journal of the mechanical behavior of biomedical materials 29 (2014) 427–437
Available online at www.sciencedirect.com
www.elsevier.com/locate/jmbbm
Research Paper
Evaluation of the interfacial work of fracture of glass-ionomer cements bonded to dentin Joshua J. Cheethama,n, Joseph E.A. Palamarab, Martin J. Tyasc, Michael F. Burrowd a
Melbourne Dental School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Australia Melbourne Dental School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Australia c Honorary Professorial Fellow, Melbourne Dental School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Australia d Clinical Associate Professor Faculty of Dentistry, The University of Hong Kong, People's Republic of China b
art i cle i nfo
ab st rac t
Article history:
Objective: The aim of this study was to investigate the interfacial work of fracture of
Received 16 July 2013
conventional (C-) and resin-modified (RM-) glass-ionomer cements (GICs) bonded to dentin.
Received in revised form
Methods: One hundred and sixty five aries-free human molars were embedded in epoxy
8 September 2013
resin, sectioned and polished with 300- and 600- grit silicon carbide paper to remove
Accepted 15 September 2013
enamel on the occlusal surface. Equilateral triangular-shaped plastic molds (4 4 4
Available online 10 October 2013
5 mm4) were clamped to the prepared dentin surfaces by a stainless steel test apparatus.
Keywords:
Teflon tape was placed under one internal vertex of the mold to create a 0.1-mm notch at
Work of fracture
the material-dentin interface. Interfacial work of fracture (γwofint) in tensile fracture mode-I
Interfacial work of fracture
(opening) was determined for six C-GIC, three RM-GIC, and two GIC luting cements at a
Resin-modified glass-ionomer
cross-head speed of 0.1 mm/min and a crosshead distance (L) from the interface of 4.3 mm.
Conventional glass-ionomer
The debonded surfaces were evaluated for the predominant failure mode. SEM analysis of examples showing interfacial and notch areas was performed. Results: ANOVA and Tukey's post hoc test demonstrated the highest mean γwofint value (90.16716.6 J/m2) of one RM-GIC was significantly different (po0.05) from the other materials. ‘High viscosity’ GICs achieved lower results with the lowest recorded at 20.4710.1 J/m2. There was a significant difference observed (po0.05) between the mean γwofint of luting C-GIC and luting RM-GIC. Although differences were observed between different material mean γwofint, when comparing groups no significant differences (p ¼0.181) were observed. For all groups, mixed GIC-interface failure (41%) was the most commonly observed, followed by cohesive failure in GIC (25%) and adhesive failure (20%). SEM analysis revealed that specimens generally fractured from the notch initiation point into the GIC or along the dentin–GIC interface. Conclusion: Within the limits of this study, significant differences (po0.05) were observed in the γwofint between different glass-ionomer materials. The null hypothesis that there is no difference in the γwofint among different glass-ionomer materials bonded to human dentin was rejected.
n Correspondence to: Melbourne Dental School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne 720 Swanston Street, Victoria, 3010 Australia. Tel.: þ61 3 93411532; fax: þ61 3 93411599. E-mail address: [email protected] (J.J. Cheetham).
1751-6161/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jmbbm.2013.09.020
428
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Relevance: In the current study, the interfacial work of fracture (γwofint) of glass-ionomer adhesive interfaces has been reported using a simple method that can be used to study the fracture mechanics of an adhesive interface without the need for complicated specimen preparation. & 2013 Elsevier Ltd. All rights reserved.
1.
Introduction
Variables such as test rig, biological substrate, dentin and enamel type and position in the tooth, storage of teeth, affect bond strength tests (Heintze, 2013). However the bond tests used are simple and easy to use, so they are still commonly reported in an effort to demonstrate the efficacy and ranking of adhesive dental materials. Comparing bond strength test results from different studies is not recommended as test conditions are invariably different among the various tests (Kelly et al., 2012; Scherrer et al., 2010; VanNoort, 1989). Several authors have questioned the validity of these tests and criticized the lack of standard protocols and relevance to clinical performance (Della Bona and Watts, 2013; Heintze, 2013; Stephen, 2012). Researchers have also suggested that a fracture mechanics approach is more appropriate than conventional shear or tensile bond strength tests (Salz and Bock, 2010). Using a fracture mechanics approach, a crack is introduced into the bond interface and the system's strength to resist crack propagation across the adhesive interface has also been investigated (Hashimoto, 2010; Salz and Bock, 2010). Critical stress intensity factor (KC), describes the ability to resits crack propagation. Linear elastic fracture toughness (KIC) studies the stress region ahead of crack propagation in tensile (mode – I) failure (Soderholm, 2010). Various methods have been employed to investigate interfacial fracture toughness of adhesives to teeth and biomaterials to determine the work of fracture which has been abbreviated as Wf andWi, plane strain interfacial fracture toughness (KICint), the adhesive (elastic-plastic) fracture energy (JIC) and the critical plane strain energy release rate (GICint) (Armstrong et al., 1998, 2001; Barker, 1977; Cheng et al., 1999; De Munck et al., 2013; Della Bona et al., 2006; Howard and Söderholm, 2010; Jancar, 2011; Lin and Douglas, 1994; Rasmussen, 1984; Rasmussen and Patchin, 1984; Tam and Pilliar, 1993, 1994; Tam and Yim, 1997; Toparli and Aksoy, 1998; Walshaw et al., 2003). Studies have also been performed on bone cements, which show adhesive failure occurring due to stress cracking at some point in their lifecycle (Lucksanasombool et al., 2003; Tong, 2006; Tong et al., 2007; Wang and Pilliar, 1989; Wang and Agrawal, 1997, 2000; Wang et al., 1994). Interfacial fracture toughness has also been measured for glass-ionomer materials (Akinmade and Hill, 1992; Mitsuhashi et al., 2003; Setien et al., 2005; Tam et al., 1995). Furthermore, a series of reviews on bond tests have recommended that a fracture mechanics approach be revisited as the preferred test method for adhesive strength evaluation (De Munck et al., 2005; Kinloch, 1979; Salz and Bock, 2010; Scherrer et al., 2010; Soderholm, 2010). A contributing factor as to why adhesive interfacial fracture toughness tests have not been commonly reported is
because current tests are complicated and often require specialized apparatus to prepare the dentin and dental material into the desired configuration. Attempts have been made to develop less complex systems, including adaptations of a common shear bond strength test using triangular shaped adhesive areas, (Cheng et al., 1999; Tantbirojn et al., 2000) and a notchless triangular prism specimen developed by Ruse et al. (Ruse et al., 1996). There is currently no available standardized test method for determination of interfacial fracture toughness properties of adhesive dental materials. Tattersall and Tappin introduced a simple method to determine the work of fracture of materials using a specimen with a square cross section and triangular fracture surface (Tattersall and Tappin, 1966). Further work by Rasmussen et al. demonstrated a test method to study the fracture properties of enamel and dentin, and determined the work of fracture (Wf or γwof), calculated by dividing the total energy (J) required to initiate fracture by twice the surface area (m2) of a triangular-shaped fractured surface (Rasmussen, 1984; Rasmussen and Patchin, 1984; Rasmussen et al., 1976). The test method was also adapted to investigate interfacial work of fracture (Wi or γwofint) for porcelain-gold and enamelcomposite adhesion (Rasmussen, 1978). Subsequent studies have also investigated fracture mechanics of different materials and interfaces using a γwof approach (Sakai and Bradt, 1993). The aim of this study was to compare the interfacial work of fracture (γwofint) and failure modes of several glassionomer cements bonded to dentin, using a new simplified test method. The null hypothesis is that there is no difference in the interfacial work of fracture (γwofint) among different glass-ionomer cement materials bonded to human dentin.
2.
Materials and methods
2.1.
Teeth preparation
Human ethics approval (♯1033315.1) for the use of human teeth was obtained from the University of Melbourne. Onehundred and sixty-five caries-free human molars were selected from a tooth bank. No information on the age of teeth was available. Teeth were stored in a refrigerated 0.5% chloramine T solution and used within six months of collection date. This method of storage followed guidelines described in ISO TS 11405 Dental materials – testing of adhesion to tooth structure. The teeth were cleaned with a slow-speed prophylaxis polisher (Zen; Philips, CA) and wet pumice, rinsed and stored in de-ionised water for approximately 24 h prior to embedding. Three stainless steel mold
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
sets were used to embed the teeth in epoxy resin, each containing 30 cylindrical mold cavities measuring 20 mm diameter x 30 mm high. To ensure the plates could be separated, low-density polyethylene film (GLADs WRAP; Glads Products Australia, Padstow, NSW) was placed between the lower plate and upper plate. Teeth were dried with tissue paper and a small amount of permanently plastic adhesive (BluTak; Bostik Australia Pty Ltd., Thomastown, Vic) was placed on the occlusal surface to maintain the tooth position within the mold cavity. Cold-cure epoxy casting resin (105 Epoxy resin, West System, MI) was placed into each cavity and allowed to cure. Mold sets were separated after 8 h and specimens were removed by a press and stored in deionised water at 37 1C for 24 h. Specimens were placed in a slow speed sectioning saw (ISOMETs 1000 precision sectioning saw; Buhler, Lake Bluff, IL) fitted with a 125-mm diameter, 0.4-mm thick diamond-coated blade and sectioned under water until enamel on the occlusal surface was removed. Exposed dentin surfaces were – ground under water at 300 rpm and 2 kg-f on a polishing machine (Phoenix Beta; Buehler, Lake Bluff, IL) with 300-grit and 600-grit silicon carbide paper. Specimens were stored in deionised water at room temperature prior to placement of the GIC.
2.2.
Fracture specimen preparation
A triangular plastics mold (Fig. 1) measuring 4 4 4 5 mm4 was placed onto the dentin surfaces to replicate the adhesive interface dimensions used by Ruse et al. (Ruse et al., 1996).
