You are on page 1of 5

Dental Materials Journal 2012; 31(3): 333337

Interfacial shear bond strength between different base metal alloys and five low fusing feldspathic ceramic systems
Cumhur SIPAHI1 and Mutlu ZCAN2
1 2

Department of Prosthodontics, Academic Center for Dental Sciences, Glhane Military Medical Academy, 06018, Etlik, Ankara, Turkey Dental Materials Unit, Center for Dental and Oral Medicine, Clinic for Fixed and Removable Prosthodontics and Dental Materials Science, Zurich, Switzerland Corresponding author, Cumhur SIPAHI; E-mail:

This study compared the bond strength between metal alloys and 5 ceramic systems. Ceramic systems (Vita VMK68, Ivoclar IPSd. SIGN, Ceramco II, Matchmaker and Finesse) were fired onto either Ni-Cr or Co-Cr base metal alloy. Metal-ceramic interfaces were subjected to shear loading until failure. The ceramic type significantly affected the bond strength results (p<0.05). For Ni-Cr alloy, the results ranged between 15.425.3 MPa and for Co-Cr alloy between 13.319.0 MPa. The highest mean bond strength value was obtained with the combination of Ni-Cr alloy-Ceramco II (25.3 MPa), the lowest bond strength was received from the combination of Co-Cr alloy-Ivoclar IPS d.SIGN ceramic (13.3 MPa). Adhesive failures between metal and ceramic were significantly more frequent with Ni-Cr alloy (31 out of 50) than with Co-Cr (20 out of 50) (p<0.05). Ceramco II presented the highest bond strength with both Ni-Cr and Co-Cr being significantly different from one another. Keywords: Base metal alloy, Low fusing feldspathic ceramic, Metal supported ceramic crown, Metal-ceramic interface, Shear bond strength

Despite the development and growing use of all ceramic systems1,2), ceramic-fused-to-metal (CFM) fixed dental prostheses (FDP) are still considered a good option for oral rehabilitation due to their mechanical strength3-6), presenting also significantly less clinical failures as opposed to all-ceramic ones7). However, the high cost of gold and gold alloys caused an increased trend towards the use of non-precious metal alloys for CFM FDPs. The favourable mechanical properties of non-precious metal alloys allows for the fabrication of restorations with less thickness but more rigidity8-11). When cost and rigidity are considered, nickel-chromium (Ni-Cr) and cobaltchromium (Co-Cr) alloys are the most preferred ones for core materials12). In spite of their satisfying resistance against compressive and lateral occlusal forces, CFM FDPs may fracture on occasion as a result of trauma, inadequate occlusal adjustment, parafunctional habits, flexural fatigue of the metal substructure, incompatibility of the coefficient of thermal expansion between the porcelain and the metal structure or inappropriate coping design13-15). The metal-ceramic bond interface is critical for the functional and aesthetic success of CFM restorations. The strength of the metal-ceramic bond is usually dictated by the chemical bond (oxide layer formed on the metal surface forms metallic, ionic, and covalent bonds with oxides in the opaque ceramic), mechanical interlocking (physical engagement of ceramic into micropores created on the metal surface), van der Waals forces (attraction based on molecular charge), and compressive forces originating from coefficient of thermal expansion6,7-11).
Received Jun 27, 2011: Accepted Jan 6, 2012 doi:10.4012/dmj.2011-143 JOI JST.JSTAGE/dmj/2011-143

