Fracture strength of machined ceramic crowns as a function of tooth preparation design and the elastic modulus of the cement



To determine, by means of static fracture testing the effect of the tooth preparation design and the elastic modulus of the cement on the structural integrity of the cemented machined ceramic crown-tooth complex.


Human maxillary extracted premolar teeth were prepared for all-ceramic crowns using two preparation designs; a standard preparation in accordance with established protocols and a novel design with a flat occlusal design. All-ceramic feldspathic (Vita MK II) crowns were milled for all the preparations using a CAD/CAM system (CEREC-3). The machined all-ceramic crowns were resin bonded to the tooth structure using one of three cements with different elastic moduli: Super-Bond C&B, Rely X Unicem and Panavia F 2.0. The specimens were subjected to compressive force through a 4 mm diameter steel ball at a crosshead speed of 1 mm/min using a universal test machine (Loyds Instrument Model LRX.). The load at the fracture point was recorded for each specimen in Newtons (N). These values were compared to a control group of unprepared/unrestored teeth.


There was a significant difference between the control group, with higher fracture strength, and the cemented samples regardless of the occlusal design and the type of resin cement. There was no significant difference in mean fracture load between the two designs of occlusal preparation using Super-Bond C&B. For the Rely X Unicem and Panavia F 2.0 cements, the proposed preparation design with a flat occlusal morphology provides a system with increased fracture strength.


The proposed novel flat design showed less dependency on the resin cement selection in relation to the fracture strength of the restored tooth. The choice of the cement resin, with respect to its modulus of elasticity, is more important in the anatomic design than in the flat design.


A major disadvantage of dental ceramic materials relates to their brittle nature owing to their atomic bonds that inhibits the atomic planes to slide apart when subjected to load .

The applied stresses resulting from masticatory loads are concentrated on the inherent flaws that exist in the ceramic, which have the ability to amplify an applied stress leading to rapid crack propagation that can in turn lead to brittle fracture of the ceramics . As a result, it is probable that, ceramic materials fracture at a fraction of their theoretical strength due to the flaws’ stress-raising effect . It has been shown that porosities and microcracks are the sites of fracture initiation .

The inherent properties of the ceramic crown, as a stand-alone item, are of limited interest as it is the overall strength of restored tooth-crown complex that is clinically relevant. The materials used, their geometrical configuration and the manner in which they are integrated and joined to each other and to the dental tissues determine the ability of the restored tooth to withstand the occlusal stresses placed upon them. Thus, in addition to the properties of the ceramic material, the performance of the other constituent parts of the compound system should be considered, principally: the geometrical configuration of the crown (thickness and occlusal cuspal morphology); the quality of the established bond between the all-ceramic crown, the resin cement and the dentin structure; the characteristics of the adhesive lute (dimensions and elastic modulus) and the amount and quality of the remaining dentin structure (support and preservation of pulp vitality). Moreover, while some ceramic systems have truly impressive fracture strength properties (e.g. zirconia core crowns), this is at the expense of aggressive tooth preparations (1.5 mm shoulder and 3 mm occlusal reduction) that compromise the amount of remaining dentin support and pulp vitality.

Considering the ceramic material per se, a wide range of ceramic systems are currently available to select from, based on an equally wide range of fabrication technologies. These ceramics, have been shown to have a fracture strength value that should resist normal functional occlusal loads (150–665 N) ; ranging from 772.3 N for machined feldspathic ceramics to 1000 N for zirconia machined crowns . Machinable ceramics, using CAD/CAM technology, are of interest as they are homogenous and stronger, than conventional sintered porcelains where voids, flaws, and cracks are reduced to minimum and the effects of distortion or shrinkage have been omitted . Moreover, CAD/CAM systems, increase production predictability and reduce the working process and the production cost.