429
Apparatus to prepare the specimens for testing (Fig. 2a) consisted of a top plate connected with three springs to a base plate. The mold and specimen was inserted into the apparatus (Fig. 2a). Teflon tape (0.05 mm 15 mm) was placed under one corner of the mold to provide a ‘notch’ in the GIC cement (Fig. 2b) and stereo-microscopic examination at magnification 130 (M205C; Leica Microsystems GmbH, Wetzlar, Germany) using linear measuring software (LAS Application suite; Leica Microsystems GmbH) was used to achieve a distance from the edge of the tape 0.1 mm to the internal vertex of the mold (Fig. 2c). Names and composition of materials used are listed in Table 1. Although the smear layer was not removed from the prepared surfaces, surface conditioning of the dentin surface was completed prior to placement of the mold if recommended by the manufacturer (Table 1). Encapsulated materials were mixed in a capsule mixer (Ultramat 2; SDI Limited, Bayswater, Australia) as per manufacturer's instructions, inserted directly into the mold and lightly compacted with a dental plugger instrument to ensure adequate filling of the mold. The apparatus shown in Fig. 2a was placed in covered plastics boxes containing approximately 10 mm depth of deionised water and stored at 37 1C for 1 h to allow initial cement setting to occur. Specimens with the attached mold were removed from the apparatus, and immersed in water and stored at 37 1C for a further 23 h prior to determining γwofint.
2.3.
De-bonding procedure
The Teflon tape was removed and specimens placed into a fixture mounted on a universal testing machine (Model 8872; Instron Ltd, High Wycombe, England) connected to a 500 N load cell. Dentin surfaces were aligned with the vertical surface of the horizontal housing (Fig. 2d). A 0.7-mm flat edge tool attached to the load cell was placed directly against the surface of the housing achieving a distance “L” of 4.3 mm from the dentinmold interface (Fig 2 f). The flat edge was brought to within 0.5 mm of the mold (Fig. 2e) and crosshead speed was set to 0.1 mm per minute until the specimen was broken. The total area under the load-extension curve, representing the total energy (J) to initiate fracture at the interface was recorded for each specimen. Compliance was tested on the surface of samples for deformation and subtracted from the displacement measured for each specimen in order to determine the final total energy for each specimen.
2.4.
Determination of the interfacial work of fracture (γwofint)
The following formula was used to determine interfacial γwof (Rasmussen, 1978; Rasmussen, 1984; Rasmussen et al., 1976; Sakai and Bradt, 1993; Tattersall and Tappin, 1966): γ wof int ¼ Wwof =ð2AÞ where
Fig. 1 – Plastics mold dimensions, including outside edge, which is clamped down by the test apparatus. All dimensions are in mm.
γwofint is the interfacial work of fracture (J/m2) Wwof is the total energy recorded to initiate the fracture process (J) A is the surface area of the fractured surface (m2)
430
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Fig. 2 – Experimental apparatus. (a) Stainless steel apparatus. Mold held down into position by the spring loaded upper plate. (b) Location of the Teflon tape under the mold vertex prior to cement placement. Note apparatus has been removed for demonstration. (c) Fine adjustment of the Teflon tape position prior to cement placement (bar¼200 lm). (d) Specimen placed in horizontal housing showing flat edge tool position against housing. (e) Alignment of the flat edge tool 0.5 mm from the mold surface prior to testing. (f) Contact point of flat edge tool onto the mold surface. (g) Free body diagram demonstrating the loading configuaration on the mold.
2.5.
Failure analysis
Stereo-microscopic analysis at 20 magnification (M205C; Leica Microsystems GmbH) was used to report the predominant model of failure of each sample as: (a) cohesive in GIC; (b) mixed GIC-interface; (c) interface (or adhesive); and (d) cohesive in dentin. Scanning electron microscope (SEM) (Quanta 200 ESEM; FEI Company, Hillsboro, OR) images were taken of a representative number of uncoated samples at low vacuum with magnifications of up to 6000 to characterize the interfacial dentin and material fracture sites.
2.6.
Statistical analysis
ANOVA and post hoc Tukey tests were used (PASW Statistics V18, SPSS INC, Chicago, IL) to compare γwofint of each material bonded to dentin and differences between each material group, at a significance level of po0.05.
3.
Results
3.1.
Interfacial work of fracture (γwofint)
Mean γwofint were normally distributed for all materials, as assessed by Shapiro–Wilks test (p40.05). The mean γwofint are shown in Table 2 and Fig. 3. A total of 23 samples debonded prior to testing. Significant differences were observed between materials and Tukey's post hoc test revealed four homogenous subsets identified in Table 2. The highest mean value was observed with one visible light cured RM-GIC (Fuji II LC capsules) which was significantly different (po0.05) than other material mean γwofint results. No significant differences were observed between all other light cured RM-GIC mean γwofint, with the lowest value of 46.2 (720.9) J/m2 recorded for Ketac Nano. The group of C-GIC restorative materials recorded significantly different lower values when compared to Fuji II LC. Chemfil Molar and Ketac Molar recorded lower mean γwofint which were significantly different than two other C-GIC materials (Fuji IX GP Extra and Fuji IX GP Fast). GIC (Chemfil
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
431
Table 1 – Cement composition. Type
Product Name
Manufacturer
Lot number
Composition
RMGICa
Fuji II LC (Improved) capsules
GC Corporation, Tokyo, Japan
1008091
Powder: alumino-fluoro-silicate glass (amorphous) (95–100%) Liquid: polyacrylic acid (20–25%), 2-hydroxymethyl methacrylate (30– 35%), 2,2,4, trimethyl hexamethylene dicarbonate (1–5%), proprietary ingredient (5–15%) Powder:liquid ¼ 3.2:1 GC cavity conditioner: Polyacrylic acid (20%) Distilled water (77%), aluminum chloride hydrate (3%), food additive Blue no. 1 (o0.1%)
RMGIC
Photac Fil Quick
3 M/ESPE, Seefeld, Germany
411820
Powder: silane treated glass powder (499%), N,N-dimethylbenzocaine (o0.5%) Liquid: copolymer of acrylic acid-maleic acid (30–50%), HEMA (25–50%), water (20–30%), mono and di-HEMA phosphate, magnesium salt (5– 15%), diurethane dimethacrylate (3–10%) Powder:liquid ¼ 3.1:1
RMGIC
Ketac Nano
3 M/ESPE, Seefeld, Germany
N156309
Paste A: silane treated glass (40–55%), silane treated zirconia (20–30%), PEGDMA (5–15%), silane treated silica (5–15%), HEMA (1–15%), glass powder o5%, BisGMA o5%, TEGMA o1% Paste B: silane treated ceramic (40–60%), copolymer of acrylic and itaconic acid (20–30%), water (10–20%), HEMA (1–10%) Ketac Nano Glass Ionomer Primer: Water (40–50%) Hema (35–40%), copolymer of acrylic and itaconic acids (10–15%)
C-GICb
GC Fuji IXTM GP Extra
GC Corporation, Tokyo, Japan
1005261
Powder: fluoro aluminosilicate glass (amorphous) (90–100%). Polyacrylic acid (5–10%) Liquid: polyacrylic acid (30–40%), proprietary ingredient (5–15%) Powder:liquid ¼ 3.4:1 GC Cavity conditioner: Polyacrylic acid (20%), distilled water (77%), aluminum chloride hydrate (3%), food additive Blue no. 1 (o0.1%)
C-GIC
GC Fuji IXTM GP Fast
GC Corporation, Tokyo, Japan
1003051
Powder: fluoro aluminosilicate glass (amorphous) (90–100%). Polyacrylic acid (5–10%) Liquid: polyacrylic acid (30–40%), proprietary ingredient (5–15%) Powder:liquid ¼ 3.7:1 GC cavity conditioner: polyacrylic acid (20%), distilled water (77%), aluminum chloride hydrate (3%), food additive Blue no. 1 (o0.1%)
C-GIC
ChemFil Rock
Dentsply DeTrey, Konstanz, Germany
108001360
Powder: glass powder, polycarboxylic acid (10–25%) Liquid: polycarboxylic acid (10–25%), tartaric acid (2.5–10%), water Powder: Liquid ¼ ¼3.7: 1
C-GIC
ChemFil Molar
Dentsply DeTrey, Konstanz, Germany
0912000748
Powder: glass powder, polyacrylic acid (2.5–10%), tartaric acid (2.5–10%), silicon dioxide (2.5–10%) Liquid: water, polyacrylic acid (50–100%) Powder:liquid ¼ 3.7:1
C-GIC
Ketac Molar Aplicap
3M/ESPE, Seefeld, Germany
0912000748
Powder: glass powder (80–90%), copolymer of acrylic acid-maleic acid (1–6%) Liquid: water (40–45%), copolymer of acrylic acid-maleic acid (35–55%), tartaric acid (5–10%) Powder:liquid ¼ 3.4:1 Ketac Conditioner: Water (70–80%) polyacrylic acid (20–30%)
432
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Table 1 (continued ) Type
Product Name
Manufacturer
Lot number
Composition
C-GIC
Ketac Molar Quick Aplicap
3M/ESPE, Seefeld, Germany
471469
Powder: glass powder (93–98%), dichloromethylsilane reaction product with silica (0–1%), copolymer of acrylic and maleic acid (1–5%). Liquid: water (60–65%), copolymer of acrylic acid-maleic acid (30–40%), tartaric acid (5–10%) Powder:liquid ¼ 3.1:1 Ketac conditioner: Water (70–80%), polyacrylic acid (20–30%)
Luting C-GIC
GC Fuji Is
GC Corporation, Tokyo, Japan
1002101
Powder: alumino-fluoro-silicate glass (amorphous) (95%), polyacrylic acid (5%) Liquid: distilled water (50–55%), polyacrylic acid (30–40%) Powder:liquid ¼ 1.8:1
Luting RMGIC
GC Fuji Plus, capsules
GC Corporation, Tokyo, Japan
1003191
Liquid: polyacrylic acid, HEMA, UDMA Powder: alumino-silicate glass Powder:liquid ¼ 2:1 GC Fuji Plus conditioner: citric acid (10%), Distilled water (87%), iron (III) chloride (ferric chloride) (3%), food additive Blue no.1 (o0.1%)
Note: Primers and conditioners were applied to dentin according to manufacturers' instructions for use. a C-GIC ¼conventional glass-ionomer cement. b RM-GIC ¼resin-modified glass-ionomer cement.
