Comparison of three and four point bending evaluation of two adhesive bonding systems for glass-ceramic zirconia bi-layered ceramics



To quantify the adhesion of two bonding approaches of zirconia to more aesthetic glass-ceramic materials using the Schwickerath (ISO 9693-2:2016) three point bend (3PB) [1] test to determine the fracture initiation strength and strain energy release rate associated with stable crack extension with this test and the Charalamabides et al. (1989) [2] four point bend (4PB) test.


Two glass-ceramic materials (VITABLOCS Triluxe forte, Vita Zahnfabrik, Germany and IPS.emax CAD, Ivoclar Vivadent, Liechtenstein) were bonded to sintered zirconia (VITA InCeram YZ). The former was resin bonded using a dual-cure composite resin (Panavia F 2.0, Kuraray Medical Inc., Osaka, Japan) following etching and silane conditioning, while the IPS.emax CAD was glass bonded (IPS e.max CAD Crystall/Connect) during crystallization of the IPS.emax CAD. Specimens (30) of the appropriate dimensions were fabricated for the Schwickerath 3PB and 4PB tests. Strength values were determined from crack initiation while strain energy release rate values were determined from the minima in the force-displacement curves with the 3PB test (Schneider and Swain, 2015) [3] and for 4PB test from the plateau region of stable crack extension.


Strength values for the resin and glass bonded glass ceramics to zirconia were 22.20 ± 6.72 MPa and 27.02 ± 3.49 MPa respectively. The strain energy release rates for the two methods used were very similar and for the glass bonding, (4PB) 15.14 ± 5.06 N/m (or J/m 2 ) and (3PB) 16.83 ± 3.91 N/m and resin bonding (4PB) 8.34 ± 1.93 N/m and (3PB) 8.44 ± 2.81 N/m respectively. The differences in strength and strain energy release rate for the two bonding approaches were statistically significant (p < 0.05). SEM observations showed fracture occurred adhesively for the resin bonding and cohesively for the glass bonding.


The present results indicate 3PB and 4PB tests have very similar values for the strain energy release rate determination. However while strength tests reveal minimal differences between resin and glass bonding, strain energy release rates for the latter are superior for bonding CAD/CAM milled glass-ceramics to zirconia.


CAD/CAM technology in prosthetic dentistry has rapidly improved in recent years, with the goal of developing reliable and cost-effective manufacture of materials and restorations for enhanced aesthetics as well as increased efficiency and productivity . Mechanical properties of materials used in restorations is a subject of importance for clinical reliability. Consequently a stiffer and stronger core substructure is commonly desired for bi-layered restorations to resist high stresses developed under occlusal load . CAD/CAM systems enable materials such as yttria-stabilized tetragonal zirconia to be milled for use as a substructure then combined with a ceramic veneer, such a zirconia ceramic bi-layer system makes for a durable and aesthetic restoration .

Studies on CAD/CAM zirconia glass-ceramic veneering systems reported a decrease in unavoidable cohesive fractures of restorations as they are a material combination with high initial fracture resistance. This combination results in a decrease in chipping, improved sensitivity against thermo-cycling and mechanical load bearing resistance . However the geometry of layered restorations may affect durability and longevity as the veneer must also exhibit adequate strength and enable force to be transmitted to the stronger substructure . CAD/CAM restorations potentially lower stresses at the bi-layer interface and are reported to exhibit high initial fracture resistance to minimize cohesive failures . Two CAD/CAM approaches have been developed to enable the bonding of stronger, tougher and more aesthetic veneer to a zirconia substrate, namely “CAD-on” as marketed by Ivoclar and the “Rapid-layer” technology by Vita. The former uses a glass bonder that is fused at elevated temperatures during the heat (ceraming) treatment of a lithium metasilicate machinable glass ceramic while the latter relies on a dual cure composite resin to bond a machinable feldspathic dental ceramic to a zirconia substrate. Standards for analysis are required to properly gauge mechanical performance of a material for dental restorations as a measurement of fracture resistance can predict material de-bonding and failure . While there is currently an international universal standard for evaluating the strength associated with porcelain bonding to zirconia ( ISO 9693-2:2016 ) no such standard exists for the evaluation of the adhesion energy.

