- Email: [email protected]
Effect of surface roughness and thermal cycling on bond strength of C.P. titanium and Ti–6Al–4V alloy to ceramic
Available online at www.sciencedirect.com
Journal of Prosthodontic Research 56 (2012) 204–209 www.elsevier.com/locate/jpor
Original article
Effect of surface roughness and thermal cycling on bond strength of C.P. titanium and Ti–6Al–4V alloy to ceramic Cherif Adel Mohsen B.D.S., M.D.Sc., D.D.Sc.* Faculty of Dentistry, University of Minia, Minia, Egypt Received 13 January 2011; received in revised form 10 September 2011; accepted 15 October 2011 Available online 10 May 2012
Abstract Purpose: Studying the effect of surface roughness and thermal cycling on titanium–ceramic bonding. Methods: One hundred fourteen samples in the form of bar for the C.P. titanium and Ti–6Al–4V alloy were used. They were divided into two groups according to the type of bar. Each group was then subdivided according to the type of surface treatment to three subgroups, control, airborne-particle abrasion and silica coated. Each subgroup was subdivided into two classes according to the type of test (surface roughness and bond strength). Samples used for the bond strength test were veneered. These samples were subdivided into two subclasses according to thermal cycling; whether without thermal cycling or after 6000 thermal cycles. Results: The surface roughness test results showed that silica coating recorded the highest surface roughness. Also C.P. titanium gave higher value of surface roughness than Ti–6Al–4V alloy. As regard the bond strength, the airborne-particle abrasion classes and the silica coated classes recorded bond strength values above the acceptable limit of 25 MPa determined in ISO 9693. As regard thermal cycling, the results showed that aging by thermal cycling decreased the metal–ceramic bond strength. Conclusions: The airborne-particle abrasion and the silica coating are acceptable treatments for titanium–ceramic restorations. Increasing surface roughness of C.P. titanium and Ti–6Al–4V alloy not necessarily results in an increase in their bond strength to ceramics. Aging affects the metal– ceramic bond strength. # 2011 Japan Prosthodontic Society. Published by Elsevier Ireland. All rights reserved. Keywords: Titanium; Bond strength; Surface roughness; Thermal cycling; Surface treatment
1. Introduction The formation of a metal–ceramic bond at porcelain firing temperatures should be strong enough to resist both transient and residual thermal stresses and mechanical forces in function. The requirements which are to provide such a bond are discussed as chemical, thermal, and mechanical [1,2]. Surface roughness of the titanium can also drastically affect the titanium–porcelain adhesion. Wagner et al. [3] found a direct correlation between surface roughness and bond strength; the greater the roughness, the higher the bond strength. The porous surface would be in favor of mechanical interlocking between the oxide layer and the porcelain [1]. On the contrary, Reyes et al. [4] concluded that increasing surface roughness does not
* Corresponding author at: Fixed Prosthodontics Department, Faculty of Dentistry, Minia University, 4 El Tharrir St., Dokki, Guiza, Egypt. Fax: +20 0233372078. E-mail address: [email protected].
necessarily result in an increase in bond strength. Kelly et al. [5] reported that surface roughness may increase the stress concentration at the metal–ceramic interface, and generate steep re-entrant angles that may prevent complete wetting and result in voids at metal–ceramic interfaces. A comparison of surface roughness between C.P. Ti and Ti– 6Al–4V specimens was carried out by Guilherme et al. They revealed C.P. Ti to be significantly rougher than Ti–6Al–4V alloy [6]. Bienias et al. [7] reported that the surface roughness Ra after sandblasting only reached 0.69 0.02 mm for C.P. titanium and 0.55 0.02 mm for Ti–6A1–4V alloy. The success of the porcelain-fused-to-alloy restorations depends acutely on the success of the strong bonding between porcelain and the metal substructure [8,9]. Bonding porcelain to titanium for porcelain fused to metal (PFM) crowns is still an outstanding problem for the current use of metal–ceramic restorations. The highly oxidative nature of the titanium surface has been postulated as the cause of the weak bond between porcelain and titanium [10,11]. Many surface treatments of
1883-1958/$ – see front matter # 2011 Japan Prosthodontic Society. Published by Elsevier Ireland. All rights reserved. doi:10.1016/j.jpor.2011.10.001
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209
titanium have been proposed to overcome this problem such as sandblasting, acid treatment, coating using silicon nitride or SiO2 or SiO2–TiO2, bonding agents and laser etching [4,7,12– 19]. These studies investigated appropriate treatments to achieve a better bonding between porcelain and titanium as quantified by the resultant bond strength [20–22]. Some authors have shown that Ti–ceramic bonding was acceptable [21,22] but variations exist between different titanium and ceramic systems [23]. The purpose of this present study is (i) to study the effect of C.P. titanium and Ti–6Al–4V alloy surface treatment with airborne-particle abrasion or silica coating on the surface roughness and on the titanium–ceramic bond strength. (ii) To investigate the relation between surface roughness of and bond strength. (iii) To study the effect of thermal cycling on Ti– ceramic bond strength.