Among these factors, coefficient of thermal expansion (CTE), the rate at which a material contracts during cooling, can create strong shear stresses at the metalceramic interface when the combined materials are mismatched7). It has been previously suggested that the CTE of the metal substructure should be slightly higher (approximately 0.5106 K1) than that of the applied ceramic, exposing the ceramic in residual compression upon cooling16). Parameters affecting the metal-ceramic bond strength (MCBS) have been widely studied. Drummond et al.17) reported that the MCBS of ceramic fused to noble metal alloys were significantly higher than the bond strengths of ceramic fused to base metal alloys. Hammad et al.18) demonstrated that firing of opaque porcelain 18C above the recommended firing temperature significantly increased the MCBS. Anusavice et al.19) stated that metal-ceramic bond tests display two main problems; the absence of pure shear loading at metal-ceramic interface, and existing discontinuities in stresses at the ceramic termination points. They added that metal-ceramic specimens contain residual thermal stresses at the interfaces due thermal expansion coefficient mismatch between ceramics and casting alloys. De Melo et al.20) determined that the longevity of metal-ceramic restorations depends on the formation of a stable adhesive layer between the two materials. Although not completely defined, they claimed that the adhesion between ceramic and metal depends on suitable oxidation of the metal and interdiffusion of ions between metal and ceramic. Additionally, Klnk et al.21) reported that the bond strength between ceramic and casting alloy is determined by the strength of chemical bonds,


Dent Mater J 2012; 31(3): 333337

(Bego, Bremen, Germany). The wax patterns were then retrieved from silicon impression and invested with phosphate-bonded investment material (Deguvest CF, Degussa, Hanau, Germany). Following wax elimination, 50 patterns were cast from a Ni-Cr alloy (Remanium CS, Dentaurum, Ispringen, Germany) and the other 50 patterns were cast from a Co-Cr alloy (Triloy, Dentaurum, Ispringen, Germany) according to the manufacturers instructions. After casting, metal specimens were divested, finished with a laboratory tungsten bur (H79NEF, 104.023, Brasseler USA, Savannah, GA), clamped on a low-speed handpiece (KaVo EWL, Type 4005, Leutkirch im Allgu, Germany) at a rotational speed of 10,000 rpm to remove positive casting defects. The specimen surfaces were cleaned using steam spraying (Triton SL, Bego, Bremen, Germany). Metal surfaces that would come into contact with ceramic materials were sandblasted with 50 m alumina particles for 30 s and all specimens were subjected to degassing and oxidation. Five low fusing feldspathic ceramic systems with different CTEs; Vita (VMK68, Vita Zahnfabrik, Bad

mechanical interlocking, the type and concentration of defects at the interface, wetting properties, and the degree of compressive stress in the ceramic layer due to differences in the coefficients of thermal expansion between the metal and ceramic. The aim of this study was to compare the interfacial shear bond strength between two different base metal alloys and five low fusing feldspathic ceramic systems and evaluate the failure types. The hypotheses tested were that; a) ceramics with CTE less (approximately 0.5106 K1) than those of the metals would result in higher bond strength and that b) there would be no differences between the alloy types.


Specimen preparation Cylindrical brass specimens (n=10) with standard dimensions were fabricated in a CNC milling machine (Tezsan, Gebze, Turkey) (Fig. 1). Each brass specimen was duplicated 10 times using condensation type silicone impression material (Zetaplus, Zhermack, Rovigo, Italy) into 100 cylindrical PVC rings of 3 cm height and 1.5 cm diameter. The impression material was condensed into the PVC ring and the brass specimen was introduced into the ring in the way that the wider extremity was positioned at the top of the ring. After putty silicone was set, brass specimens were retrieved from the rings. The negative spaces in the putty silicone bulks were filled with light body silicone and the brass specimens were replaced into the rings to obtain an accurate and precise impression. After retrieval of the brass specimens, the impression cavities were filled with casting wax
Table 1

Fig. 1

Dimensions of cylindrical cast metal support (mm)