Material testing is commonly done under shear, tensile, and flexural but there arises a question of validity due to an array of results across the community under multiple test methodologies . Common test methods of flexural strength for ceramic materials are shear and tensile where a static load is applied until failure occurs . While tensile bond strength tests are sometimes considered better evaluators for adhesive failures due to the uniform interfacial stresses , shear tests are far more popular and often include comments on failure modes. However, some studies have contradictory views finding shear bond strength tests to be seriously flawed . The results are influenced by the non-homogenous shear and bending stress distribution with shear bond strength tests, in addition these tests result in predominately cohesive failures .

Strength tests have tended to dominate the mechanical characterisation of dental materials especially adhesion, which has become ever more critical for modern dentistry. However strength tests measure a combination of the applied and residual stress state plus the defect size(s) present in the vicinity of the maximum stress. A major consequence of the above is that considerable scatter in recorded values is measured and the values are specimen size dependent . Fracture toughness is an established method for quantifying a more intrinsic property of a material. Genuine measurements of toughness are associated with an equilibrium condition of crack extension not the unstable initiation of a crack . The situation for bonded systems becomes somewhat more complex as often materials with very different elastic modulus values are bonded resulting in a complex stress intensity factor determination for extension of an interface crack . For the past decade the 4PB Charalambides interfacial fracture toughness test has been used by a number of researchers to quantify the strain energy release rate G (N/m) for bonded dental materials . This test enables stable peeling of a bonded material from a substrate and shows a plateau in the associated force-displacement curve during crack extension over many (5–10) mm. More recently it has been shown that the Schwickerath 3PB test is also capable of generating stable crack extension from which the strain energy release rate can be determined .

The aim of this study is to evaluate four-point and three-point flexure tests to further the understanding of these two fracture mechanics testing methods for adhesion assessment of bonded zirconia ceramic systems. Our hypothesis is that the Schwickerath 3PB and Charalambides 4PB fracture toughness tests for determination of the strain energy release rate for adhesion bonding can be correlated. In addition it is hypothesised that the 3PB determined strength as well as the 3PB and 4PB strain energy release rate estimations of adhesion for resin and glass bonding of glass ceramics to zirconia are comparable.

Materials and method

Two different bonding systems were investigated under two testing methods: composite resin bonding and glass ceramic bonding, using three-point and four-point bend testing. A total of 30 zirconia plates were made for 15 composite resin bonded glass-ceramic and 15 glass bonded glass-ceramic specimen groups. 15 specimens in each group were tested under four-point bend test to determine the strain energy release rate, while 15 of each group were tested under the Schwickerath crack initiation test (ISO9693-2:2016), a three-point bend test for porcelain-ceramic systems that also enables strain energy release rate to be measured .

Specimen preparation

Four-point bending preparation

Pre-sintered yttria-stabilised zirconia blocks (VITA InCeram YZ) were sectioned with a Struers sectioning machine (Accutom-50), using a Struers Diamond cut-off wheel (M1D13) with dimensions of 37.5 mm × 10 mm × 1.875 mm. This dimension compensated for the ≈20% zirconia sintering shrinkage giving a 30 mm × 8 mm × 1.5 mm substrate dimension post sinter, at a holding temperature of 1530 °C (VITA ZYrcomat) under pre-set settings. The Struers machine was used to section the veneering materials, a feldspathic ceramic (VITABLOCS Triluxe forte VITA RLT, Zahnfrabrik, Germany) and lithium disilicate (IPS.emax CAD, Ivoclar Vivadent, Liechtenstein), to dimensions of 15 mm × 8 mm × 1.5 mm.

