- Email: [email protected]
Initial versus final fracture of metal-free crowns, analyzed via acoustic emission
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Initial versus final fracture of metal-free crowns, analyzed via acoustic emission Nadia Ereifej a , Nick Silikas a,b,∗ , David C. Watts a,b a b
Biomaterials Science Research Group, School of Dentistry, The University of Manchester, UK Photon Science Institute, The University of Manchester, UK
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. To discriminate between initial and final fracture failure loads of four metal-free
Received 22 January 2008
crown systems by the conjoint detection of acoustic emission signals during compressive
Received in revised form
loading.
6 March 2008
Methods. Teeth were prepared and used for crown construction with four crown systems;
Accepted 22 April 2008
Vita Mark II (VM II) (Vita Zahnfabrik), IPS e.max Ceram/CAD (CAD) (Ivoclar-Vivadent), IPS e.max Ceram/ZirCAD (ZirCAD) (Ivoclar-Vivadent) and BelleGlass/EverStick (BGES) (Kerr/Stick Tech Ltd.). All samples were loaded in compression via a Co/Cr maxillary first molar tooth at
Keywords:
0.2 mm/min and released acoustic signals were collected and analyzed. A minimum number
Acoustic emission
of 15 crowns per group were loaded to final failure and values of loading at initial and
Ceramics
final fracture were compared. Additional four samples per group were loaded till fracture
Fiber-reinforced composites
initiation and were fractographically examined under the optical microscope.
Metal-free crowns
Results. A lower threshold of 50 dB was selected to exclude spurious background signals.
Mechanical testing
Initial fracture forces were significantly lower than those of final fracture (p < 0.05) in all
Microcracks
groups and initial failure AE amplitudes were lower than those of final fracture. Mean initial
Fracture strength
fracture force of ZirCAD samples (1029.1 N) was higher than those of VMII (744.4 N), CAD (808.8 N) and BGES (979.7 N). Final fracture of ZirCAD also occurred at significantly higher force values (2091.7 N) than the rest of the groups; VMII (1120.9 N), CAD (1468.9 N) and BGES (1576.6 N). Significantly higher values of initial failure AE amplitude were found in VMII than CAD and BGES while those of final fracture were similar. All crowns observed under the microscope at initial fracture had signs of failure. Significance. Whereas the metal-free crowns examined showed significant variations in final failure loads, acoustic emission data showed that they all manifested initial failures at significantly lower load values. © 2008 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Increasing demands for restorative dental materials that combine pleasing aesthetics and optimal physical and chemical properties have led to the introduction of various types of
dental ceramics [1]. Fiber-reinforced composites are alternative materials to ceramics and have been successfully used in dental restorations [2]. Intra-oral restorations are challenged by complex loading conditions including both functional and parafunctional
∗ Corresponding author at: Biomaterials Science Research Group, School of Dentistry, Higher Cambridge Street, M15 6FH Manchester, UK. Tel.: +44 16 12756474. E-mail address: [email protected] (N. Silikas). 0109-5641/$ – see front matter © 2008 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2008.04.010
1290
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
stresses and fracture is still one of the most common causes of failure of these restorations [3–6]. Fracture occurs when a particular stress limit is exceeded after which brittle materials, like ceramics, fail abruptly while more ductile materials, like composites, undergo a phase of elastic deformation before they fail [7]. Therefore, the load-bearing capacity of a particular material is an important criterion influencing the performance and serviceability of this material in the clinical situations [8]. Various testing methods have been used to investigate the mechanical properties of dental materials. However, strength is not an inherent property of a certain material and it is dependant on the condition of the tested material in addition to other factors like stressing rate, stress concentration, specimen size and failure mode. Consequently, testing conditions should be controlled very carefully [9,10]. Many concerns have been raised lately regarding the clinical relevance of simple traditional mechanical testing methodologies. These tests used unrealistically high loading values and the modes of failure were different than those reported in real clinical situations. Furthermore, the results of these tests lack significant information considering the site, time of fracture initiation and fracture mechanics [11,12]. The concept of Acoustic Emission (AE) was first proposed by Joseph Kaiser in the early 1950s and later gained wider acceptance in the 1960s [13]. AE is a sound wave produced as a result of energy release from a material at which external stimuli are applied. Upon loading, areas with flaws develop high stress energies and some of this energy is released in the form of a pressure wave or sound which occurs long before the specimen fails catastrophically. As these waves propagate through the surface of the object being tested, they are collected by piezoelectric transducers coupled on the surface that convert part of these waves into electrical charges and then pass them through filters in order to eliminate background noise, then through amplifiers and finally to a computer processing system for data analysis [13–16]. The AE technique offers the advantages of being a non-stop technique which can monitor the condition of the material under investigation throughout the test. It is non-localized and has the ability to examine large volume objects at the same time [17]. Combined with conventional fracture testing machines, AE can identify failure initiation, the initial site of damage, damage propagation, and catastrophic failure of the material and help elucidate the complex failure mechanism. It is considered as an early warning of failure of the material, regarded to have high sensitivity as it registers even the small cracks as long as a noise is generated and thus is claimed to provide real time data [16,18,19]. AE is widely used in material testing and has been applied in the analysis of fracture behavior of different types of biomaterials like ceramics [7,20], dental composites [18,21–24] and porcelain [25]. However, AE technique was rarely employed in fracture testing of dental ceramics. The aim of this study was to investigate the fracture behavior of different all-ceramic and fiber-reinforced composite materials used in prosthodontics applications. The specific objectives were
(1) To use AE technique during compressive strength testing. (2) To select AE parameters for fracture analysis and set a threshold for AE signals reflecting crack initiation. (3) To compare the values of loading at initial and final failure of the materials tested using AE analysis. The null hypotheses were (i) AE cannot detect initial fracture of all-ceramic and fiberreinforced composite materials. (ii) There are no significant differences in initial and final fracture strength values or amplitudes of released AE signals among the different materials.
2.
Materials and methods
2.1.