The brands, abbreviations, types, manufacturers, firing temperatures for ceramics, melting temperatures of alloys and thermal expansion coefficients of the ceramics and alloys used in this study Material type Low fusing feldspathic ceramic Low fusing feldspathic ceramic Low fusing feldspathic ceramic Low fusing feldspathic ceramic Ultra low fusing feldspathic ceramic Ni-Cr (61% Ni, 26% Cr) Co-Cr (62% Co, 24% Cr) Manufacturer Vita Zahnfabrik, Bad Sckingen, Germany Ivoclar Vivadent, Schaan, Liechtenstein Ceramco Inc, Burlington, NJ Letchworth, Hertfordshire, UK Dentsply, Ceramco, York, PA Dentaurum, Ispringen, Germany Dentaurum, Ispringen, Germany Firing Temperature 1st Opaque: 950C Dentin: 930C 1st Opaque: 900C Dentin: 880C 1st Opaque: 915C Dentin: 890C 1st Opaque: 980C Dentin: 950C 1st Opaque: 790C Dentin: 760C Melting temperature: 1,668C Melting temperature: 1,750C Coefficient of Thermal Expansion (K1 m/m) 15.2106 12106

Brand name Vita VMK68 (VMK) Ivoclar IPS d.SIGN (IVO) Ceramco II (CER) Matchmaker (MAT) Finesse (FIN) Remanium CS Triloy

13.5106 13.8106 14.5106 14106 14.4106

Dent Mater J 2012; 31(3): 333337


Fig. 3 Fig. 2 Dimensions of metal-ceramic specimens

Loading pattern interfaces




Sckingen, Germany), Ivoclar (IPS d.SIGN, Ivoclar Vivadent, Schaan, Liechtenstein), Ceramco (Ceramco II, Ceramco Inc, Burlington, NJ), Matchmaker (Letchworth, Garden City, Hertfordshire, UK ) and Finesse (Dentsply, Ceramco, York, PA) were fired onto either Ni-Cr or Co-Cr base metal alloy (Table 1). Additional prefabricated PVC rings of 5 mm height and 5 mm in diameter with two apertures in both extremities were placed at the centre of circumferential metal specimen surfaces and secured with sticky wax. These rings were used to provide uniform shaping of ceramic materials on metal surfaces (Fig. 2). Opaque ceramic of each brand was prepared according the instructions of the manufacturers and condensed onto the metal surfaces, fixed in PVC rings to form a layer of 0.5 mm high. Following firing of one opaque layer in a porcelain firing oven (VITA Vacumat 40T, VITA Zahnfabrik, Germany), a second opaque layer was condensed and fired on the initial opaque layer to obtain 1 mm thick opaque ceramic layer for each specimen. Dentin ceramic was then condensed onto the opaque ceramic layers into the PVC rings at a height of 4 mm, to reach a total ceramic height of 5 mm (Fig. 2). Subsequently, cylindrical ceramics were fired according to the instructions of the manufacturers. After cooling, metal-ceramic specimens were finished with a medium-grit laboratory diamond bur (836-11, Brasseler, Savannah, GA) clamped on a low-speed hand-piece (KaVo EWL, Type 4005, Leutkirch im Allgu, Germany). In total 100 metal-ceramic specimens were assigned to 10 experimental groups (n=10 per group). Testing procedure and failure analysis The specimens were fixed to the aluminium base of the testing machine, positioned perpendicular to the horizontal plane. Shear force was applied to the metalceramic interface in a Universal Testing Machine (Instron 1195, Instron Corp, Canton, MA) with a shearing blade at a crosshead speed of 0.5 mm/min until failure

occurred (Fig. 3). The failure threshold was defined as the point at which the loading force reached the maximum value for breaking off the metal-ceramic junction or for fracturing the ceramic. The applied load was automatically stopped at the first instance of specimen fracture. Peak force values at failure were recorded in Newtons and divided by the surface area to obtain the bond strength values (MPa). After debonding the specimens were evaluated by two operators, and failure types were classified as follows: Adhesive: No ceramic left on the metal surface, Cohesive: Failure within the opaque ceramic without metal exposure, and Mixed: Partially ceramic left on metal surface. Statistical analysis Data were analyzed with a statistical software program (SPSS 9.0 for Windows, SPSS Inc, Chicago, Ill). Bond strength data (MPa) were submitted to Mann-Whitney U test with the bond strength as the dependent variable and metal type (2 levels; Ni-Cr, Co-Cr) and ceramic types (5 levels) as independent variables. Multiple comparisons were made using Tukeys test. Failure types were statistically analyzed using t-test. P values less than 0.05 were considered to be statistically significant in all tests.

