The fracture resistance of a CAD/CAM Resin Nano Ceramic (RNC) and a CAD ceramic at different thicknesses

Abstract

Objectives

This study aimed to investigate the influence of restoration thickness to the fracture resistance of adhesively bonded Lava™ Ultimate CAD/CAM, a Resin Nano Ceramic (RNC), and IPS e.max CAD ceramic.

Methods

Polished Lava™ Ultimate CAD/CAM (Group L), sandblasted Lava™ Ultimate CAD/CAM (Group LS), and sandblasted IPS e.max CAD (Group ES) discs ( n = 8, Ø = 10 mm) with a thickness of respectively 0.5 mm, 1.0 mm, 1.5 mm, 2.0 mm, and 3.0 mm were cemented to corresponding epoxy supporting discs, achieving a final thickness of 3.5 mm. All the 120 specimens were loaded with a universal testing machine at a crosshead speed of 1 mm/min. The load (N) at failure was recorded as fracture resistance. The stress distribution for 0.5 mm restorative discs of each group was analyzed by Finite Element Analysis (FEA). The results of facture resistances were analyzed by one-way ANOVA and regression.

Results

For the same thickness of testing discs, the fracture resistance of Group L was always significantly lower than the other two groups. The 0.5 mm discs in Group L resulted in the lowest value of 1028 (112) N. There was no significant difference between Group LS and Group ES when the restoration thickness ranged between 1.0 mm and 2.0 mm. There was a linear relation between fracture resistance and restoration thickness in Group L ( R = 0.621, P < 0.001) and in Group ES ( R = 0.854, P < 0.001). FEA showed a compressive permanent damage in all groups.

Significance

The materials tested in this in vitro study with the thickness above 0.5 mm could afford the normal bite force. When Lava Ultimate CAD/CAM is used, sandblasting is suggested to get a better bonding.

Introduction

Due to increasing concerns about the aesthetics and biocompatibility of dental restorations, patients and dentists have become more and more interested in metal-free tooth-colored materials. Ceramic materials were developed in response to this increasing demand . Although ceramics are routinely used for dental restorations, a major drawback is their high clinical failure rate in posterior sites . All-ceramic crowns are often replaced because of bulk fracture, a catastrophic failure mode noted in both monolithic and layered crowns . The fracture resistance of a layered ceramic crown can be influenced by its core design such as the thickness of the core or veneering ceramic . Most dental ceramics are considered brittle because of their low tensile strength and fracture toughness, which can be influenced by the presence of inherent flaws within the material . Tensile stresses caused by external loading can lead to a propagation of cracks starting at these inherent flaws and other defects . Therefore, cracks usually initiate from the inner surface of ceramics, i.e. the cementation surface, where tensile strength is the highest, and then propagate through the material to the outer surface, ultimately leading to bulk fracturing .

In an attempt to improve the mechanical properties, industrially made CAD/CAM ceramics blocks have been introduced to dentistry . Processing ceramics under industrial conditions resulted in remarkable reduction in voids, flaws, and cracks in comparison with laboratorial processing . One of these CAD/CAM ceramics is IPS e.max CAD (Ivoclar Vivadent AG, Liechtenstein), an improved glass-ceramic material with a relatively high fracture strength .

Lava™ Ultimate CAD/CAM Restorative (3M-ESPE™, St Paul, USA) is another material for CAD/CAM technique. As introduced by its manufacturer, this material is called Resin Nano Ceramic (RNC), which is supposed to be unique in durability and function by the manufacturer. However, from the material science perspective, this material is still belonging to the resin composite category. It was reported that the CAD/CAM resin composites that are marketed as ‘classified as ceramic’ may show less crack propagations under fatigue forces than that of some CAD/CAM ceramics . They may even provide better fracture resistance for non-retentive occlusal veneers in posterior teeth than some CAD/CAM ceramics .

Despite the material properties and the restoration designs, the thickness of ceramic restorations can also be an important factor in fracture resistance . While the usual recommendation for porcelain restoration thickness is 1.5–2.0 mm , with the development of stronger materials in combination with CAD/CAM techniques and innovative adhesive technology, a more conservative thinner crown can also be considered .