Sample preparation
Human mandibular third and second molar teeth were collected from several dental practices, disinfected with formalin, cleaned of debris and then kept in water to prevent dehydration. Upon examination of the collected teeth, only sound teeth which had dimensions falling within two standard deviations of the means (of M-D and B-L dimensions) were included. A mandibular second molar plastic tooth with average dimensions, used as a master tooth, was prepared using a high speed air turbine with coarse grit round end diamond bur. Preparation was performed so that all undercuts were eliminated, occlusal surface was reduced by 2 mm with a functional cusp bevel, all sharp angles were rounded and a 1 mm deep chamfer finish line was prepared and located 1 mm coronal to cemento-enamel junction. A Celay machine (Celay, Mikronna Technologie AG, Spreitenbach, Switzerland) was used to prepare the specimen teeth in this study. The master tooth was attached to the machine at one side and the specimen tooth was attached on the other side. While the master tooth was being scanned, the specimen tooth was being milled, so that at the end of the procedure, the milled tooth had an identical preparation to the master tooth. Finally, the dimensions of the prepared teeth were verified using a caliper to ensure that all the specimens had similar standardized dimensions (M-D = 9.4 ± 0.14, B-L = 8.4 ± 0.14). Prepared teeth were numbered and an impression was taken for each tooth using an addition-cured polyvinylsiloxane impression material (Provil, Heraeus Kulzer, Germany) via a one stage putty/wash technique. The impressions were cast using vacuum-mixed dental stone. Dies were then randomly divided into 4 groups and used for crown construction. All crown fabrication procedures were performed by one trained operator. Table 1 shows the materials used, the manufacturers and the number of samples in each group. For the construction of Vita Mark II (VMII) crowns optical images were taken for the dies of the prepared teeth and were then used to design full-veneer crowns of a mandibular second molar with CEREC 3 CAD software, Version 280 r2400 (Sirona Dental System GmbH, Fabrikstrasse, Germany). The restora-
1291
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
Table 1 – The different materials used in the study Group 1 2 3 4
Number of samples
Material
24 22 25 27
Vita Mark II (VMII) IPS e.max Ceram/CAD (CAD) IPS e.max Ceram/ZirCAD (ZirCAD) BelleGlass/EverStick (BGES)
tions were then milled from Vita Mark II blocks at the milling unit of the machine. IPS e.max Ceram/IPS e.max CAD (CAD) samples were prepared as the following: cores with minimum thicknesses of 1 mm occlusally and 0.8 mm circumferentially were designed. These were milled from IPS e.max CAD blocks and fired to achieve complete crystallization in a Programat P10 furnace (Ivoclar-Vivadent, Schaan, Lichtenstein). The cores were then veneered with IPS e.max Ceram to a thickness of 2.0–2.5 mm. For the construction of IPS e.max Ceram/IPS e.max ZirCAD (ZirCAD) samples, cores (minimum 0.7 mm occlusally and 0.5 mm circumferentially thick) were designed and milled from IPS e.max ZirCAD blocks using CEREC 3 InLab machine. These were then placed in a special furnace, Sintramat (Ivoclar-Vivadent, Schaan, Lichtenstein) for sintering. A layer of ZirLiner was added and fired to enhance bonding with the veneering porcelain and veneering was consequently performed using IPS e.max Ceram. BelleGlass/EverStick (BGES) crowns were constructed as follows: two layers of EverStick Net were cut to the appropriate size, adapted onto the dies and light cured for 40 s. BelleGlass HP composite (BG) was placed on top of the fiber cores and cured for 40 s. The partially cured crowns were then placed in a special BG oven and cured in a nitrogen chamber under 145 ◦ C and 60 psi pressure. All crowns were then adhesively cemented to their corresponding teeth using Multilink resin-based dual-cured cement (Ivoclar-Vivadent, Schaan, Lichtenstein). Before cementation, the teeth and the inner surfaces of the crowns were prepared as recommended by the manufacturers. The samples were then stored in water at 37 ◦ C for 24 h before they were tested.
2.2.
Fracture strength (quantitative)
Load was applied on the specimens using a Zwick/Roell Z020 Universal Testing Machine (Zwick GmbH & Co. KG, Ulm, Germany) shown in Fig. 1. Specimens were placed at the lower part of the machine while the upper one used a cast Co–Cr upper molar tooth moving at 0.2 mm/min speed for load application. A 2-channel AE system (Physical Acoustics Corporation, NJ, USA) was used to detect audible signals produced while the specimens were being loaded. One of the sensors was fixed at the metal base of the lower component of the Zwick machine just next to the sample and the other was attached to the loading component of the Zwick machine. Signals detected by the sensors were passed through preamplifiers of 40 dB gain with band pass of 100 kHz to 2 MHz and a threshold set at 50 dB. The data obtained from the AE machine and those obtained from the Zwick machine were combined and used to analyze the values of load at which initial cracking and final failure occurred. A graph of load and AE amplitude versus time was
Manufacturers Vita Zahnfabrik, Bad Sackingen-Germany Ivoclar-Vivadent, Schaan, Lichtenstein Ivoclar-Vivadent, Schaan, Lichtenstein Kerr, CA, USA/Stick Tech Ltd., Oy, Turku, Finland
obtained for each sample using Sigmaplot software (V 8.0) as shown in Fig. 2. The exact time at which the first AE signal ≥50 dB was released was obtained and the value of loading force (N) at this point was extracted and regarded as the value of initial failure. Further loading resulted in crack propagation, thus further release of AE signals continued until final failure occurred at which loading stopped automatically and AE signals with high amplitudes were released. The value of loading at this point was considered as the final fracture force. The same procedure was performed for all the samples in the four groups under investigation.
2.3.
Optical microscope (qualitative)
For four samples per group, the test was stopped when an AE signal of an amplitude ≥50 dB was recorded. The crowns were then tested visually and microscopically using Meiji EMZ-TR optical microscope (Meiji Techno Co., Ltd., Tokyo, Japan) at a maximum magnification of 40× to detect signs of failure. Those which did not have obvious initial cracks on the occlusal surfaces were sectioned to identify subsurface signs of failure. The rest of the samples of all the groups were tested till final failure.
2.4.
Statistical analysis
Wilcoxon-Signed ranks and paired t-tests using SPSS (Version 13.0) were conducted to detect significant differences between the initial fracture force and final fracture force values in each group. A series of Mann–Whitney tests and t-independent tests was conducted to find statistically significant differences in the initial and final fracture values between the different groups. Furthermore, a series of paired t-tests was used to detect significant differences between the average initial fracture AE signal amplitude and that of final fracture in each group followed by one-way ANOVA test to detect differences in AE amplitudes between the different materials.
3.
Results
3.1.
Quantitative findings
Table 2 shows the means and standard deviations of initial and final fracture force values (N), the percentage of initial fracture load to the final fracture load and AE amplitudes (dB) at initial as well as those at final fracture for all materials. These values are also represented in Figs. 3 and 4. Initial fracture values were always significantly lower than final fracture values with the highest difference among ZirCAD samples (1052.6 N) and the lowest in VMII (376.6 N).
1292
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
Fig. 1 – One of the crown samples being tested. (a): The basic set up of the test. (b): A drawing showing the maximum intercuspation between the loading tooth and the specimen.
