Test the hypothesis that monolithic ceramics can be developed with combined esthetics and superior fracture resistance to circumvent processing and performance drawbacks of traditional all-ceramic crowns and fixed-dental-prostheses consisting of a hard and strong core with an esthetic porcelain veneer. Specifically, to demonstrate that monolithic prostheses can be produced with a much reduced susceptibility to fracture.
Protocols were applied for quantifying resistance to chipping as well as resistance to flexural failure in two classes of dental ceramic, microstructurally-modified zirconias and lithium disilicate glass–ceramics. A sharp indenter was used to induce chips near the edges of flat-layer specimens, and the results compared with predictions from a critical load equation. The critical loads required to produce cementation surface failure in monolithic specimens bonded to dentin were computed from established flexural strength relations and the predictions validated with experimental data.
Monolithic zirconias have superior chipping and flexural fracture resistance relative to their veneered counterparts. While they have superior esthetics, glass–ceramics exhibit lower strength but higher chip fracture resistance relative to porcelain-veneered zirconias.
The study suggests a promising future for new and improved monolithic ceramic restorations, with combined durability and acceptable esthetics.
More than a decade of clinical trials recording survival rates for all-ceramic posterior crowns and fixed dental prostheses (FDPs) indicate vulnerabilities to various failure modes . The trend in circumventing this problem has been toward a strong and tough yttria tetragonal zirconia polycrystal (Y-TZP) core veneered with an esthetic porcelain . Such bilayer systems have several major drawbacks: (i) their fabrication is a multistep process; (ii) the veneer has a low toughness and is consequently susceptible to chipping ; (iii) the bonding between veneer and core can be weak relative to the toughness of the constituent material layers, with ensuing potential for delamination and (iv) residual tensile stresses can develop during the veneering process, further degrading the porcelain and its bond to the zirconia core . Veneered zirconia prostheses do not appear to perform as well as their veneered metal counterparts .
The obvious way to circumvent all these drawbacks is to replace the veneer/core bilayer with a monolithic ceramic. This has not been straightforward, because the microstructural qualities that confer good mechanical properties do not lend themselves to good esthetics, and vice versa. However, some approaches are emerging with the promise of improved clinical performance. One such is to begin with translucent but inherently weak glass–ceramics, such as Dicor , and to refine the compositions and microstructures to produce a tougher ceramic without compromising the esthetics. The lithium disilicate IPS e.max glass–ceramics manufactured by Ivoclar-Vivadent fall into this category . Those materials have strong needle-like crystals embedded within a glass matrix, mimicking natural enamel and thereby inhibiting crack propagation. They have performed well in crown applications . However, they are not as strong as the zirconias, and are less suitable for applications where stress concentrations can be high, e.g. FDP connectors .
An alternative approach has been to begin with a strong and dense zirconia, and to manipulate additive components and heat treatments to produce an acceptable translucency. The monolithic zirconias BruxZir by Glidewell and LAVA Plus by 3 M ESPE are examples. The translucency of BruxZir is achieved via the elimination of light-scattering alumina sintering aids and porosities, along with the utilization of a higher sintering temperature (1530 °C) and longer dwell time (6 h). The translucency of LAVA Plus is attained by reducing alumina sintering aids, but also by increasing the density of the green compact in order to reduce the sintering temperature (1450 °C) and dwell time (2 h), with resultant finer grain size. Glass-infiltrated zirconia (GZG) is another route to improved translucency . While remaining highly fracture resistant, zirconia-based ceramics do not yet match the esthetics of the glass–ceramics.
In assessing potential lifetimes of any material type it is important to consider the different modes in which fracture may occur. Foremost among these modes is top-surface chipping and subsurface flexural fracture . Resistance to one mode does not necessarily imply resistance to the other. The hypothesis in this work was that monolithic crowns can sustain uncommonly high bite forces, provided certain dimensional requirements are met. To test that hypothesis, the chipping resistance and flexural strength were evaluated for candidate lithium disilicate glass–ceramics and modified zirconia materials. To place the results into perspective, comparative data for a veneering porcelain and unmodified Y-TZP were also evaluated.
Materials and methods
Material selection and specimen preparation
The restorative materials selected for this study are listed in Table 1 , along with values of Young’s modulus E , toughness T and strength S . The properties data for commercial veneering porcelain, as well as for Y-TZP and glass-infiltrated zirconia (GZG), have been documented in earlier reports .