Furthermore, the fracture resistance of ceramic restorations can be also influenced by the properties of the support , for example, by its elastic modulus , or the bond strength. It was reported that well-luted specimens were usually more fracture resistant . It was supposed that luting agents had a possible bridging effect on the interfacial surface defects, which restricted and resisted against the propagation of cracks from the internal surface of the bonded resin and led to a higher fracture resistance.

The purpose of this study was to investigate the influence of thickness to the fracture resistance of Lava Ultimate CAD/CAM RNC and the IPS e.max CAD ceramic. Furthermore, the effect of bonding on the fracture resistance of Lava Ultimate CAD/CAM RNC was evaluated and the results were rationalized with Finite Element Analysis.

Materials and methods

A simplified tri-layer onlay model was designed to mimic the restoration for posterior tooth ( Fig. 1 ). The testing disc, which represented the occlusal restoration, had a diameter of 10 mm to mimic the average dimension of molars. The testing disc was cemented to the substrate epoxy disc, e.g. simulated dentin, with an equal diameter. The epoxy material (similar to G10, previously used in other studies; see Discussion) had an elastic modulus of 18 MPa, which was similar to dentin . The bonded two-layer disc had a final thickness of 3.5 mm, which was chosen as equivalent to the average thickness from pulp wall to occlusal surface . Then the two-layer disc was bonded to a steel ring with an inner diameter of 6.5 mm, an outer diameter of 10 mm, and a thickness of 1.5 mm mimicking the pulp chamber.

Fig. 1
Schematic representation of the set-up model use for this study.

Fracture resistance

The testing specimens were divided into three groups: L, LS and ES, according to testing materials and surface treatment. The testing discs in Group L were made of Lava™ Ultimate CAD/CAM Restorative (3M™ ESPE™, USA), cemented to polished epoxy substructures (Epoxydplatte, Carbotec GmbH & Co. KG, Aachen, Germany). Specimens in Group LS were also made of Lava Ultimate, with a sandblasted surface and cemented to sandblasted epoxy discs. In Group ES, testing discs were made of an IPS e.max CAD (Ivoclar Vivadent AG, Liechtenstein), also with a sandblasted surface and cemented to sandblasted epoxy discs. Each group was further divided into five subgroups ( n = 8) according to the thickness of the testing disc. The thicknesses of the testing discs were 0.5 mm, 1.0 mm, 1.5 mm, 2.0 mm, and 3.0 mm; the thicknesses of their corresponding epoxy discs were 3.0 mm, 2.5 mm, 2.0 mm, 1.5 mm, and 0.5 mm, respectively, achieving a final thickness of 3.5 mm.

For Groups L and LS, RNC cylinders were made from Lava Ultimate blocks, by a water-cooled bur with an inner diameter of 10 mm. The cylinders were further cut into 80 discs by a sawing machine (Isomet 1000; Buehler, USA), according to the five demanded thicknesses. For Group ES, 40 ceramic discs were made the same way from IPS e.max CAD blocks. 120 epoxy discs were cut from five large epoxy plates, with the corresponding five demanded thicknesses, by the same drilling machine and drill. All the 240 discs were polished by 600 grit SiC polish papers. 160 discs for Groups LS and ES were sandblasted for 10 s on one surface, by 50 μm Al 2 O 3 at a distance about 10 mm. All the 240 discs were cleaned by an ultrasonic machine (Bransonic 3510; Branson Ultrasonics Corp, USA) for 5 min.

