Abstract
Background. Ceramic fracture is a common problem in metal-ceramic restorations (MCRs). The advent of computer-aided design and computer-aided manufacturing (CAD-CAM) technology eliminated the lost-wax technique, which was responsible for many of the problems associated with framework fabrication. However, the role of the CAD-CAM technology in decreasing porcelain fracture is not yet known.
Objectives. The aim of the present in vitro study was to compare the fracture strength of porcelain in MCRs with metal frameworks fabricated with the use of the lost-wax and CAD-CAM techniques.
Material and methods. Twenty metal dies were prepared with a deep chamfer finish line, with a depth of 1.2 mm and the occlusal taper of the walls of 8°, a 2-millimeter occlusal reduction of the functional cusp, a 1.5-millimeter occlusal reduction of the nonfunctional cusp, and the functional cusp bevel. Ten frameworks were fabricated using the CAD-CAM system and 10 with the lost-wax technique. After porcelain veneering, the specimens underwent thermocycling and cyclic loading to simulate the aging process. The load test was then performed. The fracture strength of porcelain was compared between the 2 groups, and the mode of failure was also determined using a stereomicroscope.
Results. Two specimens were excluded from the CAD-CAM group. Thus, 18 specimens were statistically analyzed. The results revealed no significant difference in fracture strength between the 2 groups (p > 0.05). The mode of failure was mixed in all specimens from both groups.
Conclusions. Our results indicated that the fracture strength of porcelain and the mode of failure did not depend on the metal framework fabrication technique (lost-wax or CAD-CAM).
Keywords: CAD-CAM, fracture strength, Co-Cr, lost-wax, metal-ceramic
Introduction
Metal-ceramic restorations (MCRs) are still widely used due to their optimal physical properties and a lower cost as compared to all-ceramic restorations.1, 2, 3 Also, MCRs reportedly have a higher 5-year survival rate than all-ceramic restorations.4 Evidence shows that dental caries, followed by porcelain fracture and chipping are the main causes of failure of MCRs.5, 6 Several strategies have been suggested to prevent porcelain fracture or chipping in MCRs. Modifying the framework design is one of such strategies.7
Cobalt-chromium (Co-Cr) alloys are among the most commonly used alloys for the fabrication of MCRs, with successful clinical application since 1930.8 Cobalt provides hardness, while Cr enhances the physical properties of the alloy and prevents its corrosion. Molybdenum (Mo), which is also present in this alloy, optimizes other particles, creates space during the solidification process and increases the strength of the alloy. It is also responsible for resistance to corrosion. Tungsten (W) has effects similar to those of Mo on the properties of the Co-Cr alloy and is sometimes used as an alternative to Mo.8 Such alloys are superior to others, e.g., nickel-chromium (Ni-Cr) ones, and have been recommended for the fabrication of dental prosthetic restorations.9 Frameworks for MCRs can be fabricated with the lost-wax technique, the computer-aided design and computer-aided manufacturing (CAD-CAM) technology, or the laser sintering technique. Among these, the lost-wax technique is most commonly used.10 This method was first introduced by Dr. B.F. Philbrook in 1897 and gained popularity in 1906.11 However, problems that may occur during the investment casting and cooling processes in the lost-wax technique are often responsible for the failure of MCRs, since they can compromise the strength of the bond between porcelain and the metal framework.12, 13 Appropriate metal–porcelain bond strength is imperative for porcelain strength.14 The advent of the CAD-CAM technology eliminated many problems associated with the use of the lost-wax fabrication technique.7, 15, 16, 17, 18 The CAD-CAM technology enables the fabrication of restorations with high precision, irrespective of the length of the restoration.19
There are different methods of assessing fracture strength, one of which is the load test. During this test, a load is applied vertically to the sample until fracture occurs, and the mode of fracture is subsequently assessed. It can be adhesive, cohesive or mixed. Fracture is considered adhesive when it occurs at the metal–porcelain interface, and it is considered cohesive when it occurs within metal or porcelain. If both adhesive and cohesive types of fracture are detected in the sample, the mode of failure is considered mixed.18, 20, 21, 22, 23
Previous studies evaluated the effect of different methods of fabrication of Co-Cr frameworks on bond strength, yielding controversial results.20, 21, 22, 23, 24 Considering the gap of information, the purpose of the present study was to compare the fracture strength of porcelain in MCRs with Co-Cr frameworks fabricated with the use of the lost-wax technique and the CAD-CAM technology.