The configuration of the crown (wall and occlusal thickness), as advocated by the manufacturers, is designed with an element of built-in ‘insurance’ (over-compensation) so that its fracture strength is optimized as a stand-alone item, with less regard for other elements of the crown-tooth complex. This increase in the thickness and overall dimensions of the ceramic walls is undertaken at the expense of conservation of tooth structure and preservation of tooth vitality. The actual geometry of the crown, in particular the shape of the tooth preparation design is based on historical empirical design configurations for non-adhesive full coverage crowns as advocated for cast metal and ceramo-metal restorations . To date, little consideration has been given to the performance of all-ceramic adhesive crowns as part of a restored tooth-crown complex and how this can be optimized for the preservation of tooth structure. There is evidence to suggest that the geometry of the crown and the stiffness distribution within it also appears to have an effect on the distribution of the stresses within the tooth-crown complex .

The propagation of cracks in the ceramic crown is affected by the support offered from the underlying tougher and more elastic structures; the cement and the underlying dentin. The adhesive nature of resin-based cements has the effect of covering the internal surfaces of microcracks and small defects of the ceramic restorations; microcracks are thus blunted and inhibited from propagating . In this way, evidence suggests that resin-based adhesive luting agents that bond to the tooth structure and the ceramic restoration can increase the fracture resistance of the restoration complex .

The overall dimensions and physical characteristics of the adhesive lute may also play an important role in stress distribution and crack propagation within the overlying ceramic restoration . However, while various resin cements have been advocated for cementation of all-ceramic crowns there is neither guidance nor consensus in the literature regarding the ideal parameters for optimum performance throughout the tooth restored with an all-ceramic crown. Thickness of the cement layer on the stress levels within the crown depends on the nature of the cement; for a thick cement lute, the stress development is faster if this is a glass ionomer and slower with a resin cement ; which may be on account of the difference of the elastic moduli of these materials. Also, for an elastic resin-based lute the thickness is only an important determining parameter of fracture strength of machined ceramics when this exceeds 300 μm .

Concerning the elastic modulus of the lute, the fracture resistance of ceramic crowns is highly influenced by the high elastic modulus substrates . Moreover, high elastic modulus resin cements, assuming that they have acceptable viscosity and thickness, have the best performance in terms of all-ceramic crown survival .

It is a more clinically relevant research question to consider how these parameters, in isolation or combined, affect the fracture strength of an adhesively bonded ceramic crown-tooth system with due consideration to the preservation of tooth structure.

The aim of this in vitro investigation was to determine, by means of static fracture testing the effect of tooth preparation design and the elastic modulus of the cement on the structural integrity of the cemented machined ceramic crown-tooth complex.

The effect of the cement elastic modulus and the tooth preparation design on the stress state of this same ceramic crown-tooth complex has been investigated by the authors in an earlier study by means of Finite Element Analysis (FEA) . The FEA study revealed significant differences in the stress state that occurs in the crown-tooth complex as a result of both the crown design and the elastic modulus of the cement.

A machined feldspathic ceramic has been selected as a popular machinable ceramic with a low documented fracture strength that can be reproduced in a predictable manner; this way enabling an assessment of the effect of the tooth preparation geometry, the occlusal ceramic thickness and cement modulus, assuming an adhesive interface.

Materials and methods

Tooth preparation

Ethical approval was obtained from the University of Sheffield for this study. Extracted human maxillary premolar teeth were selected for the purpose of this investigation. The teeth were cleaned of any calculus deposits and soft tissue with hand scalers and stored in 0.9% normal saline solution. The teeth were measured in Bucco-Lingual (BL, 9.46 ± 0.11), Occluso-Gingival (OG, 5.12 ± 0.09), and Mesio-distal (MD, 7.44 ± 0.10) directions using a digital calliper (Mitutoyo, Japan), and the teeth with the dimensions beyond these ranges were excluded from the study. Seventy caries free and crack free teeth with the accepted dimensions were implemented.

The teeth were randomly divided into seven groups of 10 teeth each ( n = 10). Ten unprepared and unrestored teeth were kept and tested as a control group ( n = 10). The remaining 60 teeth were assigned to two equal groups ( n = 30), according to the occlusal reduction design: Group #1 Anatomic reduction and Group #2 Flat occlusal reduction ( Table 1 ). Each group of 30 was further sub-divided into three groups ( n = 10) in accordance with three different cementation protocols ( Table 2 ).

