- Email: [email protected]
Tensile bond strength of dual curing resin-based cements to commercially pure titanium
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Tensile bond strength of dual curing resin-based cements to commercially pure titanium Rafael Schneider a,∗ , Mario Fernando de Goes b , Guilherme Elias Pessanha Henriques c , Daniel C.N. Chan d a
` Lutheran University of Brazil, School of Dentistry, Dental Materials Area, Av. Universitaria, 95560-00 Torres, RS, Brazil Department of Restorative Dentistry, Dental Materials Area, Piracicaba Dental School at Campinas State University, Piracicaba, Brazil c Department of Periodontics and Prosthodontics, Piracicaba Dental School at Campinas State University, Piracicaba, Brazil d Department of Oral Rehabilitation, Division of Operative Dentistry, Medical College of Georgia, Augusta, GA, USA b
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. The aim of this study was to evaluate the tensile bond strength of dual curing
Received 27 July 2005
luting resin cements to commercially pure titanium at 10 min and 24 h after removal of the
Accepted 5 December 2005
oxide layer. Methods. One hundred and twenty titanium discs were obtained by casting and polishing with silicon carbide papers. The titanium discs were sandblasted with 50 m aluminum
Keywords:
oxide, ultrasonic cleaned and bonded in pairs with the resin-based cements Panavia F and
Titanium
Rely X ARC at 10 min and 24 h after the sandblasting. The tensile test was performed with a
Adhesion
crosshead speed of 0.5 mm/min in an Instron Universal testing machine.
Resin-based cement
Results. The Rely X ARC reached the highest tensile strength value at 24 h after sandblast-
Primer
ing (18.27 MPa), but there was no statistically significant difference between the two dual curing resin cements for both times tested. All specimens showed a mixture of cohesive fracture in the resin cement and adhesive failure. However, the predominant failure mode for Panavia F was cohesive in resin cement, and the Rely X ARC exhibited a greater proportion of specimens with adhesive failure between the alloy and resin luting cement at 10 min and 24 h. Significance. Both cements had, statistically, the same tensile bond strength. But in the fracture mode analysis, the adhesive predominant fracture mode of Rely X ARC cement indicates a premature clinical adhesive failure. On the other hand, the cohesive predominant fracture mode of Panavia F indicates a longer clinical adhesive bond with titanium. © 2005 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Titanium is popular in dentistry as a substitute material for the noble and non-noble alloys because of its superior biocompatibility, good mechanical properties and high corrosion resistance [1–4]. Prostheses with titanium have been indicated for a metallic framework in fixed partial denture, full
∗
crowns, prostheses with multiple units, adhesive prostheses, root posts and framework for dentures [5–8]. In spite of these characteristics, the relationship between titanium and the adhesive luting materials remains unclear. In an attempt to reach an effective adhesion between titanium and resin-based cement, a PMMA composite resin with the acidic monomer 4-methacryloyloxymethyl trimellitate anhy-
Correspondence address: R. Luiz Linck Barcelos 220, CEP 94015-590, Gravatai, RS, Brazil. Tel.: +55 51 30424340; fax: +55 51 6262000. E-mail address: [email protected] (R. Schneider). 0109-5641/$ – see front matter © 2005 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2005.12.006
82
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
dride (4-META) was used on the titanium surface and showed stable bond strength [9]. The acidic monomers derived from the carboxylic (4-META) or phosphoric (methacryloyloxydecyl di-hydrogen phosphate) acids are able to bond with the superficial oxide layer by Bolger’s mechanism through electrostatic interaction between the acids of the monomers and the OH− groups of the superficial oxide layer, depending on the isoelectric metallic oxide points and the constant acid dissociation of the monomers [10]. The oxide layer is formed during titanium casting. Due to its great reactivity with elements like O, C, H and N at high temperatures, it produces a firm oxide layer adherent with the metallic substrate. The oxide layer is formed very quickly ˚ (10−9 s), has high chemical stability and is very thin (50–100 A) [11–13]. When it is mechanically removed from the surface, it is formed again just by contact with oxygen. However, the oxide layer formation is very unstable. The exposition period to the high temperatures and to ambient contaminants has greatly affect the layer adherence and thickness. Thus the storage period, time and temperature of casting are determinant factors of the composite resin bond strength to the metal surface before the luting procedures. The use of light-cured resin cements may not be possible because the visible light is blocked by the titanium metallic structure. Besides, the self-curing resin cements have low bond strength in the first hour of the luting procedure. This means that the restoration cannot be submitted under mastication stress before the first hour to avoid its dislodgement. If the restoration moves, the results will be microleakage and second caries [14]. Thus, dual curing resin-based cements must be chosen as the luting material for cement metallic prostheses. Considering the use of titanium as the metallic framework in fixed partial dentures, the bonding capacity of the acidic monomers to the metal oxide layer, the ambient exposition time and the effect of temperature on the formation of titanium oxide layer, the aim of this study was to determine the tensile bond strength of dual curing resin-based cements to commercially pure titanium at 10 min and 24 h after the removal of oxide layer. Additionally, the specimens’ fracture mode was observed after the tensile test.
Table 1 – Luting agents used and their compositions Compositiona
Material
Panavia F
Rely X ARC
a
Paste A: colloidal silica, Bis-GMA, hydrophobic and hydrophilic dimethacrylate, benzoic peroxide Paste B: silanized barium glass, titanium oxide, sodium fluoride, colloidal silica, Bis-GMA, hydrophobic and hydrophilic dimethacrylate, diethanol-p-toluidine, T-isopropylic benzenic sodium sulfinate ED Primer A: hydroxyethylmethacrylate (HEMA), 10-methacryloyloxydecyl di-hydrogen phosphate (MDP), NM-aminosalicilic acid, diethanol-p-toluidine, water ED Primer B: NM-aminosalicilic acid, T-isopropylic benzenic sodium sulfinate, diethanol-p-toluidine, water Alloy Primer: 6-n-4-vinylbenzyl propylamino di-thione triazine (VBATDT), 10-methacryloyloxydecyl di-hydrogen phosphate (MDP) Paste A: Bis-GMA, tri-ethylene glycol dimethacrylate, zircon/silica filler (68 wt%), photoinitiators, amine, pigments Paste B: Bis-GMA, tri-ethylene glycol dimethacrylate, benzoic peroxide, zircon/silica filler (67 wt%) Ceramic Primer: gamma-methacryloxypropyl trimethoxysilane, ethanol, water
Manufacturers’ information.