Three-point bending preparation

Following the Schwickerath ISO9693-2:2016 crack initiation standard, pre-sintered zirconia beams were prepared with a final sintered dimension of 25 mm × 3 mm × 0.5 mm. A cleaning cycle for zirconia involved firing in an Ivoclar Vivadent programat P500 before sintering in the VITA ZYrcomat furnaces. The veneer materials were sectioned using similar methods to dimension of 8 mm × 3 mm × 1 mm.

Composite resin bonding

Zirconia surface preparation before cementation was by ultrasonic and steam cleaning. Feldspathic ceramic was steam-cleaned before etching with 5% HF acid gel (VITA Ceramics Etch) for 60 s on the cementing surface. Removal of the acid gel was by distilled water in an ultrasonic bath for 120 s before being rinsed and left to air-dry for 20 s. A successful HF etch resulted in a frosted opaque surface to which a silane coupling agent (Clearfil Porcelain Bond Activator, Kuraray Medical Inc., Osaka, Japan) was applied for 120 s then air-dried before cementing. The composite resin bonding cement selected was a dual-cure composite resin (Panavia F 2.0, Kuraray Medical Inc., Osaka, Japan), a recommended resin composite by VITA.

For generating a uniform 0.5 mm pre-notch width between the cemented veneering plates for the four-point bending specimens, a thin polymer sheet was placed between two veneer plates and zirconia centred on a glass jig allowing visual determination of adequate cement and bilayer alignment. Three-point bend specimens bonded with composite resin cement were made in a similar manner. Curing composite resin cemented specimens required a five second light cure (Demi Plus light curing system, Kerr Corporation), before gentle cleaning of excess cement. An oxyguard gel (Panavia F 2.0, Kuraray Medical Inc., Osaka, Japan) was applied onto exposed cemented areas before a 20 s cure on each surface of the veneer plates. The plates were left overnight to allow complete cure and the oxyguard gel removed.

Glass ceramic bonding

Zirconia substrate and lithium disilicate veneer plates were ultrasonically cleaned prior to bonding. A pre-dosed thixotropic mixture (IPS e.max CAD Crystall./Connect) was vibrated for 10 s on an Ivomix (Ivoclar Vivadent) following manufacturer’s instructions. Following the pre-notched method described above for the resin bonding, application of the mixture was applied to allow for vibrating out excess between the plates to ensure adequate paste mixture distributed. Specimens were dried and brushed clean with a soft dry brush before firing to fuse the glass in a Programat P500 under manufacturer’s setting (IPS e.max CAD-on technique Fusion/Crystallization). Specimens bonded with glass ceramic were then stored in kerosene until testing to minimize the effect of water vapour from the atmosphere influencing the bond strength .

Four-point testing

The stable crack extension test in four-point bending was carried out on an Instron Universal Testing Machine (Bluehill 2.3 software). Flexural loading was in a one fourth test configuration (inner loading at one fourth of the outer span, which were 10 mm inner and 20 mm outer span respectively), with a crosshead speed of 0.5 mm/min along with a 500N load cell.

Strain energy release rate is calculated as G-value (units: N/m or J/m 2 )

G = η ( P 2 l 2 ( 1 − ν z 2 ) ) E z b 2 h 3

Where η is the non-dimensional parameter, P is the load to stably propagate the crack, l is the moment arm or distance between inner and outer rollers on the universal testing machine, ν z and E z are the Poisson’s ratio and elastic modulus of the zirconia respectively, and b and h are the width and total thickness of the specimens. The non-dimensional parameter is calculated by:

η = ( 3 2 ) [ 1 ( h z h ) 3 − λ { ( h p h ) 3 + λ ( h z h ) 3 + 3 λ ( h p h z h 2 ) [ h p h + λ h z h ] − 1 } ]
Only gold members can continue reading. Log In or Register to continue

Nov 22, 2017 | Posted by in Dental Materials | Comments Off on Comparison of three and four point bending evaluation of two adhesive bonding systems for glass-ceramic zirconia bi-layered ceramics
Premium Wordpress Themes by UFO Themes