Mean bond strength results and significant differences between the experimental groups are presented in Table 2. While the metal type did not affect the results significantly (p>0.05), ceramic type (p<0.05) significantly affected the bond strength results. Interaction terms were also significant (p<0.05). For Ni-Cr alloy, the results ranged between 15.425.3 MPa and for Co-Cr alloy between 13.319.0 MPa. While the highest mean bond strength value was obtained with the combination of Ni-Cr alloy and CER (25.3 MPa), the lowest bond strength was received from the combination of Co-Cr alloy and IVO (13.3 MPa).

Table 2

Dent Mater J 2012; 31(3): 333337

The mean shear bond strength values (MPa) and standard deviations of experimental groups. The same letters in the same column indicate no significant differences (=0.05) Mean Bond Strength (MPa) 18.2e 25.3a 15.4c 15.4c 18.4c,e 18.0e 19.0b,e 13.3d 14.8d 18.3e Standard Deviations 6.26 2.02 1.62 1.0 1.45 4.53 1.38 1.34 0.6 1.0 Table 3 Distribution and frequency of failure types per experimental group analyzed after bond strength test. (Adhesive: No ceramic left on metal, Cohesive: Failure within ceramic material without metal exposure, Mixed: Some ceramic left on metal) Adhesive 2 0 2 2 1 2 2 5 3 2 Cohesive 3 3 2 2 4 3 3 2 2 3 Mixed 5 7 6 6 5 5 5 3 5 5

Groups (n=10) Ni-Cr/VMK Ni-Cr/CER Ni-Cr/IVO Ni-Cr/MAT Ni-Cr/FIN Co-Cr/VMK Co-Cr/CER Co-Cr/IVO Co-Cr/MAT Co-Cr/FIN

Groups Ni-Cr/ VMK Ni-Cr/ CER Ni-Cr/ IVO Ni-Cr/ MAT Ni-Cr/ FIN Co-Cr/ VMK Co-Cr/ CER Co-Cr/ IVO Cr-Co/ MAT Cr-Co/ FIN

CER presented the highest bond strength with both Ni-Cr (25.3 MPa) (p<0.001) and Co-Cr (19 MPa) (p<0.001) being significantly different from one another. The modes of failure were presented at Table 3.

This study was undertaken in order to compare the interfacial bond strength between two base metal alloys and low fusing feldspathic ceramic systems with the objective of finding the best combination. The alloy type did not significantly affect the bond strength results (p>.05) but the interaction terms were significant and failure types also varied significantly between the alloy types. Thus, the second hypothesis that there would be no differences between the alloy types could only be partially accepted. On the other hand, it was mentioned in the first hypothesis that ceramics with CTE less (approximately 0.5106) than those of the metals would result in higher bond strength16). In fact, ceramic types significantly affected the results (p<0.05). The Ni-Cr and Co-Cr base alloys had CTE of 14106 and 14.4106, respectively. The highest bond strength values were obtained with the group Ni-Cr/CER (25.3 MPa). Such a result may be attributed to the CTE of Ceramco II ceramic system (13.5106) which has the second closest CTE to that of Remanium CS (Ni-Cr) metal alloy (14.0106) after the Matchmaker ceramic system. However, contrary to this result, the Matchmaker ceramic system, having a very close CTE (13.8106) to that of the metal showed lower bond strength results. This paradoxal result may be explained with the highest firing temperature of the opaque ceramic layer of Matchmaker ceramic system (980C). Such a high firing temperature probably increases the oxide layer formation of base metal alloys, resulting to a thick oxide layer which causes failure in