Products, Rio de Janeiro, RJ, Brazil) and polished with #320, #400 and #600 silicon carbide paper. The cylinders’ edges and acrylic resin around the disc were removed to facilitate the bonding and photocuring procedures. A hole was made in the bottom of cylinders to attach it to the tensile test system. A 4 mm diameter central area of discs’ surfaces was sandblasted with aluminum oxide (50 m) for 5 s at 5 mm from the surface with 80-psi pressure. The specimens were ultrasonically cleaned for 10 min and divided into four groups (30 discs
Table 2 – Manipulation sequence of the luting agents
2.
Materials and methods
The resin-based cements tested in this study are summarized in Table 1. Commercially pure titanium ingots (Rematitan® , Dentaurum, Ispringen, BW, Germany) (Ti > 99%) were used. One hundred and twenty cone trunk-shaped casting wax discs were made using a metallic matrix, fixed in a crucible forming base with wax rods and embedded in phosphate investment (Rematitan® Plus-Dentaurum, Ispringen, BW, Germany). After the wax elimination and investment expansion thermal cycle in an electric furnace (EDGCON 5N-EDG ˜ Paulo, SP, Brazil), the commercially pure titaEquipments, Sao nium ingots were cast in an electric arc casting machine (Rematitan® ) and injected into the investment, following the manufacturer’s instructions. After casting, the discs were separated from the supplying ducts, embedded in a plastic cylinder with self-cured acrylic resin (JET-Classico Dental
Material
Panavia F
Rely X ARC
Manipulation Applying of the Alloy Primer on bonding surface Mixture of one drop of ED Primer A and ED Primer B Applying of the mixture on the bonding surface for 1 min Gently air drying Mixture of equal lengths of Paste A and B for 10 s and applying on bonding surface Excess removal Photocuring for 40 s in each surface Applying the Oxyguard II on the composite resin exposed surface for 3 min and washing Applying the Ceramic Primer on the bonding surface Gently air drying Mixture of equal lengths of Pastes A and B for 10 s and applying on bonding surface Excess removal Photocuring for 20 s in each surface
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
83
per each group). Fingernail varnish was applied on the nonsandblasted surfaces to avoid any interference in the bond strength. The bonding procedures were performed immediately in the Groups 1 and 3 and 24 h after the oxide-layer removal in the Groups 2 and 4. Panavia F (Kuraray, Medical Inc., Tokyo, Japan) was used to bond the specimens in the Groups 1 and 2 and Rely X ARC (3 M Dental Products, Saint Paul, MN, USA) in the Groups 3 and 4 following the manufacturer’s instructions (Table 2). In the Groups 1 and 2, a primer containing MDP and VBATDT monomers (Alloy Primer, Kuraray Medical Inc., Tokyo, Japan) was applied on the metal surface to bond composite resin. In Groups 3 and 4, a silane-coupling agent (Ceramic Primer-3 M Dental Products, Saint Paul, MN, USA) was applied on the metal bond surface.
Fig. 2 – Tensile test device. (A) Universal test machine crosshead. (B) Specimen held by the tensile system bar at the tensile test device and (C) tensile system bars.
Fig. 1 – Specimens bonded in pairs: (A) plastic cylinder, (B) hole to attach the tensile system bar, (C) titanium discs and (D) resin-based luting agent.
Thereafter, equal lengths of composite resins base paste and catalyst paste were mixed for 10 s. The mixture was applied on one of the isolated sandblasting surfaces and the titanium discs were bonded in pairs with finger pressure. A cylindrical plastic matrix was used to keep the discs centered and aligned in the correct place. Then, the excesses were removed with a paint brush and the composite resins were photocured. Panavia F was photocured for 20 s in each of the four different positions and Rely X ARC 40 s in each position. On the external surface of Panavia F composite resin, the Oxyguard II alcoholic gel (Kuraray Medical Inc., Tokyo, Japan) was applied to permit the curing of this surface. After the curing, the specimens (Fig. 1) were stored in distilled water at 37 ◦ C for 24 h. Hence, the tensile test was performed in a Universal Testing machine (Instron Co., Canton, MA, USA) at 0.5 mm/s crosshead speed (Fig. 2). The tensile test values were recorded in MPa and submitted to two-way ANOVA (˛ = 0.05). To determine the fracture mode, the fractured specimens in the tensile test were observed in a 60× stereoscopic loupe (Carls-Zeiss, Oberkochen, BA, Germany). Thereafter, the discs were removed from the acrylic resin, ultrasonically cleaned, gold sputtered and observed by a scanning electron microscope (Leo 435 VP, Orsay Phisics, Fuveau, ZA, France).
84
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
Table 3 – Tensile bond strength values (means and standard deviations) between the tested groups Groups
Tensile bond strength meansa
Rely X ARC—24 h (Group 4) Panavia F—24 h (Group 2) Panavia F—10 min (Group 1) Rely X ARC—10 min (Group 3)
18.27 (7.34) 14.76 (7.71) 13.96 (3.60) 13.90 (6.08)
Values in parentheses are standard deviations. a
Values without statistically significant difference in two-way ANOVA and F-test (p > 0.05).
3.
in the Groups 1 and 2, probably because of the Panavia F consistency.
Results
The tensile bond strength values were transformed (x + 0 square root) prior to statistical analysis. The means and standard deviations are listed in Table 3. Higher mean was obtained in the Group 4 (Rely X ARC—24 h), followed by Group 2 (Panavia F—24 h), Group 1 (Panavia F—10 min) and Group 3 (Rely X ARC—10 min). The bond strengths ranged from 5.15 to 30.23 MPa for Panavia F composite resin and from 3.9 to 29.34 MPa for the Rely X ARC. Table 4 shows the results analyzed according to two-way ANOVA. The bond strength was neither influenced by storage time nor luting material and there was no statistically significant difference among the bond strengths of the tested groups. Optical and SEM observations revealed that all specimens showed a mixture of adhesive and cohesive failures on the bond surface regardless of the luting material or the storage time (10 min or 24 h after the oxide removal). Hence, a composite resin cohesive failure in some areas as well as total debonding between titanium and composite resins (adhesive failure), characterizing the mixed failure, was observed. However, in the Groups 1 and 2, where Panavia F composite resin was used 10 min and 24 h after sandblasting, respectively, there was the predominance of the composite resin cohesive failure in more than 50% bonding area in all specimens (Figs. 3 and 4). In Groups 3 and 4, where Rely X ARC composite resin was used 10 min and 24 h after sandblasting, respectively, the predominant failure was adhesive between the titanium and the composite resin in more than 50% bonding area in all specimens (Figs. 5 and 6). It was observed that air bubbles were incorporated in the composite resins in all groups, which would have been caused by the composite resins manipulation. Nevertheless, there was a major bubble entrapping
4.