205
at 2.8-bar pressure from a distance of 10 mm for 20 s. All specimens were subjected to ultrasonic cleaning using distilled water for 10 min and then rinsed in distilled water. Specimens were left to dry before the application of porcelain. 2.4. Ceramic application (for the bond strength test samples) Next, in the middle of the specimens (8 mm 3 mm 1 mm) Ceramics2in1 (Biodenta, Berneck, Switzerland) lowfusing dental porcelain was fired on C.P. Ti and Ti–6Al–4V alloy according to the manufacturer’s recommendations, as listed in Table 1A custom-made jig was used to ensure the precision of porcelain thickness. 3. Testing procedure
2. Materials and methods 3.1. Surface roughness 2.1. Samples construction One Hundred and fourteen samples of C.P. Ti (grade 2) and Ti alloy Ti–6Al–4V (Titanium Int., USA) were used in this study. The samples were cut using an electric saw to form a bar with the following dimension (25.0 mm 3.0 mm 0.5 mm) as recommended with the ISO 9693 [24]. The samples were then ground on all sides with 600 grit SiC paper under continuous water cooling. Finally the samples were ultrasonically rinsed in deionized water for 30 min. 2.2. Grouping The samples were divided into two groups (57 samples each) according to the type of bar (C.P. Ti or Ti–6Al–4V alloy). Each group was then subdivided according to the type of surface treatment to three subgroups (19 samples each). The first subgroup was not subjected to any surface treatment and served as a control. The second subgroup was subjected to airborneparticle abrasion, while the third subgroup was silica coated. Each subgroup was further subdivided into two classes according to the type of test (five samples for the surface roughness test and 14 samples for the bond strength test). Then the bond strength classes were subdivided into two subclasses according to thermal cycling; whether without thermal cycling or after 6000 thermal cycles. 2.3. Surface treatment - Airborne-particle abrasion: It was accomplished using 250 mm aluminum oxide particles at 2–3 bars air pressure from a distance of 10 mm for 30 s [17,25]. - Silica coating: The samples were first blasted with 250 aluminum oxide particles at 2–3 bars air pressure from a distance of 10 mm for 30 s.The samples were then ultrasonically cleaned for 10 min in a water bath, then dried. Finally, the samples were blasted with a mixture of 50 mm Al2O3 and 50 mm silica particles (Rolloblast 50 mm, Renfert, Germany)
The surface roughness (Ra) was measured in microns using a mechanical profiler (Form Talysurf i60, Taylor Hobson Precision, England). The surface profile included the variation on the surface topography which filtered to obtain surface roughness profile only. The filtration was done using Gaussian filter with cut-off wavelength = 0.25 mm and measuring length = 5 mm. 3.2. Thermal cycling Thermal cycling was performed with bath temperatures of 5 2 8C and 55 2 8C, respectively. The dwell time in each bath was 20 s, the transfer time between the two baths was 10 s, resulting in a frequency of 1 cycle/min. Each sample was subjected to 6000 thermal cycles to simulate a 5-year clinical service [22,26]. 3.3. Bond strength The bond strength testing was performed with a 3-point bending test on universal testing machine (LLOYD Instruments, UK). The specimens were positioned on a specially fabricated metal support with the ceramic positioned on the side opposite that contacted by the applied force. The distance between the supports was 20 mm. A compressive load was applied at the midpoint of the metal strip with a rounded-tip loading rod at a crosshead speed of 0.5 mm/min until a sudden drop in load occurred in the load-deflection curve, indicating the bond failure. The failure load was recorded and calculated by the following equation given in ISO 9693 [24] s ¼kF where F is the maximum force applied in Newton before debonding (failure load), k is a constant determined from a graph in ISO 9693 with units of mm2.The value of k depends
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209
206
Table 1 Firing schedules of Ceramics2in1 veneering material. Ceramics2in1 layers
Standby t (8C)
Closing time (min)
Temp. (8C rise/min)
1st Firing temp. (8C)
2nd Firing temp. (8C)
Holding time
Vacuum start (8C)
Vacuum end (8C)
Special
Opaquer Dentin Glaze
400 400 400
6 4 4
55 45 45
800 740 710
800 730 –
2 1 1
450 450 450
799 739/729 709
– – Slower cooling approx 6 min
Table 2 The surface roughness of the tested groups (mm). Upper: mean, lower: SD. Groups
Control
Airborne-particle abrasion
Silica coating
C.P. Ti
0.25 (0.03)
1.49 (0.11)
1.76 (0.08)
Ti–6Al–4V
0.09 (0.02)
1.02 (0.07)
1.25 (0.08)
Fig. 1. Comparison between the surface roughnesses of the tested groups (mm).