Table 2 – Mean γwofint (J/m2), standard deviations (J/m2) and sample size. Group Fuji II LC Photac FIL QUICk Ketac Nano Fuji IX GP Extra Fuji IX GP Fast Chemfil Rock Chemfil Molar Ketac Molar Ketac Molar Quick Fuji I Fuji Plus
Mean γwofint (SD) 90.2 50.0 46.2 54.6 44.3 41.0 20.4 28.9 38.6 33.1 49.9
a
(16.6) (14.9)b,c (20.9)b,c (12.1)b (10.2)b,c (13.1)b,c,d (10.1)d (18.3)c,d (9.6)b,c,d,e (11.0)b,c,d (35.2)b,c
n 10 14 13 15 14 13 15 13 15 12 8
Mean γwofint with the same superscript are not significantly different (p40.05).
Molar) recorded the lowest value of 20.38 (710.1) J/m2, which was four times lower than the highest results recorded for RM-GIC. The two luting materials had similar means (p40.05). Although differences were observed between materials, when comparing groups no significant differences (p¼ 0.181) were observed between mean γwofint. The predominant modes of failure are shown in Table 3. For all groups, mixed GIC-interface failure (41%) was the most frequent one observed, followed by cohesive failure in GIC (25%) and interfacial (adhesive) failure (20%). Dentin pullout (cohesive failure) was not observed in any specimens. Dentin pullout is a significant energy dissipating mechanism in resin-based dental composites. The absence of dentin pullout
indicates that RM-GIC may not have penetrated the dentin tubules in a similar way demonstrated by solvent based adhesives. Ketac Nano predominantly failed at the interface (i.e., adhesive failure), whereas Fuji II LC had a similar number of failures for each mode. C-GIC tended to fail cohesively in the GIC and in mixed mode, whereas RM-GIC failed more frequently in mixed or interface (adhesive) modes. SEM analysis (Figs. 4, 5) revealed that specimens generally fractured from the notch initiation point into the GIC material or along the dentin–GIC interface. There were no instances of dentin “pull-out”. The region near the notched crack tip for Fuji II LC showed fracture occurring from the notch into the material and then down a plane parallel to the dentin surface. Lower mean γwofint compared to Fuji II LC was recorded for Ketac Nano, and images showed that the fracture that initiated at the notch typically followed along the dentin–GIC interface. The material's physical appearance at the notch area was observed to be different, for example the luting RM-GIC (Fuji Plus) and Ketac Nano did not show the larger particles which were present in other GICs at the notch area.
4.
Discussion
It is important to optimize the “adhesive” strength, including durability, of dental materials to provide “safe and effective products”(Della Bona and Watts, 2012). In the current study, the γwofint of glass-ionomer–dentin interfaces is reported, which provide a method that can be used to study the fracture mechanics of an adhesive interface. This study investigated a simple mold and apparatus that can be used
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
433
Fig. 3 – Comparison of the γwofint results (J/m2) for each glass-ionomer bonded to dentin.
Table 3 – Predominant mode of failure. Group
Cohesive in GIC
Mixed GIC-interface
Interface (adhesive)
De-bond prior to test
Fuji II LC Photac Fil Quick Ketac Nano Fuji IX GP Extra Fuji IX GP Fast Chemfil Rock Chemfil Molar Ketac Molar Ketac Molar Quick Fuji I Fuji Plus
3 2 0 9 5 3 4 4 6 4 2
3 7 2 6 9 8 11 6 7 6 2
4 5 11 0 0 2 0 3 2 2 4
5 1 2 0 1 2 0 2 0 3 7
Total n (%)
42 (25%)
67 (41%)
33 (20%)
23 (14%)
Note: no instances of cohesive in dentin failure mode were observed.
to determine the γwofint of dental materials without the need for complicated specimen shaping and preparation. The validity of bond strength tests has been questioned since their introduction (VanNoort, 1989). Scherrer (Scherrer et al., 2010) and others (Heintze, 2013; Soderholm, 2010; Tantbirojn et al., 2000) have recognized the deficiencies in current bond strength test methods, including fracture toughness methods previously reported (Lin and Douglas, 1994; Soderholm, 2010; Tong, 2006). Tattersall and Tappin developed a method of studying the work of fracture of monolithic materials such as ceramics, metals and plastics (Tattersall and Tappin, 1966). It involved simple specimen geometry, and fracture began at the notched “v” shaped section. This was further adapted for dentin and enamel fracture studies (Rasmussen, 1984; Rasmussen and Patchin, 1984; Rasmussen et al., 1976) and
interfacial adhesion (Rasmussen, 1978), however these tests also required detailed specimen preparation to achieve the triangular-shaped adhesive surface which is not required in the current test method. Although investigating adhesion of composite to enamel and using three point bending, Rasmussen (Rasmussen, 1978) found average γwofint results to be between 61 and 73 J/m2 which were higher than all of the mean γwofint results reported in the current study except for one RM-GIC (Fuji II LC). Other tests (Rasmussen et al., 1976) revealed average work of fracture of dentin to be in the range of 270–550 J/m2 which is well above results reported in the current study and could help to explain why there were no instances of cohesive failure in dentin. An advantage of the work of fracture technique is the simplicity in the calculation that is used, compared to other test methods. No other material properties used in other
434
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Fig. 4 – SEM images examples of fractured interfaces. (a) Overview of RM-GIC (Ketac Nano) mold and corresponding dentin surface and (b) at low magnification showing the full surface. Failure occurred at the cement–dentin interface. Note areas of GIC pullout that did not cause cohesive failure in the material. (c) Notch area (N) of nanoparticle containing RM-GIC (Ketac Nano) sample showing an absence of large particles. Failure occurred from the notched area and continued towards the perimeter of the specimen. (d) Dentin surface of C-GIC (Ketac Molar) cohesive failure at the notched area. Failure initially occurred along the dentin–GIC interface prior to moving from a mixed to cohesive GIC failure mode. (e) C-GIC specimen notch area. The specimens' filler particles provide sites for crack propagation due to their irregular shape. Note the presence of large cylindrical particles. (f) Example of spherical voids present in C-GIC surfaces after testing. (g) C-GIC specimen showing fracture along large filler particles towards a large void caused by filler pull out. (h) C-GIC (Fuji IX GP Extra) notch area (bar¼20 lm) demonstrating the inhomogeneity of material appearance in the area. Failure has occurred from this point under the large particles. (i) C-GIC (Fuji IX GP Fast) notch area showing similar material topography to Fuji IX GP Extra. Note the multiple fracture points and large particles. The top region demonstrates the non-bonded material notch area on the GIC side remains during the test. It is important to note that some images show artificial “cracks” in the GIC material due to desiccation cause by SEM vacuum environment.
Fig. 5 – SEM images of the dentine surface shows the predominant mode of failure under low magnification (a) Interfacial. (b) Cohesive in GIC. (c) Mixed GIC-interface.
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
tests, such as a materials modulus of elasticity are required for this test (Cheng et al., 1999; Sakai and Bradt, 1993; Soderholm, 2010; Tantbirojn et al., 2000; Tattersall and Tappin, 1966). Fig. 7 shows an example of a load-displacement curve with and without compliance correction which is an important consideration when determining work of fracture. The method described in this experiment has further advantages over current methods of investigating interfacial work of fracture material properties. Firstly, a standard triangular mold is used for placing the dental material. Secondly, preparation of the dentin requires only surface sectioning whereas other methods require intricate specimen preparation and complicated test apparatus which make it more difficult to achieve large sample sizes in a reasonable amount of time (Soderholm, 2010). Thirdly, the crack initiated by the tape at a vertex in the mold and the adhesive interface perimeter is not affected by grinding or cutting, thus preventing the introduction of further flaws at these critical points. Lastly, the test apparatus can be easily adapted for users that have facilities for bond strength testing and if required modifications (such as crack length and L) can be made easily. Fatigue testing, thermocycling and aging procedures could also be applied to the test specimens prior to the final fracture test to provide more relevant performance data. Table 3 details the number of specimens that de-bonded prior to testing. Unlike other interfacial fracture toughness tests, the specimens in this study remained in the mold. These specimens typically failed when the embedded tooth/ mold assembly was removed from the apparatus for immersion in water, as described in Fig. 1a. The failure was caused by operator error and/or incomplete filling during the material insertion process. Cement failure was observed in most of the materials; however this was more evident in the luting RM-GIC possibly due to air entrapment in this low viscosity cement preventing compaction by an instrument. Compaction of all materials in each group was manually achieved
Fig. 6 – Example of notched area SEM image after a predominantly interfacial break (bar¼500 lm).
435
Fig. 7 – Example of load – displacement curves with (line A) and without (line B) compliance correction.