bonding. Regarding to the possible effect of CTE of the ceramics on metal-ceramic bond, it has been previously suggested that the CTE of the metal substructure should be approximately 0.5106 higher than that of the applied ceramic16). With the findings of this study, a trend for increased bond strength was not observed when CTE of the ceramics were lower than those of the alloys. With the ceramics having similar or even higher CTE as in the case of FIN and VMK respectively, also high bond strength values were obtained. In a similar study, bond strength of a feldspathic ceramic system (Ivoclar IPS d.SIGN) combined with two types of Ni-Cr alloys (4-ALL and Wiron 99) and two types of Co-Cr alloys (IPS d.SIGN 20 and Argeloy NP) ranged between 71.719.2 MPa and 5420 MPa20). Considering the high standard deviations reported20), in the present study with the same ceramic, lower but more consistent results were obtained. In this study, the highest and lowest shear bond strength values varied between 13.3 and 25.3 MPa in general and with IVO between 13.3 and 15.4 MPa. The differences between the results of this study and those of de Melo et al.20) could be due to the variations in the test method. Metal-ceramic adhesion could be tested with various methods such as shear strength test, three- or four-point bending and biaxial flexural test4,6,12,17,18). Although ceramic is resistant to compressive and tensile forces, it shows less resistance to shear and oblique forces where the latter represents the worse-case situation9,11). Similarly, slow cross-head speed was reported to deliver more accurate results9,11). For these reasons, oblique forces were applied under a slower cross-head speed compared to that of the study of de Melo et al.20). The results of bond strength tests should also be coupled with the failure analysis. In Co-Cr groups, regardless of the ceramic type, 23 mixed, 13 cohesive,

Dent Mater J 2012; 31(3): 333337

and 14 adhesive type failures were seen among 50 samples. In Ni-Cr groups, 29 mixed, 14 cohesive, and 7 adhesive type failures were seen regardless of the ceramic type. Cohesive failures within the ceramic indicate that the adhesion between the metal-ceramic exceeded the cohesive strength of the ceramic. This phenomenon can be attributed to the thicker oxide layer formed on cobalt containing alloy surfaces that occurs during firing10,11). While certain level of oxide thickness facilitates better adhesion with the ceramic, increased thickness may weaken the chemical bond yielding to cohesive failures within the oxide layer that can be observed in the form of adhesive failure between the metal-ceramic interfaces with no remnants of ceramic on the metal surface. It was determined that adhesive failure types were 50% lower in Ni-Cr groups (7 failures) compared to Cr-Co groups (14 failures). In Ni-Cr/CER group which displayed the highest bond strength value (25.3 MPa), not any adhesive failure type was seen (3 cohesive and 7 mixed type failures). However, in Cr-Co/IVO group in which lowest bond strength values were obtained (13.3 MPa), 5 adhesive type failures were seen, with 2 cohesive and 3 mixed failure types. In the present study, the Co-Cr and Ni-Cr metal surfaces were sandblasted with 50 alumina particles before ceramic firing. Klnk et al.21) used 50 and 110 alumina particles to determine the effect of alumina particle size on metal-ceramic bond strength and concluded that higher bond strength values were obtained with 110 alumina particles. On the other hand, the difference of chemical composition and stiffness between Co-Cr and Ni-Cr alloys may also affect the surface roughness of tested metal surfaces after a standard sandblasting procedure. The surface treatment of metal substrates may also play an important role in the formation and thickness of the interactive oxide layer which directly affects the bond between ceramic and metal. It was reported that the thinner the oxide layer on the metal surface, the stronger the bond between the two materials19). It is often difficult to predict the metal types but clinical findings indicated that CFM FPD failures are frequently experienced in the form of ceramic fractures with metal exposures3,4). Whether the values obtained in this study are clinically sufficient or not needs to be verified in clinical studies. Future investigations should verify the findings of this study with other test methods where thinner ceramics are used and fatigue parameters are also incorporated in the study design.


Co-Cr alloys. Not only the bond strength but also the failure types should be considered in future studies when evaluating metal-ceramic bond strengths.