Material and methods
In this in vitro experimental study, 20 MCRs were fabricated. Ten metal frameworks were fabricated with the lost-wax technique, while another 8 were fabricated using the CAD-CAM technology (2 specimens were excluded).7, 20, 25, 26 For this purpose, maxillary second premolar acrylic teeth received deep chamfer preparation at the cervical region all-around, with a depth of 1.2 mm, the occlusal taper of the walls of 8°, a 2-millimeter occlusal reduction of the functional cusp, a 1.5-millimeter occlusal reduction of the nonfunctional cusp, and the functional cusp bevel.24 Acrylic teeth were then scanned with a Ceramill Map 400 scanner (Amann Girrbach, Pforzheim, Germany) and the information was sent to a Ceramill Motion 2 milling machine (Amann Girrbach). Based on this information, wax blocks (Yamahachi Dental, Gamagori, Japan) were carved and sprued. The metal die was invested with phosphate-bonded investment (Z4 Universal Investment; N&V Belgium, Sint-Niklaas, Belgium) and cast using a Ducatron Quattro casting machine (Ugin Dentaire, Seyssinet-Pariset, France) and the Co-Cr alloy (Magnum Ceramic Co; Mesa Italia, Travagliato, Italy).15, 27, 28 The die was then prepared with a tapered bur on a surveyor (Figure 1). The prepared metal die was scanned and the scan data was transferred to the CAD software in a Ceramic Mind CAD workstation (Amann Girrbach). After assessing the finish line and ensuring that there were no undercuts, the framework was designed with an equal thickness of 0.5 mm, and a die spacer of 50 µ at a 1-millimeter distance from the finish line (Figure 2).7, 23
Fabrication of specimens
with the lost-wax technique
A CAD-CAM machine was used to increase accuracy, and also for the standardization of specimens. The data regarding the framework design was transferred from the Ceramill Mind CAD workstation to the communicating milling machine (Ceramill Motion 2) and 10 wax patterns were carved out of the wax disks (Figure 3, Figure ).7, 24 Next, the specimens were sprued and invested using phosphate-bonded investment (Z4 Universal Investment).
Fabrication of specimens
with the CAD-CAM technique
The data regarding the framework design was transferred from the Ceramill Mind CAD workstation to the communicating milling machine (Ceramill Motion 2). Next, 8 specimens were dry-milled by using Co-Cr blocks (Ceramill Sintron blanks; Amann Girrbach) (Figure 3, Figure ).7, 20, 24 The specimens were then sintered in a Ceramill Argovent 2 sintering compartment (the Ceramill Argotherm 2 system; Amman Girrbach), in an atmosphere of argon gas at 1,280°C for 6 h. The specimens were all homogenous and had no distortions. For surface treatment prior to veneering, according to the manufacturer’s instructions, the specimens in the lost-wax group were sandblasted with 150-micrometer aluminum oxide particles, while the specimens in the CAD-CAM group were sandblasted with 50-micrometer aluminum oxide particles for 20 s at an angle of 45° and a distance of 10 mm, under pressure of 3 bars (Basic eco microblaster; Renfert, Hilzingen, Germany). All specimens were subsequently cleaned with 80% ethanol in an ultrasonic bath for 5 min, and then placed in a furnace for oxidation and degaussing (Programat P310; Ivoclar Vivadent, Schaan, Liechtenstein).7, 22, 23 The thickness of all specimens was measured with a digital caliper with an accuracy to 0.01 mm. Table 1 shows the information regarding the alloys used in the lost-wax and CAD-CAM techniques.
Porcelain veneering
To standardize the shape and amount of porcelain in all specimens, a full-contour crown was designed using the Ceramill Mind software and carved out of a wax block. A putty index (Speedex; Coltène/Whaledent, Altstätten, Switzerland) was made based on the wax model. Two layers of paste (InLine Opaquer A3; Ivoclar Vivadent) were applied to all specimens, and then the porcelain body (InLine Dentin A3; Ivoclar Vivadent) was applied using the silicon index (Figure 4). Lastly, the specimens were glazed (Figure 5). One technician performed all the baking procedures according to the manufacturer’s instructions (Ivoclar Vivadent).
Simulation of oral conditions
All specimens underwent thermocycling and cyclic loading to simulate normal oral conditions. The specimens were placed in a thermocycler (SD Mechatronik, Feldkirchen-Westerham, Germany) and subjected to 3,000 thermal cycles at 5–55°C. Each cycle lasted 60 s, and included 20 s of dwell time and 20 s of transfer time.29 For cyclic loading, a chewing simulator (CS-4; SD Mechatronik) was used (Figure 6). Using deionized water, 100,000 cycles were applied with a load of 100 N and a frequency of 1 Hz, corresponding to 2–3 months of clinical service.30, 31
Load test
The acrylic die stand was first replaced with the Co-Cr alloy to resist forces. The specimen was then cemented on the metal die with the Temp-Bond™ cement (Kerr, Brea, USA). Next, a load was applied to the specimen by means of a round-end stainless steel bar with a diameter of 5 mm at a crosshead speed of 1 mm/min along the tooth longitudinal axis in a universal testing machine (Santam Co., Tehran, Iran) with a capacity of 200 kgf. The load was applied vertically until the fracture of the specimen occurred (Figure 6).28 The load causing fracture was recorded for each specimen and the mode of failure was determined under a stereomicroscope (SMZ800; Nikon, Tokyo, Japan).