and variation in this compaction method could have influenced every groups results. Similar to other interfacial fracture toughness tests there was diversity in the predominant failure mode of each material (Salz and Bock, 2010; Setien et al., 2005; Soderholm, 2010). Fracture could not be expected to proceed in a predictable direction, because of the complexity of material structure at the interface (Fig. 4g, h, i) and inconsistencies such as voids (Fig. 4f) present in the GIC material, which could contribute to variation in results. It is important to obtain a uniform notch area. Fig. 6 shows a typical notched area of a RM-GIC (Ketac Nano) that exhibited predominant interfacial debonding during the test. Unlike the C-GIC materials in the group, hardening of light-cured RM-GIC occurs rapidly after a short light-curing period due to the free radical polymerization reaction. The resin component of the RM-GIC could have contributed to a higher resistance to work of fracture compared to the C-GIC acid–base complex at 24 h after cement placement. Ketac Nano is supplied in a paste–paste form compared to capsule delivery for Fuji II LC and Photac-Fil. A benefit of paste systems is a reduction of entrapped air in the final paste compared to mixed capsule systems, which introduces porosity during mixing (Boehm et al., 2010). The SEM image of Ketac Nano (Fig. 4c) indicates a well formed notch area lacking voids, which could be expected to produce fewer regions for crack initiation during the test. However a high instance of predominantly interface failure, and lower mean γwofint indicates that interfacial adhesion was reduced compared to the two other RM-GICs in this test. C-GIC materials varied in paste consistency from a firm paste (Chemfil Rock, Ketac Molar) to relatively low viscosity pastes (Fuji IX GP Fast), although only firm C-GIC pastes had significantly lower mean γwofint to that achieved by Ketac Nano. Luting cements (both RM-GIC and C-GIC) are required to achieve a low film thickness and have smaller glass particles and lower powder to liquid ratios (1.8–2) than restorative GICs (3.1–3.7). This could contribute to lutingRM-GIC having a similar γwofint compared to high viscosity C-GICs which have a higher powder to liquid ratio, with a potential for more fracture regions at the notched area however no information to support this correlation is currently available. Luting-RM-GIC demonstrated a greater range of results compared to other materials, and this could be due
436
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
to the inability of the current test method to place pressure on the material forcing it onto the dentin surface. The current test method separates the crack surface in mode I (opening mode). γwofint is determined from the energy required to form two new surfaces. This parameter is important when determining the critical crack length, required to propagate a crack in catastrophic failure at the adhesive interface. The “chevron” shape of this adhesive interface, together with the introduction of a 0.1-mm crack depth initiation point, ensured that the crack opened and fracture began at the notch tip. The effect of crosshead speed and distance “L” of the crosshead to the adhesive interface have not been investigated for this method and should be standardized to avoid issues associated with other tests such as shear bond strength (VanNoort, 1989). Furthermore, as with shear bond and tensile studies, a detailed finite element analysis would provide valuable insight into the stress distribution in fracture mode-I (opening) at the interface and surrounding interface perimeter including the notch areas.
5.
Conclusions
Within the limits of this study, a visible-light-cured RM-GIC exhibited significantly higher (po0.05) γwofint compared to other RM-GIC, C-GIC, luting-C-GIC and luting-RM-GIC materials. Significant differences (po0.05) were observed between materials. The null hypothesis, that there is no difference in the γwofint among different glass-ionomer materials bonded to human dentin, was rejected.
references
Akinmade, A.O., Hill, R.G., 1992. Influence of cement layer thickness on the adhesive bond strength of polyalkenoate cements. Biomaterials 13, 931–936. Armstrong, S.R., Boyer, D.B., Keller, J.C., Park, J.B., 1998. Effect of hybrid layer on fracture toughness of adhesively bonded dentin-resin composite joint. Dental Materials 14, 91–98. Armstrong, S.R., Keller, J.C., Boyer, D.B., 2001. Mode of failure in the dentin-adhesive resin-resin composite bonded joint as determined by strength-based (muTBS) and fracture-based (CNSB) mechanical testing. Dental Materials 17, 201–210. Barker, L.M., 1977. A simplified method for measuring plane strain fracture toughness. Engineering Fracture Mechanics 9, 361–369. Boehm, A., Bronstrom, M., Broyles, B., Falsafi, A., 2010. Voidcontent and customer-satisfaction of glass-ionomers: novel automix versus triturated capsules. IADR 88th General Session Barcelona, Spain. Abstract #398. Cheng, Y.S., Douglas, W.H., Versluis, A., Tantbirojn, D., 1999. Analytical study on a new bond test method for measuring adhesion. Engineering Fracture Mechanics 64, 117–123. De Munck, J., Luehrs, A.-K., Poitevin, A., Van Ende, A., Van Meerbeek, B., 2013. Fracture toughness versus micro-tensile bond strength testing of adhesive–dentin interfaces. Dental Materials 29, 635–644. De Munck, J., Van Landuyt, K., Peumans, M., Poitevin, A., Lambrechts, P., Braem, M., Van Meerbeek, B., 2005. A critical review of the durability of adhesion to tooth tissue: methods and results. Journal of Dental Research 84, 118–132.
Della Bona, A., Anusavice, K.J., Mecholsky Jr., J.J., 2006. Apparent interfacial fracture toughness of resin/ceramic systems. Journal of Dental Research 85, 1037–1041. Della Bona, A., Watts, D.C., 2012. Discussing the future of dental materials, processes and products. Dental Materials 28, 1–2. Della Bona, A., Watts, D.C., 2013. Evidence-based dentistry and the need for clinically relevant models to predict material performance. Dental Materials 29, 1–2. Hashimoto, M., 2010. A review—micromorphological evidence of degradation in resin-dentin bonds and potential preventional solutions. Journal of Biomedical Material Research B: Applied Biomaterials 92, 268–280. Heintze, S.D., 2013. Clinical relevance of tests on bond strength, microleakage and marginal adaptation. Dental Materials 29, 59–84. Howard, K., Söderholm, K.-J.M., 2010. Fracture toughness of two dentin adhesives. Dental Materials 26, 1185–1192. Jancar, J., 2011. Bond strength of five dental adhesives using a fracture mechanics approach. Journal of the Mechanical Behavoir of Biomedical Materials 4, 245–254. Kelly, J.R., Benetti, P., Rungruanganunt, P., Bona, A.D., 2012. The slippery slope – critical perspectives on in vitro research methodologies. Dental Materials 28, 41–51. Kinloch, A.J., 1979. Interfacial fracture mechanical aspects of adhesive bonded joints—a review. The Journal of Adhesion 10, 193–219. Lin, C.P., Douglas, W.H., 1994. Failure mechanisms at the human dentin-resin interface: a fracture mechanics approach. Journal of Biomechanics 27, 1037–1047. Lucksanasombool, P., Higgs, W.A., Higgs, R.J., Swain, M.V., 2003. Interfacial fracture toughness between bovine cortical bone and cements. Biomaterials 24, 1159–1166. Mitsuhashi, A., Hanaoka, K., Teranaka, T., 2003. Fracture toughness of resin-modified glass ionomer restorative materials: effect of powder/liquid ratio and powder particle size reduction on fracture toughness. Dental Materials 19, 747–757. Rasmussen, S.T., 1978. Fracture studies of adhesion. Journal of Dental Research 57, 11–20. Rasmussen, S.T., 1984. Fracture properties of human teeth in proximity to the dentinoenamel junction. Journal of Dental Research 63, 1279–1283. Rasmussen, S.T., Patchin, R.E., 1984. Fracture properties of human enamel and dentin in an aqueous environment. Journal of Dental Research 63, 1362–1368. Rasmussen, S.T., Patchin, R.E., Scott, D.B., Heuer, A.H., 1976. Fracture properties of human enamel and dentin. Journal of Dental Research 55, 154–164. Ruse, N.D., Troczynski, T., MacEntee, M.I., Feduik, D., 1996. Novel fracture toughness test using a notchless triangular prism (NTP) specimen. Journal of Biomedical Material Research 31, 457–463. Sakai, M., Bradt, R.C., 1993. Fracture toughness testing of brittle materials. International Materials Reviews 38, 53–78. Salz, U., Bock, T., 2010. Testing adhesion of direct restoratives to dental hard tissue – a review. Journal of Adhesive Dentistry 12, 343–371. Scherrer, S.S., Cesar, P.F., Swain, M.V., 2010. Direct comparison of the bond strength results of the different test methods: a critical literature review. Dental Materials 26, e78–e93. Setien, V.J., Armstrong, S.R., Wefel, J.S., 2005. Interfacial fracture toughness between resin-modified glass ionomer and dentin using three different surface treatments. Dental Materials 21, 498–504. Soderholm, K.J., 2010. Review of the fracture toughness approach. Dental Materials 26, e63–77. Stephen, C.B., 2012. Correlation of clinical performance with ‘in vitro tests’ of restorative dental materials that use polymer-based matrices. Dental Materials 28, 52–71.
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Tam, L.E., Dev, S., Pilliar, R.M., 1995. Fracture toughness of conventional or photopolymerized glass ionomer/dentin interfaces. Operative Dentistry 20, 144–150. Tam, L.E., Pilliar, R.M., 1993. Fracture toughness of dentin/resincomposite adhesive interfaces. Journal of Dental Research 72, 953–959. Tam, L.E., Pilliar, R.M., 1994. Effects of dentin surface treatments on the fracture toughness and tensile bond strength of a dentin-composite adhesive interface. Journal of Dental Research 73, 1530. Tam, L.E., Yim, D., 1997. Effect of dentine depth on the fracture toughness of dentine-composite adhesive interfaces. Journal of Dentistry 25, 339–346. Tantbirojn, D., Cheng, Y.-S., Versluis, A., Hodges, J.S., Douglas, W.H., 2000. Nominal shear or fracture mechanics in the assessment of composite-dentin adhesion? Journal of Dental Research 79, 41–48. Tattersall, H.G., Tappin, G., 1966. The work of fracture and its measurement in metals, ceramics and other materials. Journal of Materials Science 1, 296–301. Tong, J., 2006. Interfacial fracture toughness of synthetic bonecement interface. Key Engineering Materials 312, 251–256. Tong, J., Wong, K.Y., Lupton, C., 2007. Determination of interfacial fracture toughness of bone-cement interface using sandwich Brazilian disks. Engineering Fracture Mechanics 74, 1904–1916.
437
Toparli, M., Aksoy, T., 1998. Fracture toughness determination of composite resin and dentin/composite resin adhesive interfaces by laboratory testing and finite element models. Dental Materials 14, 287–293. VanNoort, R., 1989. A critique of bond strength measurements. Journal of Dentistry 17, 61–67. Walshaw, P.R., Tam, L.E., McComb, D., 2003. Bond failure at dentin-composite interfaces with ‘single-bottle’ adhesives. Journal of Dentistry 31, 117–125. Wang, C.T., Pilliar, R.M., 1989. Bone cement bonding—interfacial fracture toughness determination. Clinical Materials 4, 135–153. Wang, X., Agrawal, C.M., 1997. Interfacial fracture toughness of tissue-biomaterial systems. Journal of Biomedical Material Research 38, 1–10. Wang, X., Agrawal, C.M., 2000. A mixed mode fracture toughness test of bone-biomaterial interfaces. Journal of Biomedical Material Research 53, 664–672. Wang, X., Lankford, J., Agrawal, C.M., 1994. Use of a compact sandwich specimen to evaluate fracture toughness and interfacial bonding of bone. Journal of Applied Biomaterials 5, 315–323.