|Material||Manufacturer||Modulus E (GPa)||Toughness T (MPa·m 1/2 )||Strength S (MPa)|
|Zirconia (3 mol% Y-TZP)||In-house||216 (4) a||3.65 (0.2) b||1050 (113) c|
|GZG (graded glass/zirconia/glass)||In-house||210 (4) a||3.76 (0.3) b||1370 (84) c|
|Lithium disilicate (Press)||Ivoclar-Vivadent||95 (5) f||1.40 (0.13) b||281 (20) d|
|Lithium disilicate (CAD)||Ivoclar-Vivadent||95 (5) f||1.31 (0.17) b||376 (57) d|
|Feldspathic porcelain (Vita VM9)||Vita Zahnfabrik||66 (3) g||0.94 (0.13) b||100 (12) g|
|Dental cement (Multilink Automix)||Ivoclar-Vivadent||7.0 (0.4) f||–||–|
|Composite (Z100)||3 M||16 (1) h||1.4 (0.1) h||120 (12) h|
|Human dentin||18 (2) e||–||–|
Zirconia plates were fabricated in-house from 3 mol% Y-TZP powder (TZ-3Y-E grade, Tosoh, Tokyo, Japan) and sintered at 1450 °C for 2 h in air at a heating/cooling rate of 10 °C/min. Glass-infiltrated zirconia (GZG) plates were produced by presintering the same Y-TZP powder compacts at 1350 °C for 1 h followed by infiltration at 1450 °C for 2 h to a depth of 100 μm with a glass of matching coefficient of thermal expansion, at a heating/cooling rate of 14 °C/min . All specimens were fabricated as plates of lateral dimension 10 mm × 10 mm. Those to be used for chipping tests were 2 mm thick ( n = 4, each material). Those for flexural testing were prepared with thicknesses 1 mm, 1.5 mm and 2 mm ( n = 6, each material and each thickness). One set was prepared with thickness 0.5 mm and subsequently veneered with porcelain (VM9, Vita, Bad Säckingen, Germany) by Marotta Dental Laboratories to provide porcelain/zirconia bilayers of total thicknesses 1.0 mm, 1.5 mm and 2.0 mm ( n = 6, each thickness).
Lithium disilicate glass–ceramics were obtained from the manufacturer (IPS e.max, Ivoclar-Vivadent, Amherst, NY). These came in two forms, Press and CAD, reflecting differences in processing condition . The IPS e.max Press ingots were heat-pressed at 915 °C for 15 min in a EP 600 furnace (Ivoclar Vivadent) and then divested. A reaction layer from the heat treatment was removed by immersing pressed specimens in an aqueous solution containing 0.6% hydrofluoric acid and 1.7% sulphuric acid followed by blasting with Al 2 O 3 particles (100 μm at 0.2 MPa pressure). The IPS e.max CAD ingots were first heat treated to form partially crystallized glass–ceramic blocks, and then were CAD/CAM machined to shape. These were then heated to 840 °C for 7 min for final crystallization. Specimens were fabricated as bars 2 mm × 3 mm × 25 mm for chipping ( n = 6, each material) and flexure ( n = 20, each material) testing. Additional IPS e.max CAD plate specimens 10 mm × 10 mm were prepared for flexure testing ( n = 20).
Data for a common dental cement (Multilink Automix, Dual-Cure, Ivoclar-Vivadent) used to bond zirconia-based and lithium disilicate specimens to a 4 mm thick composite base (Z100, 3 M ESPE, St. Paul, MN), as well as for natural dentin itself, are included in Table 1 . This composite base was used to stabilize the specimens in the chipping tests, and to provide a dentin-like support substrate in the flexure tests . Prior to cementation, top and adjacent side surfaces of the ceramic plates for the chipping tests were first polished to 1 μm finish. The intaglio surfaces of the zirconia plates for flexure tests were likewise pre-polished, while those for the lithium disilicate and GZG plates were pre-etched (4% HF for 30 s and 5 min, respectively).
Scanning electron microscope images of each restorative material were then taken to reveal the microstructures. Sections of the specimens selected for examination were polished to 1 μm, and then either thermally etched (zirconia, 1250 °C for 20 min) or chemically etched (4% hydrofluoric acid – 20 s for e.max Press and CAD, 30 s for porcelain, 5 min for GZG).
Critical forces to produce chipping
Following previous protocols used to evaluate the Y-TZP, GZG and porcelain materials listed in Table 1 , chipping tests were carried out on the lithium disilicate bars . A Vickers diamond pyramid was located normally at specified distances h from a side wall, at crosshead speed 0.1 mm min −1 , with indentation diagonals parallel and perpendicular to the edge. The critical forces P C at which a chip spalled off the side wall were then recorded for each indent. These forces are directly proportional to the ceramic toughness T , as expressed in the chipping relation .
P C = β T h 3 / 2
where β = 9.3 is a dimensionless constant.
Critical forces to produce flexural fracture
Estimates of the capacity for each material system to sustain flexural stress from the applied loading were calculated using documented flexure formulae for flat-layer systems cemented onto dentin-like substrates. Flexure tests were performed on a universal testing machine (Model 5566, Instron, Norwood, MA), with contact load applied at the top surface with a hard tungsten carbide spherical indenter at crosshead speed 1 mm min −1 . A thin plastic sheet was placed between the indenter and the ceramic surface to prevent fracture from spurious contact-induced top-surface cone cracks. The critical load for the onset of radial cracking at the intaglio surface was recorded as a load drop. This kind of subsurface fracture occurs when the tensile stress at the intaglio surface of the ceramic layer equals the strength S of the material. (The radial cracks continue to expand outward at the inner surface of the ceramic layer with further increase in load, ultimately breaking through to the top surface .) For a monolithic ceramic layer (C) of thickness d and modulus E C on a compliant substrate (S) of modulus E S , the critical loads for flexural fracture are given by .
P F = 1.35 S d 2 log ( E C / E S )
Likewise, for a veneer/ceramic (V/C) bilayer of net thickness d = d V + d C free of any residual expansion mismatch stresses on the same compliant substrate (S), the corresponding critical loads are .
P F = 1.35 S d 2 ( E C / E * ) log ( E * / E S )