Discussion
The Rely X ARC is a Bis-GMA based composite resin and was developed to bond with all indirect restorative materials (polymers, metals or ceramics). In this study, the Rely X ARC composite resin showed a mean bond strength of 13.9 MPa, 10 min after sandblasting and 18.27 MPa, 24 h after sandblasting. These values were not statistically significantly different, which means the oxide layer removal and neo-formation did not have any effect on the composite resins’ tensile bond strength to titanium, although the higher numeric difference was seen at 24 h period. The self-limiting characteristic of superficial oxide layer formation on titanium surface might have been responsible for the results of statistical equality among the tested groups. Actually, during the oxide layer formation and lengthening, it becomes a protector barrier itself, avoiding the oxygen contact with the underneath metal, interrupting its formation. This phenomenon occurred in the same way with Panavia F composite resin, which showed mean values of 13.96 MPa, 10 min after sandblasting and 14.76 MPa, 24 h after the sandblasting, which are not statistically significantly different. The utilization of acidic resin monomers, like MDP contained in the Panavia F Alloy Primer, would be able to produce an effective and lasting bond between composite resins and basic metals [15–25]. This bond would occur throughout chemical links between the monomer phosphate radicals and the basic metal oxide layer. Ohno et al. [10,17] described the Bolger’s mechanism which would be responsible for this bond. In this mechanism, an electrostatic interaction between polymer acids or bases and hydroxyl groups of the metal surface would occur, depending on the isoelectric point of metal oxides and acid dissociation constants of acidic adhesive monomers. Although the efficacy of an acid monomer, like MDP, has been considered by many authors, the mean tensile bond strength reached in this study by Panavia F composite resin that contains MDP in its composition, did not show statistically significant difference with Rely X ARC composite resin, which the manufacturer recommends only for the application of a silane agent. Silicon, the principal component of silane agents, has a great affinity for metal ions of the superficial oxide layer of titanium, producing a chemical bond like Panavia F Alloy Primer [26,27].
Table 4 – Results of analysis of variance for tensile bond strength Source of variation
d.f.
Material Time Mat/time Residual
1 1 1 56
0.5633769 1.1581963 1.1048372 40.5206772
Total
59
43.3470877
General mean = 3.808355, variation ratio = 22.336%.
Sum of squares
Mean square 0.5633769 1.1581963 1.1048372 0.7235835
F-value
p-Value
0.7786 1.6006 1.5269
0.61484 0.20852 0.21948
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
85
Fig. 3 – SEM of fracture mode of Groups 1 and 2 specimens, showing the mixed mode of fracture with predominance of composite resin cohesive failure. (A) Titanium surface with partially bonded composite resin, (B) higher magnification SEM of the marked area in (A): A, composite resin cohesive failure area; B, adhesive failure area; C, pore.
The numerical match between Rely X ARC and Panavia F could occur because all specimens were sandblasted, creating an equal surface pattern in each specimen. Thus, the micromechanical retention became the principal bonding to titanium for both cements, which could explain the similar results of the tested groups [28]. Other sandblasting influence is in the phenomenon of alumina particles incrustation during the sandblasting procedure. These particles stay stuck on the titanium surface because of the velocity and the pressure that they hit it with and cannot be removed even by ultrasonic cleaning or acid etching. So, these non-removable
alumina particles become responsible for the chemical bond of the Alloy Primer and silane agents to themselves, increasing the bond strength of both cements, the Panavia F and Rely X ARC, decreasing the numerical difference between them and resulting in similar results [27,28]. A fact, which could be an influencing factor in the results, was the incorporation of lot of bubbles in the Panavia F manipulation. This composite resin is a little putty, which could contribute to the incorporation of maximum bubbles. Rely X, on the other hand, is a fluid which could minimize this phenomenon.
Fig. 4 – SEM of fracture mode of Groups 1 and 2 specimens, showing the mixed mode of fracture with predominance of composite resin cohesive failure. (A) Opposite disc surface showed in Fig. 3A. (B) Higher magnification SEM of marked area in (A): A, composite resin cohesive failure area; B, adhesive failure area; C, pore; D, sample preparation artifact.
86
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
Fig. 5 – SEM of fracture mode of Groups 3 and 4 specimens, showing the mixed mode of fracture with predominance of adhesive failure. (A) Titanium surface with partially bonded composite resin, (B) higher magnification SEM of marked area in (A): A, composite resin cohesive failure area; B, adhesive failure area; C, pore.
Fig. 6 – SEM of fracture mode of Groups 3 and 4 specimens, showing the mixed mode of fracture with predominance of adhesive failure. (A) Opposite disc surface showed in Fig. 5A. (B) Higher magnification SEM of marked area in (A): A, composite resin cohesive failure area; B, adhesive failure area; C, pore.
Though both composite resins had shown similar tensile bond strengths to titanium, the fracture shape on all Panavia F specimens (Groups 1 and 2) after the tensile test was always mixed, regardless of the storage time (10 min or 24 h after sandblasting). However, there was predominance of composite resin cohesive failure on more than 50% bonding surface area (Figs. 3 and 4). In all Rely X ARC specimens as well, the mixed failure occurred, but there was predominance of the adhesive failure on more than 50% bonding surface area (Figs. 5 and 6). This shape fracture difference seems to be linked to the most wetting ability of Panavia F, due to MDP and VBATDT, resulting in the best contact area between the Panavia F and titanium surface. Another reason for the Panavia F cohesive predominant fracture pattern is its
high adhesion capacity that exceeds its own cohesive strength [29–32].
5.