Available online at www.sciencedirect.com
www.elsevier.com/locate/jmbbm
Research Paper
Evaluation of the interfacial work of fracture of glass-ionomer cements bonded to dentin Joshua J. Cheethama,n, Joseph E.A. Palamarab, Martin J. Tyasc, Michael F. Burrowd a
Melbourne Dental School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Australia Melbourne Dental School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Australia c Honorary Professorial Fellow, Melbourne Dental School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Australia d Clinical Associate Professor Faculty of Dentistry, The University of Hong Kong, People's Republic of China b
art i cle i nfo
ab st rac t
Article history:
Objective: The aim of this study was to investigate the interfacial work of fracture of
Received 16 July 2013
conventional (C-) and resin-modified (RM-) glass-ionomer cements (GICs) bonded to dentin.
Received in revised form
Methods: One hundred and sixty five aries-free human molars were embedded in epoxy
8 September 2013
resin, sectioned and polished with 300- and 600- grit silicon carbide paper to remove
Accepted 15 September 2013
enamel on the occlusal surface. Equilateral triangular-shaped plastic molds (4 4 4
Available online 10 October 2013
5 mm4) were clamped to the prepared dentin surfaces by a stainless steel test apparatus.
Keywords:
Teflon tape was placed under one internal vertex of the mold to create a 0.1-mm notch at
Work of fracture
the material-dentin interface. Interfacial work of fracture (γwofint) in tensile fracture mode-I
Interfacial work of fracture
(opening) was determined for six C-GIC, three RM-GIC, and two GIC luting cements at a
Resin-modified glass-ionomer
cross-head speed of 0.1 mm/min and a crosshead distance (L) from the interface of 4.3 mm.
Conventional glass-ionomer
The debonded surfaces were evaluated for the predominant failure mode. SEM analysis of examples showing interfacial and notch areas was performed. Results: ANOVA and Tukey's post hoc test demonstrated the highest mean γwofint value (90.16716.6 J/m2) of one RM-GIC was significantly different (po0.05) from the other materials. ‘High viscosity’ GICs achieved lower results with the lowest recorded at 20.4710.1 J/m2. There was a significant difference observed (po0.05) between the mean γwofint of luting C-GIC and luting RM-GIC. Although differences were observed between different material mean γwofint, when comparing groups no significant differences (p ¼0.181) were observed. For all groups, mixed GIC-interface failure (41%) was the most commonly observed, followed by cohesive failure in GIC (25%) and adhesive failure (20%). SEM analysis revealed that specimens generally fractured from the notch initiation point into the GIC or along the dentin–GIC interface. Conclusion: Within the limits of this study, significant differences (po0.05) were observed in the γwofint between different glass-ionomer materials. The null hypothesis that there is no difference in the γwofint among different glass-ionomer materials bonded to human dentin was rejected.
n Correspondence to: Melbourne Dental School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne 720 Swanston Street, Victoria, 3010 Australia. Tel.: þ61 3 93411532; fax: þ61 3 93411599. E-mail address: [email protected] (J.J. Cheetham).
1751-6161/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jmbbm.2013.09.020
428
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Relevance: In the current study, the interfacial work of fracture (γwofint) of glass-ionomer adhesive interfaces has been reported using a simple method that can be used to study the fracture mechanics of an adhesive interface without the need for complicated specimen preparation. & 2013 Elsevier Ltd. All rights reserved.
1.
Introduction
Variables such as test rig, biological substrate, dentin and enamel type and position in the tooth, storage of teeth, affect bond strength tests (Heintze, 2013). However the bond tests used are simple and easy to use, so they are still commonly reported in an effort to demonstrate the efficacy and ranking of adhesive dental materials. Comparing bond strength test results from different studies is not recommended as test conditions are invariably different among the various tests (Kelly et al., 2012; Scherrer et al., 2010; VanNoort, 1989). Several authors have questioned the validity of these tests and criticized the lack of standard protocols and relevance to clinical performance (Della Bona and Watts, 2013; Heintze, 2013; Stephen, 2012). Researchers have also suggested that a fracture mechanics approach is more appropriate than conventional shear or tensile bond strength tests (Salz and Bock, 2010). Using a fracture mechanics approach, a crack is introduced into the bond interface and the system's strength to resist crack propagation across the adhesive interface has also been investigated (Hashimoto, 2010; Salz and Bock, 2010). Critical stress intensity factor (KC), describes the ability to resits crack propagation. Linear elastic fracture toughness (KIC) studies the stress region ahead of crack propagation in tensile (mode – I) failure (Soderholm, 2010). Various methods have been employed to investigate interfacial fracture toughness of adhesives to teeth and biomaterials to determine the work of fracture which has been abbreviated as Wf andWi, plane strain interfacial fracture toughness (KICint), the adhesive (elastic-plastic) fracture energy (JIC) and the critical plane strain energy release rate (GICint) (Armstrong et al., 1998, 2001; Barker, 1977; Cheng et al., 1999; De Munck et al., 2013; Della Bona et al., 2006; Howard and Söderholm, 2010; Jancar, 2011; Lin and Douglas, 1994; Rasmussen, 1984; Rasmussen and Patchin, 1984; Tam and Pilliar, 1993, 1994; Tam and Yim, 1997; Toparli and Aksoy, 1998; Walshaw et al., 2003). Studies have also been performed on bone cements, which show adhesive failure occurring due to stress cracking at some point in their lifecycle (Lucksanasombool et al., 2003; Tong, 2006; Tong et al., 2007; Wang and Pilliar, 1989; Wang and Agrawal, 1997, 2000; Wang et al., 1994). Interfacial fracture toughness has also been measured for glass-ionomer materials (Akinmade and Hill, 1992; Mitsuhashi et al., 2003; Setien et al., 2005; Tam et al., 1995). Furthermore, a series of reviews on bond tests have recommended that a fracture mechanics approach be revisited as the preferred test method for adhesive strength evaluation (De Munck et al., 2005; Kinloch, 1979; Salz and Bock, 2010; Scherrer et al., 2010; Soderholm, 2010). A contributing factor as to why adhesive interfacial fracture toughness tests have not been commonly reported is
because current tests are complicated and often require specialized apparatus to prepare the dentin and dental material into the desired configuration. Attempts have been made to develop less complex systems, including adaptations of a common shear bond strength test using triangular shaped adhesive areas, (Cheng et al., 1999; Tantbirojn et al., 2000) and a notchless triangular prism specimen developed by Ruse et al. (Ruse et al., 1996). There is currently no available standardized test method for determination of interfacial fracture toughness properties of adhesive dental materials. Tattersall and Tappin introduced a simple method to determine the work of fracture of materials using a specimen with a square cross section and triangular fracture surface (Tattersall and Tappin, 1966). Further work by Rasmussen et al. demonstrated a test method to study the fracture properties of enamel and dentin, and determined the work of fracture (Wf or γwof), calculated by dividing the total energy (J) required to initiate fracture by twice the surface area (m2) of a triangular-shaped fractured surface (Rasmussen, 1984; Rasmussen and Patchin, 1984; Rasmussen et al., 1976). The test method was also adapted to investigate interfacial work of fracture (Wi or γwofint) for porcelain-gold and enamelcomposite adhesion (Rasmussen, 1978). Subsequent studies have also investigated fracture mechanics of different materials and interfaces using a γwof approach (Sakai and Bradt, 1993). The aim of this study was to compare the interfacial work of fracture (γwofint) and failure modes of several glassionomer cements bonded to dentin, using a new simplified test method. The null hypothesis is that there is no difference in the interfacial work of fracture (γwofint) among different glass-ionomer cement materials bonded to human dentin.
2.
Materials and methods
2.1.
Teeth preparation
Human ethics approval (♯1033315.1) for the use of human teeth was obtained from the University of Melbourne. Onehundred and sixty-five caries-free human molars were selected from a tooth bank. No information on the age of teeth was available. Teeth were stored in a refrigerated 0.5% chloramine T solution and used within six months of collection date. This method of storage followed guidelines described in ISO TS 11405 Dental materials – testing of adhesion to tooth structure. The teeth were cleaned with a slow-speed prophylaxis polisher (Zen; Philips, CA) and wet pumice, rinsed and stored in de-ionised water for approximately 24 h prior to embedding. Three stainless steel mold
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
sets were used to embed the teeth in epoxy resin, each containing 30 cylindrical mold cavities measuring 20 mm diameter x 30 mm high. To ensure the plates could be separated, low-density polyethylene film (GLADs WRAP; Glads Products Australia, Padstow, NSW) was placed between the lower plate and upper plate. Teeth were dried with tissue paper and a small amount of permanently plastic adhesive (BluTak; Bostik Australia Pty Ltd., Thomastown, Vic) was placed on the occlusal surface to maintain the tooth position within the mold cavity. Cold-cure epoxy casting resin (105 Epoxy resin, West System, MI) was placed into each cavity and allowed to cure. Mold sets were separated after 8 h and specimens were removed by a press and stored in deionised water at 37 1C for 24 h. Specimens were placed in a slow speed sectioning saw (ISOMETs 1000 precision sectioning saw; Buhler, Lake Bluff, IL) fitted with a 125-mm diameter, 0.4-mm thick diamond-coated blade and sectioned under water until enamel on the occlusal surface was removed. Exposed dentin surfaces were – ground under water at 300 rpm and 2 kg-f on a polishing machine (Phoenix Beta; Buehler, Lake Bluff, IL) with 300-grit and 600-grit silicon carbide paper. Specimens were stored in deionised water at room temperature prior to placement of the GIC.
2.2.
Fracture specimen preparation
A triangular plastics mold (Fig. 1) measuring 4 4 4 5 mm4 was placed onto the dentin surfaces to replicate the adhesive interface dimensions used by Ruse et al. (Ruse et al., 1996).