Statistical analysis
The normality of the data was assessed using the Shapiro–Wilk test, and the homogeneity of the data was evaluated using Levene’s test. The data was analyzed using IBM SPSS Statistics for Windows, v. 24.0 (IBM Corp., Armonk, USA), and the t tests were applied at the significance level of 0.05.
Results
Two specimens were excluded from the CAD-CAM group due to fracture during the study, and 10 specimens in the lost-wax group and 8 specimens in the CAD-CAM group remained in the study. As shown in Table 2, the mean fracture strength was 2,271 ±420 N in the lost-wax group and 2,379 ±531 N in the CAD-CAM group. This difference was not significant (p > 0.05). The mode of failure was mixed in all specimens from both groups (Figure 7).
Discussion
The present study assessed the effect of the lost-wax and CAD-CAM framework fabrication techniques on the fracture strength of porcelain in MCRs. The results showed that the technique of framework fabrication had no significant effect on the fracture strength or the mode of failure of porcelain.
Porcelain chipping/fracture imposes extra costs on patients, and its repair is time-consuming. Both of these factors are clinically important.25 Thus, adequate metal–ceramic bond strength, metal support for the ceramic material, and/or thickness of the ceramic material are prerequisites in MCRs. Metal–ceramic bond strength depends on many factors, and one of the most influential ones is the metal framework fabrication technique.15, 22
The composition of the Co-Cr alloy, the porcelain composition, and the difference in the coefficient of thermal expansion (CTE) of the porcelain and the metal are other factors affecting the fracture strength of porcelain.7, 25, 32, 33 In this study, the CTE of porcelain was 12.9 ±0.5 × 10–6/°C, the CTE of the Magnum Ceramic Co alloy was 14.1 × 10–6/°C and that of the Ceramill Sintron alloy was 14.5 × 10–6/°C; they were all similar. In 2007, Kellerhoff and Fischer measured the fracture strength and thermal shock resistance of MCRs with gold-titanium (Au-Ti) frameworks fabricated with the use of the casting and milling methods.15 Their results were in contrast to our findings, indicating that the fracture strength of the milling group was significantly lower than that of the casting group. This can be due to the different structure of Au-Ti alloy, since during milling, a soft smear of the Au phase is created on the surface, which serves as a barrier against the diffusion of Ti and prevents the formation of a chemical bond to the ceramic.15
No significant difference was noted in the fracture strength of porcelain between the lost-wax and CAD-CAM groups in this study, which may be due to the fact that the specimens had similar oxidation patterns after air abrasion and heat treatment, resulting in similar chemical and mechanical bonding mechanisms. This finding is in agreement with the results of previous studies.20, 21, 23, 32, 33 Previous studies, however, evaluated rectangular or cylindrical specimens, and most of them did not perform thermocycling and/or cyclic loading. In this study, the specimens had the anatomical form of natural teeth, and underwent both thermocycling and cyclic loading to better simulate the clinical setting.34 The adopted thermocycling protocol in this study simulated 2.5 years of clinical service,29 while the cyclic loading protocol simulated 2–3 months of clinical service.31 Also, water has been suggested to play a role in the propagation of small cracks. Thus, we used deionized water instead of artificial saliva, since it has no significant effect on the coefficient of friction between natural teeth.7, 30 Since MCRs have complex geometries, and the effect of the material properties on fracture strength has not been well elucidated, in vitro studies should preferably simulate the clinical setting as much as possible.15 Two previous studies used specimens with natural tooth anatomy, performed thermocycling and cyclic loading, and reported results similar to our findings.7, 24 Another study evaluated the properties of the metal–ceramic bond in restorations with Co-Cr frameworks fabricated by means of the casting, milling and selective laser melting (SLM) techniques.22 They found that metal–ceramic bond strength in the casting group was lower than that in the milling and SLM groups. The variability in the results is likely due to not using specimens with tooth-like anatomy, and not performing thermocycling and cyclic loading.22
The mode of failure in this study was mixed for all specimens in both groups. However, cohesive failure within the ceramic was dominant in most mixed fractures. Some previous studies reported that the metal framework fabrication technique had no significant effect on the mode of failure.15, 20, 21 Similar to our study, Suleiman and von Steyern did not report any adhesive failure and most fractures were mixed; however, in contrast to our findings, some fractures were purely cohesive in their study.7 These results may indicate the insignificant effect of debonding forces on the mode of failure.
In the load test, only a vertical load is applied to specimens. Thus, it only simulates vertical load application in the oral environment and misses the loads applied from other directions. Accordingly, the findings of this in vitro study, like many others, cannot be acceptably generalized to the clinical setting. Future clinical trials and prospective in vivo studies are required to more validly elucidate this topic.
Conclusions
Our results indicated that the fracture strength of porcelain and its mode of failure are independent of the metal framework fabrication technique (lost-wax or CAD-CAM). Therefore, both metal framework fabrication techniques can be recommended in clinical practice.
Ethics approval and consent to participate
Not applicable.
Data availability
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.
Consent for publication
Not applicable.