Table 2 – Mean and standard deviation values for force and AE amplitude of initial and final fracture of the materials tested Material VMII CAD ZirCAD BGES
Initial fracture force (N)
Final fracture force (N)
% of initial to final force
744.4 (151.8) 808.8 (259.2) 1029.1 (333.5) 979.7 (243.9)
1121.0 (248.8) 1469.0 (290.6) 2081.7 (552.8) 1576.6 (327.4)
66.4% 55.0% 49.4% 62.1%
Correlation and regression analyses were performed to detect the relationship between initial fracture strength and final fracture strength. These are illustrated in Fig. 5. Although the correlation was found insignificant (p > 0.05), values of final fracture force tended to increase as the values of initial fracture force increased (r2 = 0.82). VMII had the lowest initial fracture value (744.4 N), followed by CAD (808.8 N). Both were significantly lower than BGES (979.7 N) and ZirCAD (1029.1 N). Mean final fracture force of VMII (1120.9 N) was significantly lower than that of CAD (1468.9 N), ZirCAD (2091.7 N) and BGES (1576.6 N) while that of ZirCAD was significantly higher than the rest of the groups. The initial fracture AE amplitude was significantly lower than the final fracture AE amplitude in all the groups. Comparing these values between the different groups, significantly higher values of initial fracture AE amplitude were found for VMII than those of CAD and BGES while the values of AE amplitude of the final fracture were similar.
3.2.
Qualitative findings
Visual and microscopic examination of the crowns that were loaded till initial failure showed the presence of small cracks within most of the samples tested as shown in Fig. 6. Five crowns, especially within VMII group, had clear cracks sometimes even visible with the unaided eye while in the rest of samples, especially in CAD, ZirCAD and BGES groups, smaller cracks were identified and in two samples, cracks were only located using the optical microscope at high magnification (up to 40×). Four samples of these groups did not have clear cracks on their occlusal surfaces but some changes in the opacity of
Initial fracture amplitude (dB) 66.8 (12.7) 56.4 (5.9) 60.5 (9.5) 57.6 (9.5)
Final fracture amplitude (dB) 90.0 (8.2) 87.0 (8.5) 94.8 (10.2) 88.6 (9.4)
one or more of the cusps were noticed indicating possible subsurface cracking or failure. Sectioning of these crowns at areas of suspected hidden failure revealed the presence of cracks that were initiated at cement/core or core/veneer interfaces.
4.
Discussion
In this study, the AE technique was employed in combination with compressive strength testing of four different types of materials used in the construction of crowns and fixed partial dentures. The technique was used to detect the time at which failure was initiated and therefore determine the loading force that caused microcrack initiation. Monitoring acoustic signals released upon further loading revealed information about further cracking and failure propagation. Finally, AE results combined with those of the loading machine were used to determine the values of loading at which final fracture occurred. Failure process starts as soon as the first microcrack is initiated which upon further loading and fatigue intra-oral conditions propagates and ultimately leads to final fracture. The presence of these microcracks weakens the structure of the material and results in its failure even at lower stress levels than those at which they were initiated. Therefore, recording the loading values at which initial microcracks occur might be the more accurate way of measuring the strength of a particular material bringing the results of the test closer to the real clinical situation. In this study, the ability of AE technique to detect initial fracture and final failure was demonstrated and therefore the first null hypothesis was rejected. Furthermore, statistically significant differences were found between the different materials tested regarding values of initial and
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
1293
Fig. 3 – Values of force (N) for initial and final fracture for all materials.
Fig. 4 – AE amplitude of initial and final fracture for all materials.
tended to fracture into many fragments in a pattern that does not exist clinically [11]. In this study, a cast Co/Cr model of an upper second molar was used to load the samples to failure. Before loading, maximum interdigitation between the loading tooth and the sample tooth was achieved mimicking their intercuspation in a real clinical situation. Upon analysis of the fractured crowns, the patterns of failure found were not very different from those observed clinically. Most of the crowns had a fragment or two chipped off, delamination of the veneer
Fig. 2 – Graphs of force (N) and AE amplitude (dB) versus time (s) of a representative specimen for each Group. (a): CAD, (b): VMII, (c): ZirCAD, (d): FRC. (): force at initial fracture, (䊉): force at final fracture.
final fracture force and the amplitudes of the released acoustic signals and thus the second null hypothesis was rejected. Most studies investigating the compressive strength of materials used metal spheres of different dimensions to apply load on tested samples [26,27]. These might not be able to distribute forces evenly on the samples and therefore samples
Fig. 5 – A plot of final fracture force versus initial fracture force.
1294
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
Fig. 6 – An image of a ceramic crown (VMII) with visible initial cracking (a), and a graph showing AE signal release at initial fracture of the crown (b).
porcelain was common in bilayered crowns and only very few samples broke into many pieces or had the underlying teeth involved in fracture. Signs of initial failure were found in all samples examined after initial fracture. Failure of VMII crowns seemed to start at their occlusal surfaces as cracks were evident there while in bilayered samples, (CAD, ZirCAD and BGES), cracks were smaller, more difficult to locate or not even visible on the outer surfaces of the crowns. Occlusal cone cracks created on the outer surfaces of the veneer layers of the crowns around the points of load concentration were found to be the most common modes of failure of all-ceramic crowns. However, cracks initiating at core/veneer interfaces or at core/cement interface were also reported in bilayered ceramic samples [28]. These more complex modes of failure in bilayered structures might explain the different microscopic appearance of bilayered crowns tested at initial failure. However, all these microscopic findings have supported the validity of the preset threshold with further evidence and proven the ability of AE technology to detect initial fracture even if it occurs in subsurface layers. Alander et al. [18] investigated the acoustic signals emitted during flexural strength testing of different types of fiberreinforced composites and found that AE signals were first released at values of loading that are 19–32% lower than the final failure points and this difference was statistically significant. Analysis of the results of the crowns tested till final fracture in this study showed that initial fracture, detected by the AE technique, always started at significantly lower loading values (33.6–51.6%) than the final loading values. This indicated that failure actually started much earlier than the
final fracture point and that loading machines recording only forces at catastrophic failure have overestimated the fracture strength of tested materials. However, a general tendency for the values of final fracture force to increase as the values of initial fracture force increased was observed and therefore initial failure point can be used to predict the final failure one. Assuming that the maximum biting force expected in normal situations in the back of the mouth is about 500 N, the four types of materials (with only a few samples per group that failed at low stress levels which can be related to manufacturing-induced defects) had average values of initial fracture that are higher than that value [29]. Therefore, it can be concluded that all materials could be potentially used in the construction of posterior full veneer crowns. Nevertheless, proper case selection and following the manufacturers’ recommendations regarding the construction processes and the indications of these materials are important factors for the success of such restorations. AE release patterns after crack initiation and before final fracture varied among the samples of the different materials. In VMII, most of the samples had limited AE activity during crack propagation indicating fewer events occurring within the samples at fracture. This can be explained by the structure and manufacturing process of these samples. These were milled using CAD/CAM machines and are composed of a homogenous structure of fine-particle feldspathic porcelain. On the other hand, in bilayered samples, higher AE activity was recorded during the fracture process as cracks propagated through the different layers of the crowns or at the interfaces between the different components. The fact that the veneer layers of these crowns were manually constructed adds to the complexity of their structure and might have induced more defects within their structure than machined samples. Construction-induced flaws might have increased AE activity of these multilayered samples. Furthermore, much fewer signals were recorded during fracture process of BGES group than all-ceramic groups. This can be related to the more flexible nature of resin-composite materials compared to brittle ceramics. Composites undergo a phase of plastic deformation before they break catastrophically. This plastic deformation is a relatively quiet process and few AE signals, above the preset threshold, are expected to be recorded. On the other hand, allceramic materials are more brittle and do not undergo plastic deformation. When loading exceeds a particular elastic limit, fracture occurs. These microfractures are noisy and therefore higher AE activity is usually recorded. At final fracture of the all-ceramic samples, parts of the crowns were chipped off. This fracture was accompanied by the release of AE signals of very high amplitude. In BGES group, the fractured fragments of the crowns were kept in place by the underlying reinforcing fibers and the more flexible nature of BG composite resulted in a gradual rather than a sudden drop of the loading force as observed in the other groups. This can be attributed to the presence of fibers that seemed to break gradually. High fracture resistance of ZirCAD group can be explained by the toughening mechanism of Y2 O3 -stabilized zirconia-based core of this group. AE amplitude at initial failure was found to be significantly lower than that at final fracture within all the groups. Signals released at initial failure were caused by small cracks while
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
those at final failure were caused by catastrophic fracture that involved in many cases total chipping of one or two cusps of the crowns. Therefore, it can be concluded that AE amplitude might be an indication of the severity of the fracture.