429
Apparatus to prepare the specimens for testing (Fig. 2a) consisted of a top plate connected with three springs to a base plate. The mold and specimen was inserted into the apparatus (Fig. 2a). Teflon tape (0.05 mm 15 mm) was placed under one corner of the mold to provide a ‘notch’ in the GIC cement (Fig. 2b) and stereo-microscopic examination at magnification 130 (M205C; Leica Microsystems GmbH, Wetzlar, Germany) using linear measuring software (LAS Application suite; Leica Microsystems GmbH) was used to achieve a distance from the edge of the tape 0.1 mm to the internal vertex of the mold (Fig. 2c). Names and composition of materials used are listed in Table 1. Although the smear layer was not removed from the prepared surfaces, surface conditioning of the dentin surface was completed prior to placement of the mold if recommended by the manufacturer (Table 1). Encapsulated materials were mixed in a capsule mixer (Ultramat 2; SDI Limited, Bayswater, Australia) as per manufacturer's instructions, inserted directly into the mold and lightly compacted with a dental plugger instrument to ensure adequate filling of the mold. The apparatus shown in Fig. 2a was placed in covered plastics boxes containing approximately 10 mm depth of deionised water and stored at 37 1C for 1 h to allow initial cement setting to occur. Specimens with the attached mold were removed from the apparatus, and immersed in water and stored at 37 1C for a further 23 h prior to determining γwofint.
2.3.
De-bonding procedure
The Teflon tape was removed and specimens placed into a fixture mounted on a universal testing machine (Model 8872; Instron Ltd, High Wycombe, England) connected to a 500 N load cell. Dentin surfaces were aligned with the vertical surface of the horizontal housing (Fig. 2d). A 0.7-mm flat edge tool attached to the load cell was placed directly against the surface of the housing achieving a distance “L” of 4.3 mm from the dentinmold interface (Fig 2 f). The flat edge was brought to within 0.5 mm of the mold (Fig. 2e) and crosshead speed was set to 0.1 mm per minute until the specimen was broken. The total area under the load-extension curve, representing the total energy (J) to initiate fracture at the interface was recorded for each specimen. Compliance was tested on the surface of samples for deformation and subtracted from the displacement measured for each specimen in order to determine the final total energy for each specimen.
2.4.
Determination of the interfacial work of fracture (γwofint)
The following formula was used to determine interfacial γwof (Rasmussen, 1978; Rasmussen, 1984; Rasmussen et al., 1976; Sakai and Bradt, 1993; Tattersall and Tappin, 1966): γ wof int ¼ Wwof =ð2AÞ where
Fig. 1 – Plastics mold dimensions, including outside edge, which is clamped down by the test apparatus. All dimensions are in mm.
γwofint is the interfacial work of fracture (J/m2) Wwof is the total energy recorded to initiate the fracture process (J) A is the surface area of the fractured surface (m2)
430
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Fig. 2 – Experimental apparatus. (a) Stainless steel apparatus. Mold held down into position by the spring loaded upper plate. (b) Location of the Teflon tape under the mold vertex prior to cement placement. Note apparatus has been removed for demonstration. (c) Fine adjustment of the Teflon tape position prior to cement placement (bar¼200 lm). (d) Specimen placed in horizontal housing showing flat edge tool position against housing. (e) Alignment of the flat edge tool 0.5 mm from the mold surface prior to testing. (f) Contact point of flat edge tool onto the mold surface. (g) Free body diagram demonstrating the loading configuaration on the mold.
2.5.
Failure analysis
Stereo-microscopic analysis at 20 magnification (M205C; Leica Microsystems GmbH) was used to report the predominant model of failure of each sample as: (a) cohesive in GIC; (b) mixed GIC-interface; (c) interface (or adhesive); and (d) cohesive in dentin. Scanning electron microscope (SEM) (Quanta 200 ESEM; FEI Company, Hillsboro, OR) images were taken of a representative number of uncoated samples at low vacuum with magnifications of up to 6000 to characterize the interfacial dentin and material fracture sites.
2.6.
Statistical analysis
ANOVA and post hoc Tukey tests were used (PASW Statistics V18, SPSS INC, Chicago, IL) to compare γwofint of each material bonded to dentin and differences between each material group, at a significance level of po0.05.
3.
Results
3.1.
Interfacial work of fracture (γwofint)
Mean γwofint were normally distributed for all materials, as assessed by Shapiro–Wilks test (p40.05). The mean γwofint are shown in Table 2 and Fig. 3. A total of 23 samples debonded prior to testing. Significant differences were observed between materials and Tukey's post hoc test revealed four homogenous subsets identified in Table 2. The highest mean value was observed with one visible light cured RM-GIC (Fuji II LC capsules) which was significantly different (po0.05) than other material mean γwofint results. No significant differences were observed between all other light cured RM-GIC mean γwofint, with the lowest value of 46.2 (720.9) J/m2 recorded for Ketac Nano. The group of C-GIC restorative materials recorded significantly different lower values when compared to Fuji II LC. Chemfil Molar and Ketac Molar recorded lower mean γwofint which were significantly different than two other C-GIC materials (Fuji IX GP Extra and Fuji IX GP Fast). GIC (Chemfil
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
431
Table 1 – Cement composition. Type
Product Name
Manufacturer
Lot number
Composition
RMGICa
Fuji II LC (Improved) capsules
GC Corporation, Tokyo, Japan
1008091
Powder: alumino-fluoro-silicate glass (amorphous) (95–100%) Liquid: polyacrylic acid (20–25%), 2-hydroxymethyl methacrylate (30– 35%), 2,2,4, trimethyl hexamethylene dicarbonate (1–5%), proprietary ingredient (5–15%) Powder:liquid ¼ 3.2:1 GC cavity conditioner: Polyacrylic acid (20%) Distilled water (77%), aluminum chloride hydrate (3%), food additive Blue no. 1 (o0.1%)
RMGIC
Photac Fil Quick
3 M/ESPE, Seefeld, Germany
411820
Powder: silane treated glass powder (499%), N,N-dimethylbenzocaine (o0.5%) Liquid: copolymer of acrylic acid-maleic acid (30–50%), HEMA (25–50%), water (20–30%), mono and di-HEMA phosphate, magnesium salt (5– 15%), diurethane dimethacrylate (3–10%) Powder:liquid ¼ 3.1:1
RMGIC
Ketac Nano
3 M/ESPE, Seefeld, Germany
N156309
Paste A: silane treated glass (40–55%), silane treated zirconia (20–30%), PEGDMA (5–15%), silane treated silica (5–15%), HEMA (1–15%), glass powder o5%, BisGMA o5%, TEGMA o1% Paste B: silane treated ceramic (40–60%), copolymer of acrylic and itaconic acid (20–30%), water (10–20%), HEMA (1–10%) Ketac Nano Glass Ionomer Primer: Water (40–50%) Hema (35–40%), copolymer of acrylic and itaconic acids (10–15%)
C-GICb
GC Fuji IXTM GP Extra
GC Corporation, Tokyo, Japan
1005261
Powder: fluoro aluminosilicate glass (amorphous) (90–100%). Polyacrylic acid (5–10%) Liquid: polyacrylic acid (30–40%), proprietary ingredient (5–15%) Powder:liquid ¼ 3.4:1 GC Cavity conditioner: Polyacrylic acid (20%), distilled water (77%), aluminum chloride hydrate (3%), food additive Blue no. 1 (o0.1%)
C-GIC
GC Fuji IXTM GP Fast
GC Corporation, Tokyo, Japan
1003051
Powder: fluoro aluminosilicate glass (amorphous) (90–100%). Polyacrylic acid (5–10%) Liquid: polyacrylic acid (30–40%), proprietary ingredient (5–15%) Powder:liquid ¼ 3.7:1 GC cavity conditioner: polyacrylic acid (20%), distilled water (77%), aluminum chloride hydrate (3%), food additive Blue no. 1 (o0.1%)
C-GIC
ChemFil Rock
Dentsply DeTrey, Konstanz, Germany
108001360
Powder: glass powder, polycarboxylic acid (10–25%) Liquid: polycarboxylic acid (10–25%), tartaric acid (2.5–10%), water Powder: Liquid ¼ ¼3.7: 1
C-GIC
ChemFil Molar
Dentsply DeTrey, Konstanz, Germany
0912000748
Powder: glass powder, polyacrylic acid (2.5–10%), tartaric acid (2.5–10%), silicon dioxide (2.5–10%) Liquid: water, polyacrylic acid (50–100%) Powder:liquid ¼ 3.7:1
C-GIC
Ketac Molar Aplicap
3M/ESPE, Seefeld, Germany
0912000748
Powder: glass powder (80–90%), copolymer of acrylic acid-maleic acid (1–6%) Liquid: water (40–45%), copolymer of acrylic acid-maleic acid (35–55%), tartaric acid (5–10%) Powder:liquid ¼ 3.4:1 Ketac Conditioner: Water (70–80%) polyacrylic acid (20–30%)
432
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Table 1 (continued ) Type
Product Name
Manufacturer
Lot number
Composition
C-GIC
Ketac Molar Quick Aplicap
3M/ESPE, Seefeld, Germany
471469
Powder: glass powder (93–98%), dichloromethylsilane reaction product with silica (0–1%), copolymer of acrylic and maleic acid (1–5%). Liquid: water (60–65%), copolymer of acrylic acid-maleic acid (30–40%), tartaric acid (5–10%) Powder:liquid ¼ 3.1:1 Ketac conditioner: Water (70–80%), polyacrylic acid (20–30%)
Luting C-GIC
GC Fuji Is
GC Corporation, Tokyo, Japan
1002101
Powder: alumino-fluoro-silicate glass (amorphous) (95%), polyacrylic acid (5%) Liquid: distilled water (50–55%), polyacrylic acid (30–40%) Powder:liquid ¼ 1.8:1
Luting RMGIC
GC Fuji Plus, capsules
GC Corporation, Tokyo, Japan
1003191
Liquid: polyacrylic acid, HEMA, UDMA Powder: alumino-silicate glass Powder:liquid ¼ 2:1 GC Fuji Plus conditioner: citric acid (10%), Distilled water (87%), iron (III) chloride (ferric chloride) (3%), food additive Blue no.1 (o0.1%)
Note: Primers and conditioners were applied to dentin according to manufacturers' instructions for use. a C-GIC ¼conventional glass-ionomer cement. b RM-GIC ¼resin-modified glass-ionomer cement.