The bonding surfaces of the Lava Ultimate discs were treated with mixed Clearfil™ SE bond primer (Kuraray CO. LTD, Japan) and a silane coupling primer (Clearfil Porcelain Bond Activator; Kuraray CO. LTD, Japan) for 10 s, according to manufacturer’s recommendation. The ceramic bonding surface was first etched with 9.5% hydrofluoric acid (PorcelEtch Syringe Green; Cosmedent, Inc., USA) for 20 s, rinsed with water, air dried and then applied 20 s of the mixture of SE bond primer and the porcelain activator. The dentin-like epoxy surface was etched with 40% phosphoric acid (K-Etchant Gel; Kuraray Medical Inc, Japan) for 30 s to clean the surface, rinsed with water and lightly air dried prior to a 10 s application of the SE bond primer. The epoxy discs were bonded to the metal rings by SE bond. For each specimen, the testing disc was then cemented to the corresponding epoxy disc with a resin cement (Panavia F 2.0; Kuraray Medical Inc, Japan). During curing a load of 50 N was applied on the specimen. The cement was cured in five directions, for each direction 30 s, by a curing light (Astralis 10, Ivoclar Vivadent, Liechtenstein). All the specimens were stored in 37 °C water for 24 h.

The completed specimens were loaded with a hemispherical steel indenter (Ø = 4.9 mm), centered on the top surface. The load was applied until failure, with a universal testing machine (Instron 6022, Instron Corp., MA, USA) at a crosshead speed of 1 mm/min. The load (N) at the failure was recorded as fracture resistance, in order to compare with the normal biting forces in the clinical situation. The results of 0.5 mm restorative discs were further used in the finite element analysis (FEA) for each group, for the analysis of the stress distribution.

Finite element analysis

Three dimensional FEA models of the test set-up with the dimensions according to the testing specimens were made (loading sphere models). The Finite Element modeling was carried out using FEMAP software (FEMAP 10.1.1; Siemens PLM software, Plano, Texas, USA), while the analysis was done with NX Nastran software (NX Nastran; Siemens PLM Software, Plano, Texas, USA). The models consisted of the layer of the supporting ring, the supporting epoxy, the testing disc, and the loading sphere. Since the models were symmetrical in geometry in two directions, they were split in quarter specimens to facilitate the border conditions, with the nodes in the centric planes allowing for sliding in the surface only. The contact surface between the loading sphere and the testing disc was modeled as contact surface with a friction coefficient of 0.45. For the models of Group L the interface of the testing disc and the epoxy was also modeled as a contact surface with a friction coefficient of 0.30, assuming an insufficient bond between these surface, while for the models of Group LS and ES the bond was assumed to be sufficient (not contact surface, but a fixed surface). The fracture load for the three models, shown in Table 3 , was loaded on the cut surface of the loading sphere ( Fig. 1 ). The nodes in the bottom of the supporting ring were fixed. The models were composed of 8,730 parabolic tetrahedron solid elements. The material properties from the manufacturer, used for the FEA were summarized in Table 1 . The Solid Maximum Principal stresses were calculated to establish the maximum tensile stress in the testing disc. Since the highest Solid von Mises stresses were in all cases higher than the Solid Maximum Principal stress, the Solid von Mises stress was used to calculate the maximum compressive stress.

Table 1
The material properties used in the FEA.
Young’s modulus (GPa) Poisson ratio
Lava™ Ultimate 12.8 0.30
IPS e-Max CAD 95.0 0.30
Substrate epoxy 18.0 0.30
Sphere/supporting ring 195.0 0.30

Since the compressive stresses in the contact surface between loading sphere and testing disc were in the FEA higher than the material strength, we assumed that the surface of the testing disc had been already permanently deformed before final fracture. For this reason, new models of the test setup (deformed surface models) were made. The deformation was designed by lowering the loading sphere into the surface of the testing disc and shaping the surface around the part of the loading sphere in the surface. The displacement of the loading sphere for the permanent deformation was calculated by the difference of the displacement of the crosshead of the universal testing machine on the test set-up, e.g. the total deformation of the test set-up, and the elastic deformation in the FEA ( Table 2 ). The models were symmetrical in geometry and were split in half specimens to facilitate the border conditions, with the nodes in the centric plane allowing for sliding in the surface only. The nodes in the center were allowed to move only in the vertical direction. For the models of Group L the interface of the testing disc and the substrate epoxy was modeled as a contact surface with a friction coefficient of 0.30, assuming an insufficient bond between these surface, while for the models of Group LS and ES the bond was assumed to be sufficient (no contact surface). The fracture load for the different models, shown in Table 3 , was loaded on the surface of the permanent deformation. The nodes in the bottom of the supporting ring were fixed. The models were composed of 8960–14,464 parabolic tetrahedron solid elements. The material properties used for the analysis were summarized in Table 1 . The Solid Maximum Principal stresses were calculated to establish the maximum tensile stress in the testing disc, since the highest Solid von Mises stresses were in all cases higher than the Solid Maximum Principal stress, the Solid von Mises stress was used to calculate the maximum compressive stress.