Conclusion
Finally, Panavia F and Rely X ARC showed tensile bond strength means without statistically significant difference (p > 0.05), but did not show the same fracture failure mode. The same tensile bond strength results leave the clinicians to assume the same clinical result with both materials. But, they must be careful because the adhesive predominant fracture mode of Rely X ARC cement indicates a premature clinical adhesive failure and, on the other hand, the cohesive predominant fracture
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
mode of Panavia F cement indicates a longer clinical adhesive bond with titanium. [18]
references
[1] Taira M, Moser JB, Greener EH. Studies of Ti alloys for dental castings. Dent Mater 1989;5:45–50. [2] King AWC, Lautenschlager E, Chai J, Gilbert J. A comparison of the hardness of different types of titanium and conventional metal ceramics. J Prosth Dent 1994;72:314–9. [3] Berg E, Davik G, Hegdahl T, Gjerdet NR. Hardness strength and ductility of prefabricated titanium rods used in the manufacture of spark erosion crowns. J Prosth Dent 1996;75:419–25. [4] Wang RR, Fenton A. Titanium for prosthodontic applications: a review of literature. Quintessence Int 1996;27:401–8. [5] Watanabe F, Powers JM, Lorey RF. In vitro bonding of prosthodontic adhesive to dental alloys. J Dent Res 1988;67:479–83. [6] Anderson M, Bergman B, Bessing C, Ericson G, Lundquist P, Nilson H. Clinical results with titanium crowns fabricted with machine duplication and spark erosion. Acta Odont Scand 1989;47:279–86. [7] Blackman R, Barghi N, Tran C. Dimensional changes in casting titanium removable partial denture frameworks. J Prosth Dent 1991;65:309–15. [8] Kononen M, Rintanen J, Waltimo A, Kempainen P. Titanium framework removable partial dentures used for patient allergic to other metals: a clinical report and literature review. J Prosth Dent 1995;73:4–7. [9] Matsumura H, Yoshida K, Tanaka T, Atsuta M. Adhesive bonding of titanium with a titanate coupler and 4-META/MMA-TBB opaque resin. J Dent Res 1990;69:1614–6. [10] Ohno H, Yamane Y, Endo K, Araki Y, Iizuka R. Adhesion of adhesive resin to dental precious metal alloys. Part I. New precious metal alloys with base metals for resin bonding. Dent Mater J 1998;17:285–93. [11] Hansson O. Strength of bond with Comspan opaque to three silicoated alloys and Titanium. Scand J Dent Res 1990;98:248–56. [12] Lautenschlager EP, Monaghan P. Titanium and titanium alloys as dental materials. Int Dent J 1993;43:245–53. [13] Strietzel R, Hosh A, Kalbfleish H, Buch D. In vitro corrosion of titanium. Biomaterials 1998;19:1495–9. [14] Burrow MF, Nikaido T, Satoh M, Tagami J. Early bonding of resin cements to dentin-effect of bonding environment. Operative Dent 1996;21:196–202. [15] Ferrari M, Cagidiaco MD, Breschi R. Microscopic examination of resin bond to enamel and retainer with a phosphate monomer resin. J Prosth Dent 1987;57:298–301. [16] Matsumura H, Nakabayashi N. Adhesive 4-META/MMA-TBB opaque resin with poly(methyl metacrylate)-coated titanium dioxide. J Dent Res 1988;63:29–32. [17] Ohno H, Endo K, Yamane Y, Kawashima I. Adhesion of adhesive resin to dental precious metal alloys. Part II. The relationship between surface structure of Au–In alloys and
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30] [31]
[32]
87
adhesive ability with 4-META resin. Dent Mater J 1998;17:285–93. Taira Y, Imai Y. Primer for bonding resin to metal. Dent Mater 1995;11:2–6. Taira Y, Matsumura H, Yoshida K, Tanaka T, Atsuta M. Adhesive bonding of titanium with a metacrilate-phosphate primer and self-curing adhesives resins. J Oral Rehabil 1995;22:409–12. Taira Y, Matsumura H, Atsuta M. Bonding of titanium with acidic primers and a tri-n-butilborane-initiated luting agent. J Oral Rehabil 1997;24:385–8. Taira Y, Yanagida H, Matsumura H, Yoshida K, Atsuta M, Suzuki S. Adhesive bonding of titanium with a thione-phosphate dual function primer and self-curing luting agents. Eur J Oral Sci 2000;108:456–60. Okhubo C, Watanabe I, Hosoi T, Okabe T. Shear bond strength of PMMA to cast titanium and Co–Cr frameworks using five metal primers. J Prosth Dent 2000;83:50–7. Yanagida H, Matsumura H, Taira Y, Atsuta M, Shimoe S. Adhesive bonding of composite material to cast titanium with varying surface preparations. J Oral Rehabil 2002;29:121–6. Yanagida H, Taira Y, Shimoe S, Atsuta M, Yoneyama T, Matsumura H. Adhesive bonding of titanium–aluminium–niobium alloy with nine surfaces preparations and three self-curing resins. Eur J Oral Sci 2003;111:170–4. Matsumura H, Tanaoue N, Yanagida H, Atsuta M, Koike M, Yoneyama T. Adhesive bonding of super elastic titanium–nickel alloy castings with a phosphate metal conditioner and an acrylic adhesive. J Oral Rehabil 2003;30:653–8. Aboush YEY, Mudassir A, Eldertn RJ. Technical note: resin-to-metal bonds mediated by adhesion promoters. Dent Mater 1991;7:279–80. Cobb DS, Vargas MA, Fridrich TA, d Boushlicher MR. Metal surface treatment: characterization and effect on composite-to-metal bond strength. Operative Dent 2000;25:427–33. Goharian R, Maleknejad F, Salari T, Ghavamnasiri M, Derhami M. Effect of adhesives on bond strength of porcelain veneer to base metal alloy. Quintessence Int 2002;33:595–9. Sahmali S, Demirel F, Saygili G. Comparison of in vitro tensile bond strengths of luting cements to metallic and tooth-colored posts. Int J Periodontics Restorative Dent June 2004;24(3):256–63. Mc Connell RJ. Metal-resin bonding. CDA J 1993;21:38–42. Matinlinna JP, Lassila LV, Kangasniemi I, Yli-Urpo A, Vallittu PK. Shear bond strength of Bis-GMA resin and methacrylated dendrimer resins on silanized titanium substrate. Dent Mater 2005;21(March (3)):287–96. Matinlinna JP, Ozcan M, Lassila LV, Vallittu PK. The effect of a 3-methacryloxypropyltrimethoxysilane and vinyltriisopropoxysilane blend and tris(3-trimethoxysilylpropyl)isocyanurate on the shear bond strength of composite resin to titanium metal. Dent Mater 2004;20(November (9)):804–13.