1) Guazzato M, Albakrya M, Ringerb SP, Swain MV. Strength, fracture toughness and microstructure of a selection of all-ceramic materials. Part I. Pressable and alumina glassinfiltrated ceramics. Dent Mater 2004; 20: 441-448. 2) Rizkalla AS, Jones DW. Mechanical properties of commercial high strength ceramic core materials. Dent Mater 2004; 20: 207-212. 3) zcan M, Niedermeier W. Clinical study on the reasons for and location of failures of metal-ceramic restorations and survival of repairs. Int J Prosthodont 2002; 15: 299-302. 4) Almilhatti HJ, Giampaolo ET, Vergani CE, Machado AL, Pavarina AC. Shear bond strength of aesthetic materials bonded to Ni-Cr alloy. J Dent 2003; 31: 205-211. 5) zcan M. Evaluation of alternative intra-oral techniques for fractured ceramic-fused-to-metal restorations. J Oral Rehabil 2003; 30: 194-203. 6) Schweitzer DM, Goldstein GR, Ricci JL, Silva NR, Hittelman EL. Comparison of bond strength of a pressed ceramic fused to metal versus feldspathic porcelain fused to metal. J Prosthodont 2005; 14: 239-247. 7) Craig RG, Powers JM. Restorative Dental Materials. 2nd Ed. St. Louis: Mosby; 2002. p. 576-581. 8) Sced IR, McLean JW. The strength of metal/ceramic bonds with base metals containing chromium. A preliminary report. Br Dent J 1972; 132: 232-234. 9) McLean JW. The Science and Art of Dental Ceramics. Volume I. The Nature of Dental Ceramics and their Clinical Use. Chicago: Quintessence Publishing Co; 1979. p. 79-82. 10) Murakami I, Schulman A. Aspects of metal-ceramic bonding. Dent Clin North Am 1987; 31: 333-343. 11) Shillingburg HT, Hobo S, Whitsett LD. Fundamentals of Fixed Prosthodontics. 3rd Ed. Chicago: Quintessence; 1997. p. 455. 12) Prado RA, Panzeri H, Fernandes Neto AJ, Neves FD, Silva MR, Mendonca G. Shear bond strength of dental porcelains to nickel-chromium alloys. Braz Dent J 2005; 16: 202-206. 13) zcan M. Fracture reasons in ceramic-fused-to-metalrestorations. J Oral Rehabil 2003; 30: 265-269. 14) Haneda IG, Fonseca RG, Adabo GL, Cruz CA. Shear bond strength of metal-ceramic repair systems. Gen Dent 2009; 57: 644-651. 15) Lombardo GH, Nishioka RS, Souza RO, Michida SM, Kojima AN, Mesquita AM, Buso L. Influence of surface treatment on the shear bond strength of ceramics fused to cobalt-chromium. J Prosthodont 2010; 19: 103-111. 16) Wataha JC. Alloys for prosthodontic restorations. J Prosthet Dent 2002; 87: 351-363. 17) Drummond JL, Randolph RG, Jekkals VJ, Lenke JW. Shear testing of the porcelain-metal bond. J Dent Res 1984; 63: 1400-1401. 18) Hammad IA, Goodkind RJ, Gerberich WW. A shear test for the bond strength of ceramo-metals. J Prosthet Dent 1987; 58: 431-437. 19) Anusavice KJ, Dehoff PH, Fairhurst CW. Comparative evaluation of ceramic-metal bond tests using finite element stress analysis. J Dent Res 1980; 59: 608-613. 20) De Melo RM, Travassos AC, Neisser MP. Shear bond strengths of a ceramic system to alternative metal alloys. J Prosthet Dent 2005; 93: 64-69. 21) Klnk T, Kurt M, Ural , Klnk , Baba S. Effect of different air-abrasion particles on metal-ceramic bond strength. J Dent Sci 2011; 6: 140-146.

The findings of bond strength data obtained in this study suggest that both Ni-Cr and Co-Cr could be used in conjunction with the ceramics tested but failure type analysis showed more adhesive failures with Ni-Cr indicating less favourable adhesion of the ceramics on this alloy compared to Co-Cr. The highest bond strength results were obtained with Ceramco II on both Ni-Cr and