Acknowledgment The authors are grateful to Ivoclar-Vivadent (Schaan, Lichtenstein) for providing IPS e.max materials and equipment and Mr. Adrian Bennett for providing the materials and equipment for the construction of VMII samples.
references
[1] McLaren EA, White SN. Glass-infiltrated zirconia/alumina-based ceramic for crowns and fixed partial dentures: clinical and laboratory guidelines. Quintessence Dent Technol 2000;23:63–75. [2] Brown D. Fibre-reinforced materials. Dent Update 2000;27:442–8. [3] Walton JN. A survey of crown and fixed partial denture failures: length of service and reasons for replacement. J Prosthet Dent 1986;56:416–21. [4] Brochu JF, El-Mowafy O. Longivity and clinical performance of IPS Empress Ceramic restorations—a literature review. J Can Dent Assoc 2002;68(4):233–7. [5] Oden A, Andersson M, Krystek-Ondracek I, Magnusson D. Five-year clinical evaluation of Procera AllCeram crowns. J Prosthet Dent 1998;80(4):450–6. [6] Taskonak B, Sertgoz A. Two-year clinical evaluation of lithia-disilicate-based all-ceramic crowns and fixed partial dentures. Dent Mater 2006;22(11):1008–13. [7] Lawn BR. Strength and reliability. In: Davis EA, Ward IM, editors. Fracture of brittle solids. 2nd ed. Cambridge: Cambridge University Press; 1993. [8] Darvell BW. Mechanical testing. In: Darvell BW, editor. Materials Science for Dentistry. 7th ed. Hong Kong: Darvell BW; 2002. [9] Kelly JR. Perspectives on strength. Dent Mater 1995;11(2):103–10. [10] Kelly JR. Dental ceramics: current thinking and trends. Dent Clin North Am 2004;48(2):513–30. [11] Kelly JR. Clinically relevant approach to failure testing of all-ceramic restorations. J Prosthet Dent 1999;81(6): 652–61. [12] Dong XD, Darvell BW. Stress distribution and failure mode of dental ceramic structures under Hertzian indentation. Dent Mater 2003;19:542–51.
1295
[13] Holroyd TJ. In: Hunt TV, editor. The acoustic emission and ultrasonic monitoring. 1st ed. Oxford: Coxmoor Publishing Company; 2000. [14] Nicholls PJ, Berg E. Acoustic emission properties of callus. Med Biol Eng Comput 1981;19(4):416–8. [15] Kim KH, Park JH, Imai Y, Kishi T. Microfracture mechanisms of dental resin composites containing spherically-shaped filler particles. J Dent Res 1994;73(2):499–504. [16] Kohn DH. Acoustic emission and nondestructive evaluation of biomaterials and tissues. Crit Rev Biomed Eng 1995;23(3–4):221–306. [17] Williams RV. Acoustic emission techniques and systems. In: Williams RV, editor. Acoustic emission. Bristol: Adam Hilger Ltd.; 1980. [18] Alander P, Lassila LV, Tezvergil A, Vallittu PK. Acoustic emission analysis of fiber-reinforced composite in flexural testing. Dent Mater 2004;20(4):305–12. [19] Watanabe Y, Takai S, Arai Y, Yoshino N, Hirasawa Y. Prediction of mechanical properties of healing fractures using acoustic emission. J Orthopaed Res 2001;19:548–53. [20] Srikanth V, Subbarao EC. Acoustic emission study of phase relations in low-Y2 O3 portion of ZrO2 –Y2 O3 system. J Mater Sci 1994;29:3363–71. [21] Duray SJ, Menis DL, Gilbert JL, Greener EH. Laser-induced acoustic emission in particle filled dental composites. J Dent Res 1995;74:184. [22] Fennis WM, Tezvergil A, Kuijs RH, Lassila LV, Kreulen CM, Creugers NH, Vallittu PK. In vitro fracture resistance of fiber reinforced cusp-replacing composite restorations. Dent Mater 2005;21(6):565–72. [23] Kim KH, Okuno O. Microfracture behaviour of composite resins containing irregular-shaped fillers. J Oral Rehabil 2002;29(12):1153–9. [24] Vallittu PK. Use of woven glass fibres to reinforce a composite veneer. A fracture resistance and acoustic emission study. J Oral Rehabil 2002;29(5):423–9. [25] Asaoka K. Effect of transient stress on acoustic emission behaviour during firing of dental porcelain. J Mater Sci 1992;27:3118–22. [26] Chen HY, Hickel R, Setcos JC, Kunzelmann KH. Effects of surface finish and fatigue testing on the fracture strength of CAD-CAM and pressed-ceramic crowns. J Prosthet Dent 1999;82:468–75. [27] Attia A, Abdelaziz KM, Freitag S, Kern M. Fracture load of composite resin and feldspathic all-ceramic CAD/CAM crowns. J Prosthet Dent 2006;95:117–23. [28] Aboushelib MN, de Jager N, Kleverlaan CJ, Feilzer AJ. Effect of loading method on the fracture mechanics of two layered all-ceramic restorative systems. Dent Mater 2007;23(8): 952–9. [29] Floystrand F, Kleven E, Oilo G. A novel miniature bite force recorder and its clinical application. Acta Odontol Scand 1982;40(4):209–14.