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Tensile bond strength of dual curing resin-based cements to commercially pure titanium Rafael Schneider a,∗ , Mario Fernando de Goes b , Guilherme Elias Pessanha Henriques c , Daniel C.N. Chan d a
` Lutheran University of Brazil, School of Dentistry, Dental Materials Area, Av. Universitaria, 95560-00 Torres, RS, Brazil Department of Restorative Dentistry, Dental Materials Area, Piracicaba Dental School at Campinas State University, Piracicaba, Brazil c Department of Periodontics and Prosthodontics, Piracicaba Dental School at Campinas State University, Piracicaba, Brazil d Department of Oral Rehabilitation, Division of Operative Dentistry, Medical College of Georgia, Augusta, GA, USA b
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. The aim of this study was to evaluate the tensile bond strength of dual curing
Received 27 July 2005
luting resin cements to commercially pure titanium at 10 min and 24 h after removal of the
Accepted 5 December 2005
oxide layer. Methods. One hundred and twenty titanium discs were obtained by casting and polishing with silicon carbide papers. The titanium discs were sandblasted with 50 m aluminum
Keywords:
oxide, ultrasonic cleaned and bonded in pairs with the resin-based cements Panavia F and
Titanium
Rely X ARC at 10 min and 24 h after the sandblasting. The tensile test was performed with a
Adhesion
crosshead speed of 0.5 mm/min in an Instron Universal testing machine.
Resin-based cement
Results. The Rely X ARC reached the highest tensile strength value at 24 h after sandblast-
Primer
ing (18.27 MPa), but there was no statistically significant difference between the two dual curing resin cements for both times tested. All specimens showed a mixture of cohesive fracture in the resin cement and adhesive failure. However, the predominant failure mode for Panavia F was cohesive in resin cement, and the Rely X ARC exhibited a greater proportion of specimens with adhesive failure between the alloy and resin luting cement at 10 min and 24 h. Significance. Both cements had, statistically, the same tensile bond strength. But in the fracture mode analysis, the adhesive predominant fracture mode of Rely X ARC cement indicates a premature clinical adhesive failure. On the other hand, the cohesive predominant fracture mode of Panavia F indicates a longer clinical adhesive bond with titanium. © 2005 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Titanium is popular in dentistry as a substitute material for the noble and non-noble alloys because of its superior biocompatibility, good mechanical properties and high corrosion resistance [1–4]. Prostheses with titanium have been indicated for a metallic framework in fixed partial denture, full
∗
crowns, prostheses with multiple units, adhesive prostheses, root posts and framework for dentures [5–8]. In spite of these characteristics, the relationship between titanium and the adhesive luting materials remains unclear. In an attempt to reach an effective adhesion between titanium and resin-based cement, a PMMA composite resin with the acidic monomer 4-methacryloyloxymethyl trimellitate anhy-
Correspondence address: R. Luiz Linck Barcelos 220, CEP 94015-590, Gravatai, RS, Brazil. Tel.: +55 51 30424340; fax: +55 51 6262000. E-mail address: [email protected] (R. Schneider). 0109-5641/$ – see front matter © 2005 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2005.12.006
82
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
dride (4-META) was used on the titanium surface and showed stable bond strength [9]. The acidic monomers derived from the carboxylic (4-META) or phosphoric (methacryloyloxydecyl di-hydrogen phosphate) acids are able to bond with the superficial oxide layer by Bolger’s mechanism through electrostatic interaction between the acids of the monomers and the OH− groups of the superficial oxide layer, depending on the isoelectric metallic oxide points and the constant acid dissociation of the monomers [10]. The oxide layer is formed during titanium casting. Due to its great reactivity with elements like O, C, H and N at high temperatures, it produces a firm oxide layer adherent with the metallic substrate. The oxide layer is formed very quickly ˚ (10−9 s), has high chemical stability and is very thin (50–100 A) [11–13]. When it is mechanically removed from the surface, it is formed again just by contact with oxygen. However, the oxide layer formation is very unstable. The exposition period to the high temperatures and to ambient contaminants has greatly affect the layer adherence and thickness. Thus the storage period, time and temperature of casting are determinant factors of the composite resin bond strength to the metal surface before the luting procedures. The use of light-cured resin cements may not be possible because the visible light is blocked by the titanium metallic structure. Besides, the self-curing resin cements have low bond strength in the first hour of the luting procedure. This means that the restoration cannot be submitted under mastication stress before the first hour to avoid its dislodgement. If the restoration moves, the results will be microleakage and second caries [14]. Thus, dual curing resin-based cements must be chosen as the luting material for cement metallic prostheses. Considering the use of titanium as the metallic framework in fixed partial dentures, the bonding capacity of the acidic monomers to the metal oxide layer, the ambient exposition time and the effect of temperature on the formation of titanium oxide layer, the aim of this study was to determine the tensile bond strength of dual curing resin-based cements to commercially pure titanium at 10 min and 24 h after the removal of oxide layer. Additionally, the specimens’ fracture mode was observed after the tensile test.
Table 1 – Luting agents used and their compositions Compositiona
Material
Panavia F
Rely X ARC
a
Paste A: colloidal silica, Bis-GMA, hydrophobic and hydrophilic dimethacrylate, benzoic peroxide Paste B: silanized barium glass, titanium oxide, sodium fluoride, colloidal silica, Bis-GMA, hydrophobic and hydrophilic dimethacrylate, diethanol-p-toluidine, T-isopropylic benzenic sodium sulfinate ED Primer A: hydroxyethylmethacrylate (HEMA), 10-methacryloyloxydecyl di-hydrogen phosphate (MDP), NM-aminosalicilic acid, diethanol-p-toluidine, water ED Primer B: NM-aminosalicilic acid, T-isopropylic benzenic sodium sulfinate, diethanol-p-toluidine, water Alloy Primer: 6-n-4-vinylbenzyl propylamino di-thione triazine (VBATDT), 10-methacryloyloxydecyl di-hydrogen phosphate (MDP) Paste A: Bis-GMA, tri-ethylene glycol dimethacrylate, zircon/silica filler (68 wt%), photoinitiators, amine, pigments Paste B: Bis-GMA, tri-ethylene glycol dimethacrylate, benzoic peroxide, zircon/silica filler (67 wt%) Ceramic Primer: gamma-methacryloxypropyl trimethoxysilane, ethanol, water
Manufacturers’ information.