Table 2 – Mean γwofint (J/m2), standard deviations (J/m2) and sample size. Group Fuji II LC Photac FIL QUICk Ketac Nano Fuji IX GP Extra Fuji IX GP Fast Chemfil Rock Chemfil Molar Ketac Molar Ketac Molar Quick Fuji I Fuji Plus
Mean γwofint (SD) 90.2 50.0 46.2 54.6 44.3 41.0 20.4 28.9 38.6 33.1 49.9
a
(16.6) (14.9)b,c (20.9)b,c (12.1)b (10.2)b,c (13.1)b,c,d (10.1)d (18.3)c,d (9.6)b,c,d,e (11.0)b,c,d (35.2)b,c
n 10 14 13 15 14 13 15 13 15 12 8
Mean γwofint with the same superscript are not significantly different (p40.05).
Molar) recorded the lowest value of 20.38 (710.1) J/m2, which was four times lower than the highest results recorded for RM-GIC. The two luting materials had similar means (p40.05). Although differences were observed between materials, when comparing groups no significant differences (p¼ 0.181) were observed between mean γwofint. The predominant modes of failure are shown in Table 3. For all groups, mixed GIC-interface failure (41%) was the most frequent one observed, followed by cohesive failure in GIC (25%) and interfacial (adhesive) failure (20%). Dentin pullout (cohesive failure) was not observed in any specimens. Dentin pullout is a significant energy dissipating mechanism in resin-based dental composites. The absence of dentin pullout
indicates that RM-GIC may not have penetrated the dentin tubules in a similar way demonstrated by solvent based adhesives. Ketac Nano predominantly failed at the interface (i.e., adhesive failure), whereas Fuji II LC had a similar number of failures for each mode. C-GIC tended to fail cohesively in the GIC and in mixed mode, whereas RM-GIC failed more frequently in mixed or interface (adhesive) modes. SEM analysis (Figs. 4, 5) revealed that specimens generally fractured from the notch initiation point into the GIC material or along the dentin–GIC interface. There were no instances of dentin “pull-out”. The region near the notched crack tip for Fuji II LC showed fracture occurring from the notch into the material and then down a plane parallel to the dentin surface. Lower mean γwofint compared to Fuji II LC was recorded for Ketac Nano, and images showed that the fracture that initiated at the notch typically followed along the dentin–GIC interface. The material's physical appearance at the notch area was observed to be different, for example the luting RM-GIC (Fuji Plus) and Ketac Nano did not show the larger particles which were present in other GICs at the notch area.
4.
Discussion
It is important to optimize the “adhesive” strength, including durability, of dental materials to provide “safe and effective products”(Della Bona and Watts, 2012). In the current study, the γwofint of glass-ionomer–dentin interfaces is reported, which provide a method that can be used to study the fracture mechanics of an adhesive interface. This study investigated a simple mold and apparatus that can be used
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
433
Fig. 3 – Comparison of the γwofint results (J/m2) for each glass-ionomer bonded to dentin.
Table 3 – Predominant mode of failure. Group
Cohesive in GIC
Mixed GIC-interface
Interface (adhesive)
De-bond prior to test
Fuji II LC Photac Fil Quick Ketac Nano Fuji IX GP Extra Fuji IX GP Fast Chemfil Rock Chemfil Molar Ketac Molar Ketac Molar Quick Fuji I Fuji Plus
3 2 0 9 5 3 4 4 6 4 2
3 7 2 6 9 8 11 6 7 6 2
4 5 11 0 0 2 0 3 2 2 4
5 1 2 0 1 2 0 2 0 3 7
Total n (%)
42 (25%)
67 (41%)
33 (20%)
23 (14%)
Note: no instances of cohesive in dentin failure mode were observed.
to determine the γwofint of dental materials without the need for complicated specimen shaping and preparation. The validity of bond strength tests has been questioned since their introduction (VanNoort, 1989). Scherrer (Scherrer et al., 2010) and others (Heintze, 2013; Soderholm, 2010; Tantbirojn et al., 2000) have recognized the deficiencies in current bond strength test methods, including fracture toughness methods previously reported (Lin and Douglas, 1994; Soderholm, 2010; Tong, 2006). Tattersall and Tappin developed a method of studying the work of fracture of monolithic materials such as ceramics, metals and plastics (Tattersall and Tappin, 1966). It involved simple specimen geometry, and fracture began at the notched “v” shaped section. This was further adapted for dentin and enamel fracture studies (Rasmussen, 1984; Rasmussen and Patchin, 1984; Rasmussen et al., 1976) and
interfacial adhesion (Rasmussen, 1978), however these tests also required detailed specimen preparation to achieve the triangular-shaped adhesive surface which is not required in the current test method. Although investigating adhesion of composite to enamel and using three point bending, Rasmussen (Rasmussen, 1978) found average γwofint results to be between 61 and 73 J/m2 which were higher than all of the mean γwofint results reported in the current study except for one RM-GIC (Fuji II LC). Other tests (Rasmussen et al., 1976) revealed average work of fracture of dentin to be in the range of 270–550 J/m2 which is well above results reported in the current study and could help to explain why there were no instances of cohesive failure in dentin. An advantage of the work of fracture technique is the simplicity in the calculation that is used, compared to other test methods. No other material properties used in other
434
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Fig. 4 – SEM images examples of fractured interfaces. (a) Overview of RM-GIC (Ketac Nano) mold and corresponding dentin surface and (b) at low magnification showing the full surface. Failure occurred at the cement–dentin interface. Note areas of GIC pullout that did not cause cohesive failure in the material. (c) Notch area (N) of nanoparticle containing RM-GIC (Ketac Nano) sample showing an absence of large particles. Failure occurred from the notched area and continued towards the perimeter of the specimen. (d) Dentin surface of C-GIC (Ketac Molar) cohesive failure at the notched area. Failure initially occurred along the dentin–GIC interface prior to moving from a mixed to cohesive GIC failure mode. (e) C-GIC specimen notch area. The specimens' filler particles provide sites for crack propagation due to their irregular shape. Note the presence of large cylindrical particles. (f) Example of spherical voids present in C-GIC surfaces after testing. (g) C-GIC specimen showing fracture along large filler particles towards a large void caused by filler pull out. (h) C-GIC (Fuji IX GP Extra) notch area (bar¼20 lm) demonstrating the inhomogeneity of material appearance in the area. Failure has occurred from this point under the large particles. (i) C-GIC (Fuji IX GP Fast) notch area showing similar material topography to Fuji IX GP Extra. Note the multiple fracture points and large particles. The top region demonstrates the non-bonded material notch area on the GIC side remains during the test. It is important to note that some images show artificial “cracks” in the GIC material due to desiccation cause by SEM vacuum environment.
Fig. 5 – SEM images of the dentine surface shows the predominant mode of failure under low magnification (a) Interfacial. (b) Cohesive in GIC. (c) Mixed GIC-interface.
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
tests, such as a materials modulus of elasticity are required for this test (Cheng et al., 1999; Sakai and Bradt, 1993; Soderholm, 2010; Tantbirojn et al., 2000; Tattersall and Tappin, 1966). Fig. 7 shows an example of a load-displacement curve with and without compliance correction which is an important consideration when determining work of fracture. The method described in this experiment has further advantages over current methods of investigating interfacial work of fracture material properties. Firstly, a standard triangular mold is used for placing the dental material. Secondly, preparation of the dentin requires only surface sectioning whereas other methods require intricate specimen preparation and complicated test apparatus which make it more difficult to achieve large sample sizes in a reasonable amount of time (Soderholm, 2010). Thirdly, the crack initiated by the tape at a vertex in the mold and the adhesive interface perimeter is not affected by grinding or cutting, thus preventing the introduction of further flaws at these critical points. Lastly, the test apparatus can be easily adapted for users that have facilities for bond strength testing and if required modifications (such as crack length and L) can be made easily. Fatigue testing, thermocycling and aging procedures could also be applied to the test specimens prior to the final fracture test to provide more relevant performance data. Table 3 details the number of specimens that de-bonded prior to testing. Unlike other interfacial fracture toughness tests, the specimens in this study remained in the mold. These specimens typically failed when the embedded tooth/ mold assembly was removed from the apparatus for immersion in water, as described in Fig. 1a. The failure was caused by operator error and/or incomplete filling during the material insertion process. Cement failure was observed in most of the materials; however this was more evident in the luting RM-GIC possibly due to air entrapment in this low viscosity cement preventing compaction by an instrument. Compaction of all materials in each group was manually achieved
Fig. 6 – Example of notched area SEM image after a predominantly interfacial break (bar¼500 lm).
435
Fig. 7 – Example of load – displacement curves with (line A) and without (line B) compliance correction.
and variation in this compaction method could have influenced every groups results. Similar to other interfacial fracture toughness tests there was diversity in the predominant failure mode of each material (Salz and Bock, 2010; Setien et al., 2005; Soderholm, 2010). Fracture could not be expected to proceed in a predictable direction, because of the complexity of material structure at the interface (Fig. 4g, h, i) and inconsistencies such as voids (Fig. 4f) present in the GIC material, which could contribute to variation in results. It is important to obtain a uniform notch area. Fig. 6 shows a typical notched area of a RM-GIC (Ketac Nano) that exhibited predominant interfacial debonding during the test. Unlike the C-GIC materials in the group, hardening of light-cured RM-GIC occurs rapidly after a short light-curing period due to the free radical polymerization reaction. The resin component of the RM-GIC could have contributed to a higher resistance to work of fracture compared to the C-GIC acid–base complex at 24 h after cement placement. Ketac Nano is supplied in a paste–paste form compared to capsule delivery for Fuji II LC and Photac-Fil. A benefit of paste systems is a reduction of entrapped air in the final paste compared to mixed capsule systems, which introduces porosity during mixing (Boehm et al., 2010). The SEM image of Ketac Nano (Fig. 4c) indicates a well formed notch area lacking voids, which could be expected to produce fewer regions for crack initiation during the test. However a high instance of predominantly interface failure, and lower mean γwofint indicates that interfacial adhesion was reduced compared to the two other RM-GICs in this test. C-GIC materials varied in paste consistency from a firm paste (Chemfil Rock, Ketac Molar) to relatively low viscosity pastes (Fuji IX GP Fast), although only firm C-GIC pastes had significantly lower mean γwofint to that achieved by Ketac Nano. Luting cements (both RM-GIC and C-GIC) are required to achieve a low film thickness and have smaller glass particles and lower powder to liquid ratios (1.8–2) than restorative GICs (3.1–3.7). This could contribute to lutingRM-GIC having a similar γwofint compared to high viscosity C-GICs which have a higher powder to liquid ratio, with a potential for more fracture regions at the notched area however no information to support this correlation is currently available. Luting-RM-GIC demonstrated a greater range of results compared to other materials, and this could be due
436
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
to the inability of the current test method to place pressure on the material forcing it onto the dentin surface. The current test method separates the crack surface in mode I (opening mode). γwofint is determined from the energy required to form two new surfaces. This parameter is important when determining the critical crack length, required to propagate a crack in catastrophic failure at the adhesive interface. The “chevron” shape of this adhesive interface, together with the introduction of a 0.1-mm crack depth initiation point, ensured that the crack opened and fracture began at the notch tip. The effect of crosshead speed and distance “L” of the crosshead to the adhesive interface have not been investigated for this method and should be standardized to avoid issues associated with other tests such as shear bond strength (VanNoort, 1989). Furthermore, as with shear bond and tensile studies, a detailed finite element analysis would provide valuable insight into the stress distribution in fracture mode-I (opening) at the interface and surrounding interface perimeter including the notch areas.