Table 1
Preparation guidelines.
Area of reduction Anatomic occlusal reduction Flat occlusal reduction
Occlusal 2 mm from occlusal center 1.2 mm from occlusal center
Axial reduction 1.2–1.5 mm 1.2–1.5 mm
Finishing line 1 mm shoulder 1 mm shoulder
Taper 12° 12°

Table 2
Sub-groups according to cementation protocols.
Group Super-Bond C&B (Group SB) Rely X Unicem (Group RX) Panavia F 2.0 (Group PN)
Hydrofluoric acid gel (5%) Vita Ceramics Etch
Vita ceramics etch, Vita Zahnfabrik, Bad Sackingen, Germany
Silane (Ceramic primer) Porcelain Liner M
Sun Medical, Tokyo, Japan
Rely X Ceramic Primer
Clearfil Porcelain Bond Activator
Kuraray Medical, Inc.
Dentin bonding Heliobond
Ivoclar Vivadent
Cement material Super-Bond C&B
Sun Medical, Tokyo, Japan
Rely X Unicem
Panavia F 2.0
Kuraray Medical, Inc.
Cement chemistry Self-cure dental adhesive resin cement Dual-polymerized self-adhesive universal resin cement Dual-polymerized phosphate-modified resin cement
Elastic modulus of cement (GPa) 1.8 8 18.3

All the specimens were mounted vertically in cylindrical moulds with the roots set in acrylic resin up to the amelo-cemental junction and the long axis of the crown perpendicular to the horizontal plane. Two different designs of preparation were chosen for this experiment: 1 . A standard anatomic occlusal reduction for Group #1 ( Fig. 1 a ) and 2 . A flat occlusal preparation for Group #2, in accordance with the guidelines detailed in Table 1 .

Fig. 1
a: Anatomic design of tooth preparation for all-ceramic crown. b: Flat design of tooth preparation for all-ceramic crown.

Crown fabrication

The “Correlation mode” of the CEREC CAD/CAM (Software version v3.6; SIRONA Dental Systems) was used to replicate the occlusal morphology of the unprepared teeth for the CEREC crowns. All teeth were coated with an optical scanning powder (CEREC powder VITA, Zahnfabric, Germany) before tooth preparation. The samples were fixed to a special base plate provided by the SIRONA Company and an optical impression of the tooth was taken with the inEos laser scanner (SIRONA Dental Systems, GmbH). The models for the different preparations were correlated by placing notches on the acrylic base to allow reproducing the original tooth morphology for the milled crown.

After completion of the preparations, the specimens were powdered and scanned in the same manner as the unprepared ones. Before designing the crowns, the spacer value in the CEREC software design parameters was set to 30 μm .

The restorations were designed using the CEREC inLab software version 3.6. The “Correlation mode” was selected for the purpose of designing the crown. The crown for the anatomic preparation was designed with 2 mm thickness over whole of the occlusal ceramic thickness (over the central fissure and over the cusps). In the flat design, the thickness of ceramic in the central fissure was 1.2 mm and 2 mm over the cusps.

The thickness of the crowns was standardized with the software, so that crowns of the same preparation design would have the same dimensions.

The “Endo mode” was selected for the milling process using the cylindrical bur (REF 58 55 734) and the step bur (REF 60 89 010) according to the manufacturer’s instructions. The Vita Mark II feldspathic ceramic blocks for CEREC/inLab (shade A 2, Vita Mark II, 2M2C I12, 12 mm long, Vita Zahnfabrik, Bad Sackingen, Germany) were used for milling. The cutting diamond burs were changed after milling 10 crowns (five for each group); and the milling unit was calibrated using the CEREC calibration kit at the beginning of the study and whenever the computer software requested it.