Products, Rio de Janeiro, RJ, Brazil) and polished with #320, #400 and #600 silicon carbide paper. The cylinders’ edges and acrylic resin around the disc were removed to facilitate the bonding and photocuring procedures. A hole was made in the bottom of cylinders to attach it to the tensile test system. A 4 mm diameter central area of discs’ surfaces was sandblasted with aluminum oxide (50 m) for 5 s at 5 mm from the surface with 80-psi pressure. The specimens were ultrasonically cleaned for 10 min and divided into four groups (30 discs
Table 2 – Manipulation sequence of the luting agents
2.
Materials and methods
The resin-based cements tested in this study are summarized in Table 1. Commercially pure titanium ingots (Rematitan® , Dentaurum, Ispringen, BW, Germany) (Ti > 99%) were used. One hundred and twenty cone trunk-shaped casting wax discs were made using a metallic matrix, fixed in a crucible forming base with wax rods and embedded in phosphate investment (Rematitan® Plus-Dentaurum, Ispringen, BW, Germany). After the wax elimination and investment expansion thermal cycle in an electric furnace (EDGCON 5N-EDG ˜ Paulo, SP, Brazil), the commercially pure titaEquipments, Sao nium ingots were cast in an electric arc casting machine (Rematitan® ) and injected into the investment, following the manufacturer’s instructions. After casting, the discs were separated from the supplying ducts, embedded in a plastic cylinder with self-cured acrylic resin (JET-Classico Dental
Material
Panavia F
Rely X ARC
Manipulation Applying of the Alloy Primer on bonding surface Mixture of one drop of ED Primer A and ED Primer B Applying of the mixture on the bonding surface for 1 min Gently air drying Mixture of equal lengths of Paste A and B for 10 s and applying on bonding surface Excess removal Photocuring for 40 s in each surface Applying the Oxyguard II on the composite resin exposed surface for 3 min and washing Applying the Ceramic Primer on the bonding surface Gently air drying Mixture of equal lengths of Pastes A and B for 10 s and applying on bonding surface Excess removal Photocuring for 20 s in each surface
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
83
per each group). Fingernail varnish was applied on the nonsandblasted surfaces to avoid any interference in the bond strength. The bonding procedures were performed immediately in the Groups 1 and 3 and 24 h after the oxide-layer removal in the Groups 2 and 4. Panavia F (Kuraray, Medical Inc., Tokyo, Japan) was used to bond the specimens in the Groups 1 and 2 and Rely X ARC (3 M Dental Products, Saint Paul, MN, USA) in the Groups 3 and 4 following the manufacturer’s instructions (Table 2). In the Groups 1 and 2, a primer containing MDP and VBATDT monomers (Alloy Primer, Kuraray Medical Inc., Tokyo, Japan) was applied on the metal surface to bond composite resin. In Groups 3 and 4, a silane-coupling agent (Ceramic Primer-3 M Dental Products, Saint Paul, MN, USA) was applied on the metal bond surface.
Fig. 2 – Tensile test device. (A) Universal test machine crosshead. (B) Specimen held by the tensile system bar at the tensile test device and (C) tensile system bars.
Fig. 1 – Specimens bonded in pairs: (A) plastic cylinder, (B) hole to attach the tensile system bar, (C) titanium discs and (D) resin-based luting agent.
Thereafter, equal lengths of composite resins base paste and catalyst paste were mixed for 10 s. The mixture was applied on one of the isolated sandblasting surfaces and the titanium discs were bonded in pairs with finger pressure. A cylindrical plastic matrix was used to keep the discs centered and aligned in the correct place. Then, the excesses were removed with a paint brush and the composite resins were photocured. Panavia F was photocured for 20 s in each of the four different positions and Rely X ARC 40 s in each position. On the external surface of Panavia F composite resin, the Oxyguard II alcoholic gel (Kuraray Medical Inc., Tokyo, Japan) was applied to permit the curing of this surface. After the curing, the specimens (Fig. 1) were stored in distilled water at 37 ◦ C for 24 h. Hence, the tensile test was performed in a Universal Testing machine (Instron Co., Canton, MA, USA) at 0.5 mm/s crosshead speed (Fig. 2). The tensile test values were recorded in MPa and submitted to two-way ANOVA (˛ = 0.05). To determine the fracture mode, the fractured specimens in the tensile test were observed in a 60× stereoscopic loupe (Carls-Zeiss, Oberkochen, BA, Germany). Thereafter, the discs were removed from the acrylic resin, ultrasonically cleaned, gold sputtered and observed by a scanning electron microscope (Leo 435 VP, Orsay Phisics, Fuveau, ZA, France).
84
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
Table 3 – Tensile bond strength values (means and standard deviations) between the tested groups Groups
Tensile bond strength meansa
Rely X ARC—24 h (Group 4) Panavia F—24 h (Group 2) Panavia F—10 min (Group 1) Rely X ARC—10 min (Group 3)
18.27 (7.34) 14.76 (7.71) 13.96 (3.60) 13.90 (6.08)
Values in parentheses are standard deviations. a
Values without statistically significant difference in two-way ANOVA and F-test (p > 0.05).
3.
in the Groups 1 and 2, probably because of the Panavia F consistency.
Results
The tensile bond strength values were transformed (x + 0 square root) prior to statistical analysis. The means and standard deviations are listed in Table 3. Higher mean was obtained in the Group 4 (Rely X ARC—24 h), followed by Group 2 (Panavia F—24 h), Group 1 (Panavia F—10 min) and Group 3 (Rely X ARC—10 min). The bond strengths ranged from 5.15 to 30.23 MPa for Panavia F composite resin and from 3.9 to 29.34 MPa for the Rely X ARC. Table 4 shows the results analyzed according to two-way ANOVA. The bond strength was neither influenced by storage time nor luting material and there was no statistically significant difference among the bond strengths of the tested groups. Optical and SEM observations revealed that all specimens showed a mixture of adhesive and cohesive failures on the bond surface regardless of the luting material or the storage time (10 min or 24 h after the oxide removal). Hence, a composite resin cohesive failure in some areas as well as total debonding between titanium and composite resins (adhesive failure), characterizing the mixed failure, was observed. However, in the Groups 1 and 2, where Panavia F composite resin was used 10 min and 24 h after sandblasting, respectively, there was the predominance of the composite resin cohesive failure in more than 50% bonding area in all specimens (Figs. 3 and 4). In Groups 3 and 4, where Rely X ARC composite resin was used 10 min and 24 h after sandblasting, respectively, the predominant failure was adhesive between the titanium and the composite resin in more than 50% bonding area in all specimens (Figs. 5 and 6). It was observed that air bubbles were incorporated in the composite resins in all groups, which would have been caused by the composite resins manipulation. Nevertheless, there was a major bubble entrapping
4.