Table 2
The deformations (in μm) used in the FEA set-up: total deformation under the universal testing machine, elastic deformation in the FEA, and the calculated permanent deformation.
L 0.5 mm LS 0.5 mm ES 0.5 mm
Total deformation 173 351 225
Elastic deformation 106 229 71
Permanent deformation 67 122 154

Table 3
Means and standard deviations of the fracture load (N) for polished LavaTM Ultimate CAD/CAM (Group L), sandblasted LavaTM Ultimate CAD/CAM (Group LS), and sandblasted IPS e.max CAD (Group ES) discs.
Group 0.5 mm 1.0 mm 1.5mm 2.0 mm 3.0 mm
L 1028 (112) a1 1201 (160) a1 1097 (149) a1 1095 (249) a1 1574 (143) b1
LS 2221 (110) a2 1764 (261) b2 1771 (265) b2 1994 (208) ab2 2174 (389) a2
ES 1418 (314) a3 1516 (309) a12 1613 (429) a2 2288 (270) b2 2754 (241) c3
For each horizontal row: values with identical letters indicate no statistically significant differences ( P > 0.05).
For each vertical column: values with identical numbers indicate no statistically significant differences ( P > 0.05).

Statistics

The values obtained for each subgroup were analyzed by one-way ANOVA, and the Tukey test was adopted for the post-hoc test. The relation between fracture resistance and thickness was analyzed by linear regression. The statistics were done with IBM SPSS statistics 20 (IBM Corp., USA) at a significance level of α = 0.05.

Materials and methods

A simplified tri-layer onlay model was designed to mimic the restoration for posterior tooth ( Fig. 1 ). The testing disc, which represented the occlusal restoration, had a diameter of 10 mm to mimic the average dimension of molars. The testing disc was cemented to the substrate epoxy disc, e.g. simulated dentin, with an equal diameter. The epoxy material (similar to G10, previously used in other studies; see Discussion) had an elastic modulus of 18 MPa, which was similar to dentin . The bonded two-layer disc had a final thickness of 3.5 mm, which was chosen as equivalent to the average thickness from pulp wall to occlusal surface . Then the two-layer disc was bonded to a steel ring with an inner diameter of 6.5 mm, an outer diameter of 10 mm, and a thickness of 1.5 mm mimicking the pulp chamber.

Fig. 1
Schematic representation of the set-up model use for this study.

Fracture resistance

The testing specimens were divided into three groups: L, LS and ES, according to testing materials and surface treatment. The testing discs in Group L were made of Lava™ Ultimate CAD/CAM Restorative (3M™ ESPE™, USA), cemented to polished epoxy substructures (Epoxydplatte, Carbotec GmbH & Co. KG, Aachen, Germany). Specimens in Group LS were also made of Lava Ultimate, with a sandblasted surface and cemented to sandblasted epoxy discs. In Group ES, testing discs were made of an IPS e.max CAD (Ivoclar Vivadent AG, Liechtenstein), also with a sandblasted surface and cemented to sandblasted epoxy discs. Each group was further divided into five subgroups ( n = 8) according to the thickness of the testing disc. The thicknesses of the testing discs were 0.5 mm, 1.0 mm, 1.5 mm, 2.0 mm, and 3.0 mm; the thicknesses of their corresponding epoxy discs were 3.0 mm, 2.5 mm, 2.0 mm, 1.5 mm, and 0.5 mm, respectively, achieving a final thickness of 3.5 mm.