Each milled crown was evaluated by means of visual examination (with loupes, ×4 magnification) to exclude any unfavorable restorations from the study. The inclusion criteria were: marginal gap of <150 μm, marginal integrity and correct crown morphology. The marginal fit of the crowns was measured using a traveling microscope (Mitutoyo™) at 30× magnification at four different points in the middle of the buccal, lingual, mesial, and distal wall. Only one crown fell outside these parameters and a replacement was milled.

The ceramic thickness for each crown was standardized so that the ceramic thickness at the central occlusal fossae for Group #1 was verified to have a thickness of 2 mm and for Group #2 this was1.2 mm.

The milled crowns were glazed using an Akzent glazing kit (Vita Zahnfabrik, Bad Sackingen, Germany) to increase the strength of the ceramic following the manufacturer’s recommendations of pre-drying at a temperature of 600 °C, followed by a temperature increase at the rate of 58 °C/min with closing time of 6 min and a final firing temperature of 950 °C with a holding time of 1 min.


Three different cementation protocols were used for the two preparation designs groups; each using cement with a significantly different modulus of elasticity, representative of the low (Super-Bond C&B; 1.8 GPa) , middle (Rely X Unicem; 8 GPa) and high (Panavia F 2.0; 18.3 GPa) values found in current commercially available materials ( Table 2 ).

A standardized cementation protocol, that combined a silane coupling agent, a dentin bonding system and a lute cement, was followed for each group, in accordance with the manufacturer’s recommendations. For all the groups, the internal surface of the crown was cleaned and etched using Hydrofluoric acid 5% (Ceramic etching gel; Vita ceramics etch, Vita Zahnfabrik, Bad Sackingen, Germany) for 120 s, and then thoroughly washed for 60 s; the outer surface of the ceramic was protected with a wax coating. All the prepared tooth surfaces were cleaned with a rubber cup and flour of pumice in a slow speed motor hand-piece washed with water spray (3-in-1 syringe) and dried with oil-free air. Specific variations in the cementation protocols, as per the manufacturer’s recommendations, were: (i) teeth in the Super-Bond C&B group were etched with phosphoric acid gel (Jumbo Total etch blue, 37% phosphoric acid) for 30 s, washed and dried to remove excess water, but not over dried; (ii) the cement in the Rely X Unicem and the Panavia F 2.0 group was photo-polymerized for 40 s for each of the five coronal surfaces with a calibrated halogen unit (Optilux 501, Kerr Demetron, Danbury, CT); (iii) the external interface crown margins of the specimens in the Panavia F 2.0 group were protected from oxygen inhibition with an air sealing gel for 3 min (Oxygaurd, Kuraray Medical, Inc.).

The cementation of all specimens was carried out in a universal materials testing machine under a load of 40 N for 3 min (Lloyds Instrument Model LRX) . A specially designed jig was constructed that housed a silicone putty impression material with an imprint of the occlusal surface of the crown (Aquasil Putty, Dentsply-Detrey, Konstanz, Germany). This was assumed to standardize the clinical cementation procedure using finger pressure ( Fig. 2 ).

Fig. 2
Cementation procedure using Tensometer.

All the cemented samples were left undisturbed for 1 h and then stored in distilled deionized water for 24 h prior to testing.

Fracture test

A universal materials test machine (Lloyds Instrument Model LRX) was used to undertake the fracture tests of all the specimens (control and test groups). The specimens were firmly retained in the fixed base grip of the test machine. A static compressive axial load was applied to the central occlusal surface of the ceramic crown at a crosshead speed of 1 mm/min via a 4.24 mm diameter steel ball with an interspaced sheet of rubber dam (1 mm thickness) designed to act as a stress breaker between the all-ceramic crown and the occluding steel ball ( Fig. 3 ).

Fig. 3
Load application through the rubber dam as stress breaker.

For each specimen, the maximum fracture was recorded in Newtons (N).

Mode of fracture

The fracture mode of the samples after the fracture strength test was reported according to Burke’s classification as shown in Table 3 .

Nov 25, 2017 | Posted by in Dental Materials | Comments Off on Fracture strength of machined ceramic crowns as a function of tooth preparation design and the elastic modulus of the cement
Premium Wordpress Themes by UFO Themes