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Initial versus final fracture of metal-free crowns, analyzed via acoustic emission Nadia Ereifej a , Nick Silikas a,b,∗ , David C. Watts a,b a b
Biomaterials Science Research Group, School of Dentistry, The University of Manchester, UK Photon Science Institute, The University of Manchester, UK
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. To discriminate between initial and final fracture failure loads of four metal-free
Received 22 January 2008
crown systems by the conjoint detection of acoustic emission signals during compressive
Received in revised form
loading.
6 March 2008
Methods. Teeth were prepared and used for crown construction with four crown systems;
Accepted 22 April 2008
Vita Mark II (VM II) (Vita Zahnfabrik), IPS e.max Ceram/CAD (CAD) (Ivoclar-Vivadent), IPS e.max Ceram/ZirCAD (ZirCAD) (Ivoclar-Vivadent) and BelleGlass/EverStick (BGES) (Kerr/Stick Tech Ltd.). All samples were loaded in compression via a Co/Cr maxillary first molar tooth at
Keywords:
0.2 mm/min and released acoustic signals were collected and analyzed. A minimum number
Acoustic emission
of 15 crowns per group were loaded to final failure and values of loading at initial and
Ceramics
final fracture were compared. Additional four samples per group were loaded till fracture
Fiber-reinforced composites
initiation and were fractographically examined under the optical microscope.
Metal-free crowns
Results. A lower threshold of 50 dB was selected to exclude spurious background signals.
Mechanical testing
Initial fracture forces were significantly lower than those of final fracture (p < 0.05) in all
Microcracks
groups and initial failure AE amplitudes were lower than those of final fracture. Mean initial
Fracture strength
fracture force of ZirCAD samples (1029.1 N) was higher than those of VMII (744.4 N), CAD (808.8 N) and BGES (979.7 N). Final fracture of ZirCAD also occurred at significantly higher force values (2091.7 N) than the rest of the groups; VMII (1120.9 N), CAD (1468.9 N) and BGES (1576.6 N). Significantly higher values of initial failure AE amplitude were found in VMII than CAD and BGES while those of final fracture were similar. All crowns observed under the microscope at initial fracture had signs of failure. Significance. Whereas the metal-free crowns examined showed significant variations in final failure loads, acoustic emission data showed that they all manifested initial failures at significantly lower load values. © 2008 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Increasing demands for restorative dental materials that combine pleasing aesthetics and optimal physical and chemical properties have led to the introduction of various types of
dental ceramics [1]. Fiber-reinforced composites are alternative materials to ceramics and have been successfully used in dental restorations [2]. Intra-oral restorations are challenged by complex loading conditions including both functional and parafunctional
∗ Corresponding author at: Biomaterials Science Research Group, School of Dentistry, Higher Cambridge Street, M15 6FH Manchester, UK. Tel.: +44 16 12756474. E-mail address: [email protected] (N. Silikas). 0109-5641/$ – see front matter © 2008 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2008.04.010
1290
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
stresses and fracture is still one of the most common causes of failure of these restorations [3–6]. Fracture occurs when a particular stress limit is exceeded after which brittle materials, like ceramics, fail abruptly while more ductile materials, like composites, undergo a phase of elastic deformation before they fail [7]. Therefore, the load-bearing capacity of a particular material is an important criterion influencing the performance and serviceability of this material in the clinical situations [8]. Various testing methods have been used to investigate the mechanical properties of dental materials. However, strength is not an inherent property of a certain material and it is dependant on the condition of the tested material in addition to other factors like stressing rate, stress concentration, specimen size and failure mode. Consequently, testing conditions should be controlled very carefully [9,10]. Many concerns have been raised lately regarding the clinical relevance of simple traditional mechanical testing methodologies. These tests used unrealistically high loading values and the modes of failure were different than those reported in real clinical situations. Furthermore, the results of these tests lack significant information considering the site, time of fracture initiation and fracture mechanics [11,12]. The concept of Acoustic Emission (AE) was first proposed by Joseph Kaiser in the early 1950s and later gained wider acceptance in the 1960s [13]. AE is a sound wave produced as a result of energy release from a material at which external stimuli are applied. Upon loading, areas with flaws develop high stress energies and some of this energy is released in the form of a pressure wave or sound which occurs long before the specimen fails catastrophically. As these waves propagate through the surface of the object being tested, they are collected by piezoelectric transducers coupled on the surface that convert part of these waves into electrical charges and then pass them through filters in order to eliminate background noise, then through amplifiers and finally to a computer processing system for data analysis [13–16]. The AE technique offers the advantages of being a non-stop technique which can monitor the condition of the material under investigation throughout the test. It is non-localized and has the ability to examine large volume objects at the same time [17]. Combined with conventional fracture testing machines, AE can identify failure initiation, the initial site of damage, damage propagation, and catastrophic failure of the material and help elucidate the complex failure mechanism. It is considered as an early warning of failure of the material, regarded to have high sensitivity as it registers even the small cracks as long as a noise is generated and thus is claimed to provide real time data [16,18,19]. AE is widely used in material testing and has been applied in the analysis of fracture behavior of different types of biomaterials like ceramics [7,20], dental composites [18,21–24] and porcelain [25]. However, AE technique was rarely employed in fracture testing of dental ceramics. The aim of this study was to investigate the fracture behavior of different all-ceramic and fiber-reinforced composite materials used in prosthodontics applications. The specific objectives were
(1) To use AE technique during compressive strength testing. (2) To select AE parameters for fracture analysis and set a threshold for AE signals reflecting crack initiation. (3) To compare the values of loading at initial and final failure of the materials tested using AE analysis. The null hypotheses were (i) AE cannot detect initial fracture of all-ceramic and fiberreinforced composite materials. (ii) There are no significant differences in initial and final fracture strength values or amplitudes of released AE signals among the different materials.
2.
Materials and methods
2.1.