For Groups L and LS, RNC cylinders were made from Lava Ultimate blocks, by a water-cooled bur with an inner diameter of 10 mm. The cylinders were further cut into 80 discs by a sawing machine (Isomet 1000; Buehler, USA), according to the five demanded thicknesses. For Group ES, 40 ceramic discs were made the same way from IPS e.max CAD blocks. 120 epoxy discs were cut from five large epoxy plates, with the corresponding five demanded thicknesses, by the same drilling machine and drill. All the 240 discs were polished by 600 grit SiC polish papers. 160 discs for Groups LS and ES were sandblasted for 10 s on one surface, by 50 μm Al 2 O 3 at a distance about 10 mm. All the 240 discs were cleaned by an ultrasonic machine (Bransonic 3510; Branson Ultrasonics Corp, USA) for 5 min.

The bonding surfaces of the Lava Ultimate discs were treated with mixed Clearfil™ SE bond primer (Kuraray CO. LTD, Japan) and a silane coupling primer (Clearfil Porcelain Bond Activator; Kuraray CO. LTD, Japan) for 10 s, according to manufacturer’s recommendation. The ceramic bonding surface was first etched with 9.5% hydrofluoric acid (PorcelEtch Syringe Green; Cosmedent, Inc., USA) for 20 s, rinsed with water, air dried and then applied 20 s of the mixture of SE bond primer and the porcelain activator. The dentin-like epoxy surface was etched with 40% phosphoric acid (K-Etchant Gel; Kuraray Medical Inc, Japan) for 30 s to clean the surface, rinsed with water and lightly air dried prior to a 10 s application of the SE bond primer. The epoxy discs were bonded to the metal rings by SE bond. For each specimen, the testing disc was then cemented to the corresponding epoxy disc with a resin cement (Panavia F 2.0; Kuraray Medical Inc, Japan). During curing a load of 50 N was applied on the specimen. The cement was cured in five directions, for each direction 30 s, by a curing light (Astralis 10, Ivoclar Vivadent, Liechtenstein). All the specimens were stored in 37 °C water for 24 h.

The completed specimens were loaded with a hemispherical steel indenter (Ø = 4.9 mm), centered on the top surface. The load was applied until failure, with a universal testing machine (Instron 6022, Instron Corp., MA, USA) at a crosshead speed of 1 mm/min. The load (N) at the failure was recorded as fracture resistance, in order to compare with the normal biting forces in the clinical situation. The results of 0.5 mm restorative discs were further used in the finite element analysis (FEA) for each group, for the analysis of the stress distribution.

Finite element analysis

Three dimensional FEA models of the test set-up with the dimensions according to the testing specimens were made (loading sphere models). The Finite Element modeling was carried out using FEMAP software (FEMAP 10.1.1; Siemens PLM software, Plano, Texas, USA), while the analysis was done with NX Nastran software (NX Nastran; Siemens PLM Software, Plano, Texas, USA). The models consisted of the layer of the supporting ring, the supporting epoxy, the testing disc, and the loading sphere. Since the models were symmetrical in geometry in two directions, they were split in quarter specimens to facilitate the border conditions, with the nodes in the centric planes allowing for sliding in the surface only. The contact surface between the loading sphere and the testing disc was modeled as contact surface with a friction coefficient of 0.45. For the models of Group L the interface of the testing disc and the epoxy was also modeled as a contact surface with a friction coefficient of 0.30, assuming an insufficient bond between these surface, while for the models of Group LS and ES the bond was assumed to be sufficient (not contact surface, but a fixed surface). The fracture load for the three models, shown in Table 3 , was loaded on the cut surface of the loading sphere ( Fig. 1 ). The nodes in the bottom of the supporting ring were fixed. The models were composed of 8,730 parabolic tetrahedron solid elements. The material properties from the manufacturer, used for the FEA were summarized in Table 1 . The Solid Maximum Principal stresses were calculated to establish the maximum tensile stress in the testing disc. Since the highest Solid von Mises stresses were in all cases higher than the Solid Maximum Principal stress, the Solid von Mises stress was used to calculate the maximum compressive stress.

Nov 25, 2017 | Posted by in Dental Materials | Comments Off on The fracture resistance of a CAD/CAM Resin Nano Ceramic (RNC) and a CAD ceramic at different thicknesses

VIDEdental - Online dental courses

Get VIDEdental app for watching clinical videos