Discussion
The Rely X ARC is a Bis-GMA based composite resin and was developed to bond with all indirect restorative materials (polymers, metals or ceramics). In this study, the Rely X ARC composite resin showed a mean bond strength of 13.9 MPa, 10 min after sandblasting and 18.27 MPa, 24 h after sandblasting. These values were not statistically significantly different, which means the oxide layer removal and neo-formation did not have any effect on the composite resins’ tensile bond strength to titanium, although the higher numeric difference was seen at 24 h period. The self-limiting characteristic of superficial oxide layer formation on titanium surface might have been responsible for the results of statistical equality among the tested groups. Actually, during the oxide layer formation and lengthening, it becomes a protector barrier itself, avoiding the oxygen contact with the underneath metal, interrupting its formation. This phenomenon occurred in the same way with Panavia F composite resin, which showed mean values of 13.96 MPa, 10 min after sandblasting and 14.76 MPa, 24 h after the sandblasting, which are not statistically significantly different. The utilization of acidic resin monomers, like MDP contained in the Panavia F Alloy Primer, would be able to produce an effective and lasting bond between composite resins and basic metals [15–25]. This bond would occur throughout chemical links between the monomer phosphate radicals and the basic metal oxide layer. Ohno et al. [10,17] described the Bolger’s mechanism which would be responsible for this bond. In this mechanism, an electrostatic interaction between polymer acids or bases and hydroxyl groups of the metal surface would occur, depending on the isoelectric point of metal oxides and acid dissociation constants of acidic adhesive monomers. Although the efficacy of an acid monomer, like MDP, has been considered by many authors, the mean tensile bond strength reached in this study by Panavia F composite resin that contains MDP in its composition, did not show statistically significant difference with Rely X ARC composite resin, which the manufacturer recommends only for the application of a silane agent. Silicon, the principal component of silane agents, has a great affinity for metal ions of the superficial oxide layer of titanium, producing a chemical bond like Panavia F Alloy Primer [26,27].
Table 4 – Results of analysis of variance for tensile bond strength Source of variation
d.f.
Material Time Mat/time Residual
1 1 1 56
0.5633769 1.1581963 1.1048372 40.5206772
Total
59
43.3470877
General mean = 3.808355, variation ratio = 22.336%.
Sum of squares
Mean square 0.5633769 1.1581963 1.1048372 0.7235835
F-value
p-Value
0.7786 1.6006 1.5269
0.61484 0.20852 0.21948
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
85
Fig. 3 – SEM of fracture mode of Groups 1 and 2 specimens, showing the mixed mode of fracture with predominance of composite resin cohesive failure. (A) Titanium surface with partially bonded composite resin, (B) higher magnification SEM of the marked area in (A): A, composite resin cohesive failure area; B, adhesive failure area; C, pore.
The numerical match between Rely X ARC and Panavia F could occur because all specimens were sandblasted, creating an equal surface pattern in each specimen. Thus, the micromechanical retention became the principal bonding to titanium for both cements, which could explain the similar results of the tested groups [28]. Other sandblasting influence is in the phenomenon of alumina particles incrustation during the sandblasting procedure. These particles stay stuck on the titanium surface because of the velocity and the pressure that they hit it with and cannot be removed even by ultrasonic cleaning or acid etching. So, these non-removable
alumina particles become responsible for the chemical bond of the Alloy Primer and silane agents to themselves, increasing the bond strength of both cements, the Panavia F and Rely X ARC, decreasing the numerical difference between them and resulting in similar results [27,28]. A fact, which could be an influencing factor in the results, was the incorporation of lot of bubbles in the Panavia F manipulation. This composite resin is a little putty, which could contribute to the incorporation of maximum bubbles. Rely X, on the other hand, is a fluid which could minimize this phenomenon.
Fig. 4 – SEM of fracture mode of Groups 1 and 2 specimens, showing the mixed mode of fracture with predominance of composite resin cohesive failure. (A) Opposite disc surface showed in Fig. 3A. (B) Higher magnification SEM of marked area in (A): A, composite resin cohesive failure area; B, adhesive failure area; C, pore; D, sample preparation artifact.
86
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
Fig. 5 – SEM of fracture mode of Groups 3 and 4 specimens, showing the mixed mode of fracture with predominance of adhesive failure. (A) Titanium surface with partially bonded composite resin, (B) higher magnification SEM of marked area in (A): A, composite resin cohesive failure area; B, adhesive failure area; C, pore.
Fig. 6 – SEM of fracture mode of Groups 3 and 4 specimens, showing the mixed mode of fracture with predominance of adhesive failure. (A) Opposite disc surface showed in Fig. 5A. (B) Higher magnification SEM of marked area in (A): A, composite resin cohesive failure area; B, adhesive failure area; C, pore.
Though both composite resins had shown similar tensile bond strengths to titanium, the fracture shape on all Panavia F specimens (Groups 1 and 2) after the tensile test was always mixed, regardless of the storage time (10 min or 24 h after sandblasting). However, there was predominance of composite resin cohesive failure on more than 50% bonding surface area (Figs. 3 and 4). In all Rely X ARC specimens as well, the mixed failure occurred, but there was predominance of the adhesive failure on more than 50% bonding surface area (Figs. 5 and 6). This shape fracture difference seems to be linked to the most wetting ability of Panavia F, due to MDP and VBATDT, resulting in the best contact area between the Panavia F and titanium surface. Another reason for the Panavia F cohesive predominant fracture pattern is its
high adhesion capacity that exceeds its own cohesive strength [29–32].
5.