Sample preparation
Human mandibular third and second molar teeth were collected from several dental practices, disinfected with formalin, cleaned of debris and then kept in water to prevent dehydration. Upon examination of the collected teeth, only sound teeth which had dimensions falling within two standard deviations of the means (of M-D and B-L dimensions) were included. A mandibular second molar plastic tooth with average dimensions, used as a master tooth, was prepared using a high speed air turbine with coarse grit round end diamond bur. Preparation was performed so that all undercuts were eliminated, occlusal surface was reduced by 2 mm with a functional cusp bevel, all sharp angles were rounded and a 1 mm deep chamfer finish line was prepared and located 1 mm coronal to cemento-enamel junction. A Celay machine (Celay, Mikronna Technologie AG, Spreitenbach, Switzerland) was used to prepare the specimen teeth in this study. The master tooth was attached to the machine at one side and the specimen tooth was attached on the other side. While the master tooth was being scanned, the specimen tooth was being milled, so that at the end of the procedure, the milled tooth had an identical preparation to the master tooth. Finally, the dimensions of the prepared teeth were verified using a caliper to ensure that all the specimens had similar standardized dimensions (M-D = 9.4 ± 0.14, B-L = 8.4 ± 0.14). Prepared teeth were numbered and an impression was taken for each tooth using an addition-cured polyvinylsiloxane impression material (Provil, Heraeus Kulzer, Germany) via a one stage putty/wash technique. The impressions were cast using vacuum-mixed dental stone. Dies were then randomly divided into 4 groups and used for crown construction. All crown fabrication procedures were performed by one trained operator. Table 1 shows the materials used, the manufacturers and the number of samples in each group. For the construction of Vita Mark II (VMII) crowns optical images were taken for the dies of the prepared teeth and were then used to design full-veneer crowns of a mandibular second molar with CEREC 3 CAD software, Version 280 r2400 (Sirona Dental System GmbH, Fabrikstrasse, Germany). The restora-
1291
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
Table 1 – The different materials used in the study Group 1 2 3 4
Number of samples
Material
24 22 25 27
Vita Mark II (VMII) IPS e.max Ceram/CAD (CAD) IPS e.max Ceram/ZirCAD (ZirCAD) BelleGlass/EverStick (BGES)
tions were then milled from Vita Mark II blocks at the milling unit of the machine. IPS e.max Ceram/IPS e.max CAD (CAD) samples were prepared as the following: cores with minimum thicknesses of 1 mm occlusally and 0.8 mm circumferentially were designed. These were milled from IPS e.max CAD blocks and fired to achieve complete crystallization in a Programat P10 furnace (Ivoclar-Vivadent, Schaan, Lichtenstein). The cores were then veneered with IPS e.max Ceram to a thickness of 2.0–2.5 mm. For the construction of IPS e.max Ceram/IPS e.max ZirCAD (ZirCAD) samples, cores (minimum 0.7 mm occlusally and 0.5 mm circumferentially thick) were designed and milled from IPS e.max ZirCAD blocks using CEREC 3 InLab machine. These were then placed in a special furnace, Sintramat (Ivoclar-Vivadent, Schaan, Lichtenstein) for sintering. A layer of ZirLiner was added and fired to enhance bonding with the veneering porcelain and veneering was consequently performed using IPS e.max Ceram. BelleGlass/EverStick (BGES) crowns were constructed as follows: two layers of EverStick Net were cut to the appropriate size, adapted onto the dies and light cured for 40 s. BelleGlass HP composite (BG) was placed on top of the fiber cores and cured for 40 s. The partially cured crowns were then placed in a special BG oven and cured in a nitrogen chamber under 145 ◦ C and 60 psi pressure. All crowns were then adhesively cemented to their corresponding teeth using Multilink resin-based dual-cured cement (Ivoclar-Vivadent, Schaan, Lichtenstein). Before cementation, the teeth and the inner surfaces of the crowns were prepared as recommended by the manufacturers. The samples were then stored in water at 37 ◦ C for 24 h before they were tested.
2.2.
Fracture strength (quantitative)
Load was applied on the specimens using a Zwick/Roell Z020 Universal Testing Machine (Zwick GmbH & Co. KG, Ulm, Germany) shown in Fig. 1. Specimens were placed at the lower part of the machine while the upper one used a cast Co–Cr upper molar tooth moving at 0.2 mm/min speed for load application. A 2-channel AE system (Physical Acoustics Corporation, NJ, USA) was used to detect audible signals produced while the specimens were being loaded. One of the sensors was fixed at the metal base of the lower component of the Zwick machine just next to the sample and the other was attached to the loading component of the Zwick machine. Signals detected by the sensors were passed through preamplifiers of 40 dB gain with band pass of 100 kHz to 2 MHz and a threshold set at 50 dB. The data obtained from the AE machine and those obtained from the Zwick machine were combined and used to analyze the values of load at which initial cracking and final failure occurred. A graph of load and AE amplitude versus time was
Manufacturers Vita Zahnfabrik, Bad Sackingen-Germany Ivoclar-Vivadent, Schaan, Lichtenstein Ivoclar-Vivadent, Schaan, Lichtenstein Kerr, CA, USA/Stick Tech Ltd., Oy, Turku, Finland
obtained for each sample using Sigmaplot software (V 8.0) as shown in Fig. 2. The exact time at which the first AE signal ≥50 dB was released was obtained and the value of loading force (N) at this point was extracted and regarded as the value of initial failure. Further loading resulted in crack propagation, thus further release of AE signals continued until final failure occurred at which loading stopped automatically and AE signals with high amplitudes were released. The value of loading at this point was considered as the final fracture force. The same procedure was performed for all the samples in the four groups under investigation.
2.3.
Optical microscope (qualitative)
For four samples per group, the test was stopped when an AE signal of an amplitude ≥50 dB was recorded. The crowns were then tested visually and microscopically using Meiji EMZ-TR optical microscope (Meiji Techno Co., Ltd., Tokyo, Japan) at a maximum magnification of 40× to detect signs of failure. Those which did not have obvious initial cracks on the occlusal surfaces were sectioned to identify subsurface signs of failure. The rest of the samples of all the groups were tested till final failure.
2.4.
Statistical analysis
Wilcoxon-Signed ranks and paired t-tests using SPSS (Version 13.0) were conducted to detect significant differences between the initial fracture force and final fracture force values in each group. A series of Mann–Whitney tests and t-independent tests was conducted to find statistically significant differences in the initial and final fracture values between the different groups. Furthermore, a series of paired t-tests was used to detect significant differences between the average initial fracture AE signal amplitude and that of final fracture in each group followed by one-way ANOVA test to detect differences in AE amplitudes between the different materials.
3.
Results
3.1.
Quantitative findings
Table 2 shows the means and standard deviations of initial and final fracture force values (N), the percentage of initial fracture load to the final fracture load and AE amplitudes (dB) at initial as well as those at final fracture for all materials. These values are also represented in Figs. 3 and 4. Initial fracture values were always significantly lower than final fracture values with the highest difference among ZirCAD samples (1052.6 N) and the lowest in VMII (376.6 N).
1292
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
Fig. 1 – One of the crown samples being tested. (a): The basic set up of the test. (b): A drawing showing the maximum intercuspation between the loading tooth and the specimen.