5.
Conclusions
Within the limits of this study, a visible-light-cured RM-GIC exhibited significantly higher (po0.05) γwofint compared to other RM-GIC, C-GIC, luting-C-GIC and luting-RM-GIC materials. Significant differences (po0.05) were observed between materials. The null hypothesis, that there is no difference in the γwofint among different glass-ionomer materials bonded to human dentin, was rejected.
references
Akinmade, A.O., Hill, R.G., 1992. Influence of cement layer thickness on the adhesive bond strength of polyalkenoate cements. Biomaterials 13, 931–936. Armstrong, S.R., Boyer, D.B., Keller, J.C., Park, J.B., 1998. Effect of hybrid layer on fracture toughness of adhesively bonded dentin-resin composite joint. Dental Materials 14, 91–98. Armstrong, S.R., Keller, J.C., Boyer, D.B., 2001. Mode of failure in the dentin-adhesive resin-resin composite bonded joint as determined by strength-based (muTBS) and fracture-based (CNSB) mechanical testing. Dental Materials 17, 201–210. Barker, L.M., 1977. A simplified method for measuring plane strain fracture toughness. Engineering Fracture Mechanics 9, 361–369. Boehm, A., Bronstrom, M., Broyles, B., Falsafi, A., 2010. Voidcontent and customer-satisfaction of glass-ionomers: novel automix versus triturated capsules. IADR 88th General Session Barcelona, Spain. Abstract #398. Cheng, Y.S., Douglas, W.H., Versluis, A., Tantbirojn, D., 1999. Analytical study on a new bond test method for measuring adhesion. Engineering Fracture Mechanics 64, 117–123. De Munck, J., Luehrs, A.-K., Poitevin, A., Van Ende, A., Van Meerbeek, B., 2013. Fracture toughness versus micro-tensile bond strength testing of adhesive–dentin interfaces. Dental Materials 29, 635–644. De Munck, J., Van Landuyt, K., Peumans, M., Poitevin, A., Lambrechts, P., Braem, M., Van Meerbeek, B., 2005. A critical review of the durability of adhesion to tooth tissue: methods and results. Journal of Dental Research 84, 118–132.
Della Bona, A., Anusavice, K.J., Mecholsky Jr., J.J., 2006. Apparent interfacial fracture toughness of resin/ceramic systems. Journal of Dental Research 85, 1037–1041. Della Bona, A., Watts, D.C., 2012. Discussing the future of dental materials, processes and products. Dental Materials 28, 1–2. Della Bona, A., Watts, D.C., 2013. Evidence-based dentistry and the need for clinically relevant models to predict material performance. Dental Materials 29, 1–2. Hashimoto, M., 2010. A review—micromorphological evidence of degradation in resin-dentin bonds and potential preventional solutions. Journal of Biomedical Material Research B: Applied Biomaterials 92, 268–280. Heintze, S.D., 2013. Clinical relevance of tests on bond strength, microleakage and marginal adaptation. Dental Materials 29, 59–84. Howard, K., Söderholm, K.-J.M., 2010. Fracture toughness of two dentin adhesives. Dental Materials 26, 1185–1192. Jancar, J., 2011. Bond strength of five dental adhesives using a fracture mechanics approach. Journal of the Mechanical Behavoir of Biomedical Materials 4, 245–254. Kelly, J.R., Benetti, P., Rungruanganunt, P., Bona, A.D., 2012. The slippery slope – critical perspectives on in vitro research methodologies. Dental Materials 28, 41–51. Kinloch, A.J., 1979. Interfacial fracture mechanical aspects of adhesive bonded joints—a review. The Journal of Adhesion 10, 193–219. Lin, C.P., Douglas, W.H., 1994. Failure mechanisms at the human dentin-resin interface: a fracture mechanics approach. Journal of Biomechanics 27, 1037–1047. Lucksanasombool, P., Higgs, W.A., Higgs, R.J., Swain, M.V., 2003. Interfacial fracture toughness between bovine cortical bone and cements. Biomaterials 24, 1159–1166. Mitsuhashi, A., Hanaoka, K., Teranaka, T., 2003. Fracture toughness of resin-modified glass ionomer restorative materials: effect of powder/liquid ratio and powder particle size reduction on fracture toughness. Dental Materials 19, 747–757. Rasmussen, S.T., 1978. Fracture studies of adhesion. Journal of Dental Research 57, 11–20. Rasmussen, S.T., 1984. Fracture properties of human teeth in proximity to the dentinoenamel junction. Journal of Dental Research 63, 1279–1283. Rasmussen, S.T., Patchin, R.E., 1984. Fracture properties of human enamel and dentin in an aqueous environment. Journal of Dental Research 63, 1362–1368. Rasmussen, S.T., Patchin, R.E., Scott, D.B., Heuer, A.H., 1976. Fracture properties of human enamel and dentin. Journal of Dental Research 55, 154–164. Ruse, N.D., Troczynski, T., MacEntee, M.I., Feduik, D., 1996. Novel fracture toughness test using a notchless triangular prism (NTP) specimen. Journal of Biomedical Material Research 31, 457–463. Sakai, M., Bradt, R.C., 1993. Fracture toughness testing of brittle materials. International Materials Reviews 38, 53–78. Salz, U., Bock, T., 2010. Testing adhesion of direct restoratives to dental hard tissue – a review. Journal of Adhesive Dentistry 12, 343–371. Scherrer, S.S., Cesar, P.F., Swain, M.V., 2010. Direct comparison of the bond strength results of the different test methods: a critical literature review. Dental Materials 26, e78–e93. Setien, V.J., Armstrong, S.R., Wefel, J.S., 2005. Interfacial fracture toughness between resin-modified glass ionomer and dentin using three different surface treatments. Dental Materials 21, 498–504. Soderholm, K.J., 2010. Review of the fracture toughness approach. Dental Materials 26, e63–77. Stephen, C.B., 2012. Correlation of clinical performance with ‘in vitro tests’ of restorative dental materials that use polymer-based matrices. Dental Materials 28, 52–71.
journal of the mechanical behavior of biomedical materials 29 (2014) 427 –437
Tam, L.E., Dev, S., Pilliar, R.M., 1995. Fracture toughness of conventional or photopolymerized glass ionomer/dentin interfaces. Operative Dentistry 20, 144–150. Tam, L.E., Pilliar, R.M., 1993. Fracture toughness of dentin/resincomposite adhesive interfaces. Journal of Dental Research 72, 953–959. Tam, L.E., Pilliar, R.M., 1994. Effects of dentin surface treatments on the fracture toughness and tensile bond strength of a dentin-composite adhesive interface. Journal of Dental Research 73, 1530. Tam, L.E., Yim, D., 1997. Effect of dentine depth on the fracture toughness of dentine-composite adhesive interfaces. Journal of Dentistry 25, 339–346. Tantbirojn, D., Cheng, Y.-S., Versluis, A., Hodges, J.S., Douglas, W.H., 2000. Nominal shear or fracture mechanics in the assessment of composite-dentin adhesion? Journal of Dental Research 79, 41–48. Tattersall, H.G., Tappin, G., 1966. The work of fracture and its measurement in metals, ceramics and other materials. Journal of Materials Science 1, 296–301. Tong, J., 2006. Interfacial fracture toughness of synthetic bonecement interface. Key Engineering Materials 312, 251–256. Tong, J., Wong, K.Y., Lupton, C., 2007. Determination of interfacial fracture toughness of bone-cement interface using sandwich Brazilian disks. Engineering Fracture Mechanics 74, 1904–1916.
437
Toparli, M., Aksoy, T., 1998. Fracture toughness determination of composite resin and dentin/composite resin adhesive interfaces by laboratory testing and finite element models. Dental Materials 14, 287–293. VanNoort, R., 1989. A critique of bond strength measurements. Journal of Dentistry 17, 61–67. Walshaw, P.R., Tam, L.E., McComb, D., 2003. Bond failure at dentin-composite interfaces with ‘single-bottle’ adhesives. Journal of Dentistry 31, 117–125. Wang, C.T., Pilliar, R.M., 1989. Bone cement bonding—interfacial fracture toughness determination. Clinical Materials 4, 135–153. Wang, X., Agrawal, C.M., 1997. Interfacial fracture toughness of tissue-biomaterial systems. Journal of Biomedical Material Research 38, 1–10. Wang, X., Agrawal, C.M., 2000. A mixed mode fracture toughness test of bone-biomaterial interfaces. Journal of Biomedical Material Research 53, 664–672. Wang, X., Lankford, J., Agrawal, C.M., 1994. Use of a compact sandwich specimen to evaluate fracture toughness and interfacial bonding of bone. Journal of Applied Biomaterials 5, 315–323.