Conclusion
Finally, Panavia F and Rely X ARC showed tensile bond strength means without statistically significant difference (p > 0.05), but did not show the same fracture failure mode. The same tensile bond strength results leave the clinicians to assume the same clinical result with both materials. But, they must be careful because the adhesive predominant fracture mode of Rely X ARC cement indicates a premature clinical adhesive failure and, on the other hand, the cohesive predominant fracture
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 81–87
mode of Panavia F cement indicates a longer clinical adhesive bond with titanium. [18]
references
[1] Taira M, Moser JB, Greener EH. Studies of Ti alloys for dental castings. Dent Mater 1989;5:45–50. [2] King AWC, Lautenschlager E, Chai J, Gilbert J. A comparison of the hardness of different types of titanium and conventional metal ceramics. J Prosth Dent 1994;72:314–9. [3] Berg E, Davik G, Hegdahl T, Gjerdet NR. Hardness strength and ductility of prefabricated titanium rods used in the manufacture of spark erosion crowns. J Prosth Dent 1996;75:419–25. [4] Wang RR, Fenton A. Titanium for prosthodontic applications: a review of literature. Quintessence Int 1996;27:401–8. [5] Watanabe F, Powers JM, Lorey RF. In vitro bonding of prosthodontic adhesive to dental alloys. J Dent Res 1988;67:479–83. [6] Anderson M, Bergman B, Bessing C, Ericson G, Lundquist P, Nilson H. Clinical results with titanium crowns fabricted with machine duplication and spark erosion. Acta Odont Scand 1989;47:279–86. [7] Blackman R, Barghi N, Tran C. Dimensional changes in casting titanium removable partial denture frameworks. J Prosth Dent 1991;65:309–15. [8] Kononen M, Rintanen J, Waltimo A, Kempainen P. Titanium framework removable partial dentures used for patient allergic to other metals: a clinical report and literature review. J Prosth Dent 1995;73:4–7. [9] Matsumura H, Yoshida K, Tanaka T, Atsuta M. Adhesive bonding of titanium with a titanate coupler and 4-META/MMA-TBB opaque resin. J Dent Res 1990;69:1614–6. [10] Ohno H, Yamane Y, Endo K, Araki Y, Iizuka R. Adhesion of adhesive resin to dental precious metal alloys. Part I. New precious metal alloys with base metals for resin bonding. Dent Mater J 1998;17:285–93. [11] Hansson O. Strength of bond with Comspan opaque to three silicoated alloys and Titanium. Scand J Dent Res 1990;98:248–56. [12] Lautenschlager EP, Monaghan P. Titanium and titanium alloys as dental materials. Int Dent J 1993;43:245–53. [13] Strietzel R, Hosh A, Kalbfleish H, Buch D. In vitro corrosion of titanium. Biomaterials 1998;19:1495–9. [14] Burrow MF, Nikaido T, Satoh M, Tagami J. Early bonding of resin cements to dentin-effect of bonding environment. Operative Dent 1996;21:196–202. [15] Ferrari M, Cagidiaco MD, Breschi R. Microscopic examination of resin bond to enamel and retainer with a phosphate monomer resin. J Prosth Dent 1987;57:298–301. [16] Matsumura H, Nakabayashi N. Adhesive 4-META/MMA-TBB opaque resin with poly(methyl metacrylate)-coated titanium dioxide. J Dent Res 1988;63:29–32. [17] Ohno H, Endo K, Yamane Y, Kawashima I. Adhesion of adhesive resin to dental precious metal alloys. Part II. The relationship between surface structure of Au–In alloys and
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30] [31]
[32]
87
adhesive ability with 4-META resin. Dent Mater J 1998;17:285–93. Taira Y, Imai Y. Primer for bonding resin to metal. Dent Mater 1995;11:2–6. Taira Y, Matsumura H, Yoshida K, Tanaka T, Atsuta M. Adhesive bonding of titanium with a metacrilate-phosphate primer and self-curing adhesives resins. J Oral Rehabil 1995;22:409–12. Taira Y, Matsumura H, Atsuta M. Bonding of titanium with acidic primers and a tri-n-butilborane-initiated luting agent. J Oral Rehabil 1997;24:385–8. Taira Y, Yanagida H, Matsumura H, Yoshida K, Atsuta M, Suzuki S. Adhesive bonding of titanium with a thione-phosphate dual function primer and self-curing luting agents. Eur J Oral Sci 2000;108:456–60. Okhubo C, Watanabe I, Hosoi T, Okabe T. Shear bond strength of PMMA to cast titanium and Co–Cr frameworks using five metal primers. J Prosth Dent 2000;83:50–7. Yanagida H, Matsumura H, Taira Y, Atsuta M, Shimoe S. Adhesive bonding of composite material to cast titanium with varying surface preparations. J Oral Rehabil 2002;29:121–6. Yanagida H, Taira Y, Shimoe S, Atsuta M, Yoneyama T, Matsumura H. Adhesive bonding of titanium–aluminium–niobium alloy with nine surfaces preparations and three self-curing resins. Eur J Oral Sci 2003;111:170–4. Matsumura H, Tanaoue N, Yanagida H, Atsuta M, Koike M, Yoneyama T. Adhesive bonding of super elastic titanium–nickel alloy castings with a phosphate metal conditioner and an acrylic adhesive. J Oral Rehabil 2003;30:653–8. Aboush YEY, Mudassir A, Eldertn RJ. Technical note: resin-to-metal bonds mediated by adhesion promoters. Dent Mater 1991;7:279–80. Cobb DS, Vargas MA, Fridrich TA, d Boushlicher MR. Metal surface treatment: characterization and effect on composite-to-metal bond strength. Operative Dent 2000;25:427–33. Goharian R, Maleknejad F, Salari T, Ghavamnasiri M, Derhami M. Effect of adhesives on bond strength of porcelain veneer to base metal alloy. Quintessence Int 2002;33:595–9. Sahmali S, Demirel F, Saygili G. Comparison of in vitro tensile bond strengths of luting cements to metallic and tooth-colored posts. Int J Periodontics Restorative Dent June 2004;24(3):256–63. Mc Connell RJ. Metal-resin bonding. CDA J 1993;21:38–42. Matinlinna JP, Lassila LV, Kangasniemi I, Yli-Urpo A, Vallittu PK. Shear bond strength of Bis-GMA resin and methacrylated dendrimer resins on silanized titanium substrate. Dent Mater 2005;21(March (3)):287–96. Matinlinna JP, Ozcan M, Lassila LV, Vallittu PK. The effect of a 3-methacryloxypropyltrimethoxysilane and vinyltriisopropoxysilane blend and tris(3-trimethoxysilylpropyl)isocyanurate on the shear bond strength of composite resin to titanium metal. Dent Mater 2004;20(November (9)):804–13.