Table 2 – Mean and standard deviation values for force and AE amplitude of initial and final fracture of the materials tested Material VMII CAD ZirCAD BGES
Initial fracture force (N)
Final fracture force (N)
% of initial to final force
744.4 (151.8) 808.8 (259.2) 1029.1 (333.5) 979.7 (243.9)
1121.0 (248.8) 1469.0 (290.6) 2081.7 (552.8) 1576.6 (327.4)
66.4% 55.0% 49.4% 62.1%
Correlation and regression analyses were performed to detect the relationship between initial fracture strength and final fracture strength. These are illustrated in Fig. 5. Although the correlation was found insignificant (p > 0.05), values of final fracture force tended to increase as the values of initial fracture force increased (r2 = 0.82). VMII had the lowest initial fracture value (744.4 N), followed by CAD (808.8 N). Both were significantly lower than BGES (979.7 N) and ZirCAD (1029.1 N). Mean final fracture force of VMII (1120.9 N) was significantly lower than that of CAD (1468.9 N), ZirCAD (2091.7 N) and BGES (1576.6 N) while that of ZirCAD was significantly higher than the rest of the groups. The initial fracture AE amplitude was significantly lower than the final fracture AE amplitude in all the groups. Comparing these values between the different groups, significantly higher values of initial fracture AE amplitude were found for VMII than those of CAD and BGES while the values of AE amplitude of the final fracture were similar.
3.2.
Qualitative findings
Visual and microscopic examination of the crowns that were loaded till initial failure showed the presence of small cracks within most of the samples tested as shown in Fig. 6. Five crowns, especially within VMII group, had clear cracks sometimes even visible with the unaided eye while in the rest of samples, especially in CAD, ZirCAD and BGES groups, smaller cracks were identified and in two samples, cracks were only located using the optical microscope at high magnification (up to 40×). Four samples of these groups did not have clear cracks on their occlusal surfaces but some changes in the opacity of
Initial fracture amplitude (dB) 66.8 (12.7) 56.4 (5.9) 60.5 (9.5) 57.6 (9.5)
Final fracture amplitude (dB) 90.0 (8.2) 87.0 (8.5) 94.8 (10.2) 88.6 (9.4)
one or more of the cusps were noticed indicating possible subsurface cracking or failure. Sectioning of these crowns at areas of suspected hidden failure revealed the presence of cracks that were initiated at cement/core or core/veneer interfaces.
4.
Discussion
In this study, the AE technique was employed in combination with compressive strength testing of four different types of materials used in the construction of crowns and fixed partial dentures. The technique was used to detect the time at which failure was initiated and therefore determine the loading force that caused microcrack initiation. Monitoring acoustic signals released upon further loading revealed information about further cracking and failure propagation. Finally, AE results combined with those of the loading machine were used to determine the values of loading at which final fracture occurred. Failure process starts as soon as the first microcrack is initiated which upon further loading and fatigue intra-oral conditions propagates and ultimately leads to final fracture. The presence of these microcracks weakens the structure of the material and results in its failure even at lower stress levels than those at which they were initiated. Therefore, recording the loading values at which initial microcracks occur might be the more accurate way of measuring the strength of a particular material bringing the results of the test closer to the real clinical situation. In this study, the ability of AE technique to detect initial fracture and final failure was demonstrated and therefore the first null hypothesis was rejected. Furthermore, statistically significant differences were found between the different materials tested regarding values of initial and
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
1293
Fig. 3 – Values of force (N) for initial and final fracture for all materials.
Fig. 4 – AE amplitude of initial and final fracture for all materials.
tended to fracture into many fragments in a pattern that does not exist clinically [11]. In this study, a cast Co/Cr model of an upper second molar was used to load the samples to failure. Before loading, maximum interdigitation between the loading tooth and the sample tooth was achieved mimicking their intercuspation in a real clinical situation. Upon analysis of the fractured crowns, the patterns of failure found were not very different from those observed clinically. Most of the crowns had a fragment or two chipped off, delamination of the veneer
Fig. 2 – Graphs of force (N) and AE amplitude (dB) versus time (s) of a representative specimen for each Group. (a): CAD, (b): VMII, (c): ZirCAD, (d): FRC. (): force at initial fracture, (䊉): force at final fracture.
final fracture force and the amplitudes of the released acoustic signals and thus the second null hypothesis was rejected. Most studies investigating the compressive strength of materials used metal spheres of different dimensions to apply load on tested samples [26,27]. These might not be able to distribute forces evenly on the samples and therefore samples
Fig. 5 – A plot of final fracture force versus initial fracture force.
1294
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
Fig. 6 – An image of a ceramic crown (VMII) with visible initial cracking (a), and a graph showing AE signal release at initial fracture of the crown (b).
porcelain was common in bilayered crowns and only very few samples broke into many pieces or had the underlying teeth involved in fracture. Signs of initial failure were found in all samples examined after initial fracture. Failure of VMII crowns seemed to start at their occlusal surfaces as cracks were evident there while in bilayered samples, (CAD, ZirCAD and BGES), cracks were smaller, more difficult to locate or not even visible on the outer surfaces of the crowns. Occlusal cone cracks created on the outer surfaces of the veneer layers of the crowns around the points of load concentration were found to be the most common modes of failure of all-ceramic crowns. However, cracks initiating at core/veneer interfaces or at core/cement interface were also reported in bilayered ceramic samples [28]. These more complex modes of failure in bilayered structures might explain the different microscopic appearance of bilayered crowns tested at initial failure. However, all these microscopic findings have supported the validity of the preset threshold with further evidence and proven the ability of AE technology to detect initial fracture even if it occurs in subsurface layers. Alander et al. [18] investigated the acoustic signals emitted during flexural strength testing of different types of fiberreinforced composites and found that AE signals were first released at values of loading that are 19–32% lower than the final failure points and this difference was statistically significant. Analysis of the results of the crowns tested till final fracture in this study showed that initial fracture, detected by the AE technique, always started at significantly lower loading values (33.6–51.6%) than the final loading values. This indicated that failure actually started much earlier than the
final fracture point and that loading machines recording only forces at catastrophic failure have overestimated the fracture strength of tested materials. However, a general tendency for the values of final fracture force to increase as the values of initial fracture force increased was observed and therefore initial failure point can be used to predict the final failure one. Assuming that the maximum biting force expected in normal situations in the back of the mouth is about 500 N, the four types of materials (with only a few samples per group that failed at low stress levels which can be related to manufacturing-induced defects) had average values of initial fracture that are higher than that value [29]. Therefore, it can be concluded that all materials could be potentially used in the construction of posterior full veneer crowns. Nevertheless, proper case selection and following the manufacturers’ recommendations regarding the construction processes and the indications of these materials are important factors for the success of such restorations. AE release patterns after crack initiation and before final fracture varied among the samples of the different materials. In VMII, most of the samples had limited AE activity during crack propagation indicating fewer events occurring within the samples at fracture. This can be explained by the structure and manufacturing process of these samples. These were milled using CAD/CAM machines and are composed of a homogenous structure of fine-particle feldspathic porcelain. On the other hand, in bilayered samples, higher AE activity was recorded during the fracture process as cracks propagated through the different layers of the crowns or at the interfaces between the different components. The fact that the veneer layers of these crowns were manually constructed adds to the complexity of their structure and might have induced more defects within their structure than machined samples. Construction-induced flaws might have increased AE activity of these multilayered samples. Furthermore, much fewer signals were recorded during fracture process of BGES group than all-ceramic groups. This can be related to the more flexible nature of resin-composite materials compared to brittle ceramics. Composites undergo a phase of plastic deformation before they break catastrophically. This plastic deformation is a relatively quiet process and few AE signals, above the preset threshold, are expected to be recorded. On the other hand, allceramic materials are more brittle and do not undergo plastic deformation. When loading exceeds a particular elastic limit, fracture occurs. These microfractures are noisy and therefore higher AE activity is usually recorded. At final fracture of the all-ceramic samples, parts of the crowns were chipped off. This fracture was accompanied by the release of AE signals of very high amplitude. In BGES group, the fractured fragments of the crowns were kept in place by the underlying reinforcing fibers and the more flexible nature of BG composite resulted in a gradual rather than a sudden drop of the loading force as observed in the other groups. This can be attributed to the presence of fibers that seemed to break gradually. High fracture resistance of ZirCAD group can be explained by the toughening mechanism of Y2 O3 -stabilized zirconia-based core of this group. AE amplitude at initial failure was found to be significantly lower than that at final fracture within all the groups. Signals released at initial failure were caused by small cracks while
d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1289–1295
those at final failure were caused by catastrophic fracture that involved in many cases total chipping of one or two cusps of the crowns. Therefore, it can be concluded that AE amplitude might be an indication of the severity of the fracture.
