFED S θ
Surface free energy of the liquid polar component Surface free energy of the solid polar component Surface free energy of the liquid dispersive component Surface free energy of the solid dispersive component Contact angle
Subsequently, the WA between the CAD/CAM resin and the non-polymerized resin composite cement was calculated using the sum of the polar and disperse shares of the SFE of the bonding agents (BA) after application on the CAD/CAM resin as well as on the resin composite cements (RC). These results were put into a formula to determine the work of adhesion. qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ D W A ¼ 2 ⋅ S FED ⋅ S FE þ 2 ⋅ S FEPBA ⋅ S FEPRC BA RC Further formulas were used to calculate the interfacial tension (IFT) and the spreading coefficient (SC). I FT ¼ S FEBA þ S FERC − 2 ⋅
qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ D S FEPBA ⋅ S FEPRC S FED BA ⋅ S FE RC − 2 ⋅
Statistical analysis The data were analysed using descriptive statistics. The normality of data distribution was tested using the Kolmogorov– Smirnov and Shapiro–Wilk tests. The differences between the groups were determined using the Mann–Whitney U and Kruskal–Wallis H tests (SPSS V20, SPSS Inc., Chicago, IL, USA). The association between fracture type and pretreatment was investigated by a Chi2 test. In addition, the relative frequencies of fracture types, together with the corresponding 95 % CI, were given using the Ciba−Geigy table . The correlation between all parameters of all used tests used was non-parametrically analysed with Spearman’s Rho test. All results for statistical analyses with p values below p=0.05 were considered to be statistically significant.
Results Tables 2 and 3 show the descriptive statistics for all test methods for each resin composite cement and each pretreatment, separately. The results are depicted in Fig. 3. The Kolmogorov–Smirnov and Shapiro tests indicated that SBS, TBS and WA groups were not normally distributed. Hence, nonparametric statistical analyses were performed. SBS results The control group and groups pretreated with Clearfil Ceramic Primer, regardless of resin composite cement, as well as VP connect combined with Clearfil SA Cement, showed no bond to the CAD/CAM resin (Table 2). Among all resin composite cements, the highest bond strength (p