Acknowledgment The authors are grateful to Ivoclar-Vivadent (Schaan, Lichtenstein) for providing IPS e.max materials and equipment and Mr. Adrian Bennett for providing the materials and equipment for the construction of VMII samples.
references
[1] McLaren EA, White SN. Glass-infiltrated zirconia/alumina-based ceramic for crowns and fixed partial dentures: clinical and laboratory guidelines. Quintessence Dent Technol 2000;23:63–75. [2] Brown D. Fibre-reinforced materials. Dent Update 2000;27:442–8. [3] Walton JN. A survey of crown and fixed partial denture failures: length of service and reasons for replacement. J Prosthet Dent 1986;56:416–21. [4] Brochu JF, El-Mowafy O. Longivity and clinical performance of IPS Empress Ceramic restorations—a literature review. J Can Dent Assoc 2002;68(4):233–7. [5] Oden A, Andersson M, Krystek-Ondracek I, Magnusson D. Five-year clinical evaluation of Procera AllCeram crowns. J Prosthet Dent 1998;80(4):450–6. [6] Taskonak B, Sertgoz A. Two-year clinical evaluation of lithia-disilicate-based all-ceramic crowns and fixed partial dentures. Dent Mater 2006;22(11):1008–13. [7] Lawn BR. Strength and reliability. In: Davis EA, Ward IM, editors. Fracture of brittle solids. 2nd ed. Cambridge: Cambridge University Press; 1993. [8] Darvell BW. Mechanical testing. In: Darvell BW, editor. Materials Science for Dentistry. 7th ed. Hong Kong: Darvell BW; 2002. [9] Kelly JR. Perspectives on strength. Dent Mater 1995;11(2):103–10. [10] Kelly JR. Dental ceramics: current thinking and trends. Dent Clin North Am 2004;48(2):513–30. [11] Kelly JR. Clinically relevant approach to failure testing of all-ceramic restorations. J Prosthet Dent 1999;81(6): 652–61. [12] Dong XD, Darvell BW. Stress distribution and failure mode of dental ceramic structures under Hertzian indentation. Dent Mater 2003;19:542–51.
1295
[13] Holroyd TJ. In: Hunt TV, editor. The acoustic emission and ultrasonic monitoring. 1st ed. Oxford: Coxmoor Publishing Company; 2000. [14] Nicholls PJ, Berg E. Acoustic emission properties of callus. Med Biol Eng Comput 1981;19(4):416–8. [15] Kim KH, Park JH, Imai Y, Kishi T. Microfracture mechanisms of dental resin composites containing spherically-shaped filler particles. J Dent Res 1994;73(2):499–504. [16] Kohn DH. Acoustic emission and nondestructive evaluation of biomaterials and tissues. Crit Rev Biomed Eng 1995;23(3–4):221–306. [17] Williams RV. Acoustic emission techniques and systems. In: Williams RV, editor. Acoustic emission. Bristol: Adam Hilger Ltd.; 1980. [18] Alander P, Lassila LV, Tezvergil A, Vallittu PK. Acoustic emission analysis of fiber-reinforced composite in flexural testing. Dent Mater 2004;20(4):305–12. [19] Watanabe Y, Takai S, Arai Y, Yoshino N, Hirasawa Y. Prediction of mechanical properties of healing fractures using acoustic emission. J Orthopaed Res 2001;19:548–53. [20] Srikanth V, Subbarao EC. Acoustic emission study of phase relations in low-Y2 O3 portion of ZrO2 –Y2 O3 system. J Mater Sci 1994;29:3363–71. [21] Duray SJ, Menis DL, Gilbert JL, Greener EH. Laser-induced acoustic emission in particle filled dental composites. J Dent Res 1995;74:184. [22] Fennis WM, Tezvergil A, Kuijs RH, Lassila LV, Kreulen CM, Creugers NH, Vallittu PK. In vitro fracture resistance of fiber reinforced cusp-replacing composite restorations. Dent Mater 2005;21(6):565–72. [23] Kim KH, Okuno O. Microfracture behaviour of composite resins containing irregular-shaped fillers. J Oral Rehabil 2002;29(12):1153–9. [24] Vallittu PK. Use of woven glass fibres to reinforce a composite veneer. A fracture resistance and acoustic emission study. J Oral Rehabil 2002;29(5):423–9. [25] Asaoka K. Effect of transient stress on acoustic emission behaviour during firing of dental porcelain. J Mater Sci 1992;27:3118–22. [26] Chen HY, Hickel R, Setcos JC, Kunzelmann KH. Effects of surface finish and fatigue testing on the fracture strength of CAD-CAM and pressed-ceramic crowns. J Prosthet Dent 1999;82:468–75. [27] Attia A, Abdelaziz KM, Freitag S, Kern M. Fracture load of composite resin and feldspathic all-ceramic CAD/CAM crowns. J Prosthet Dent 2006;95:117–23. [28] Aboushelib MN, de Jager N, Kleverlaan CJ, Feilzer AJ. Effect of loading method on the fracture mechanics of two layered all-ceramic restorative systems. Dent Mater 2007;23(8): 952–9. [29] Floystrand F, Kleven E, Oilo G. A novel miniature bite force recorder and its clinical application. Acta Odontol Scand 1982;40(4):209–14.