Table 3 The bond strength of the tested groups (MPa). Upper: mean, lower: SD.
on the thickness and the elastic modulus of the tested titanium samples. All data obtained in this research were calculated, tabulated and statistically analyzed. One-way ANOVA and Tukey tests were used for the surface roughness data while a Two-way ANOVA and Fisher’s LSD tests was used for the bond strength data.
Groups
Thermal cycling
Control
Airborne-particle abrasion
Silica coating
C.P. Ti
Without
12.49 (2.70) 9.37 (0.56)
29.40 (3.73) 25.10 (5.52)
33.20 (5.40) 26.45 (5.89)
13.34 (2.13) 10.02 (1.11)
33.14 (6.04) 27.56 (4.39)
37.80 (5.39) 30.79 (7.12)
4. Results The means and standard deviations of the surface roughness for the two tested groups are presented in Fig. 1 and Table 2. A one-way ANOVA test was used to determine significant differences between the tested groups, subgroups and classes ( p < 0.05). The Tukey test for multiple comparisons of means at ( p < 0.05) was done following the one-way analysis of variance. The results showed a statistical difference between the treated surfaces and the control subgroups (no surface treatments). The results showed that there was a statistical difference between the two tested surface treatments. Silica coating recorded the highest surface roughness 1.76 and 1.25 mm for C.P. Ti and Ti–6Al–4V alloy, respectively. While airborne-particle abrasion recorded: 1.49 and 1.02 mm for C.P. Ti and Ti–6Al–4V alloy, respectively. The results also showed that there was a significant difference between cp Ti and Ti– 6Al–4V alloy for the same surface treatment as well as for the control classes: C.P. Ti gave higher value of surface roughness than the Ti–6Al–4V alloy. The means and standard deviations of the bond strength for the two tested groups are presented in Fig. 2 and Table 3. A two way ANOVA Test was used to determine significant differences between the tested groups, subgroups and subclasses ( p < 0.05). Fisher’s LCD method for multiple comparisons of means at ( p < 0.05) was done following the two-way analysis of variance. The results showed a statistical difference
After Ti–6Al–4V
Without After
between the treated surfaces and the control subgroups (no surface treatments). The results showed that for both groups tested, there was a statistical difference between the airborneparticle abrasion subclasses and the silica coated subclasses. At the same time, the results showed that the airborne-particle abrasion subclasses and the silica coated subclasses recorded bond strength values above the acceptable limit of 25 MPa determined in ISO 9693 [24] for the bond strength between titanium or its alloys and different low-fusing dental porcelains. For the same type of surface treatment tested, there was a significant difference between the C.P. Ti and Ti–6Al–4V alloy which recorded higher bond strength values. Thermal cycling resulted in a significant decrease in bond strength for all the tested subclasses. As regard the mode of failure, the C.P. Ti and Ti–6Al–4V alloy surface after debonding demonstrated a combination of cohesive and adhesive bond failures, i.e. traces of porcelain were observed on specimens that were treated with the of airborne-particle abrasion and that silica coated. 5. Discussion In general, the roughness of the surface can be connected to the enhancement of the bond strength of the metal–ceramic
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209
207
Bond strength (MPa) 29.4 12.49
9.37
without aging
aer aging
Control
25.1
without aging
aer aging
Airborne-parcle abrasion
33.2
without aging
26.45
aer aging
Silica coang
c.p.Ti
13.34
10.02
without aging
aer aging
Control
33.14
27.56
without aging
aer aging
Airborne-parcle abrasion
37.8
without aging
30.79
aer aging
Silica coang
Ti-6Al-4V
Fig. 2. Comparison between the bond strength of the tested groups (MPa).
system, which significantly influences the achievement of sufficient bonding, observed in previous studies [27,28]. Roughness can create an increase in the surface area of a metal substrate for porcelain and enhance the metal–ceramic bond [7]. The results of the surface roughness measurements in this study showed that for the same tested group (C.P. Ti/Ti– 6Al–4V), a statistical difference between the treated surfaces and the control subgroups (no surface treatments), also there was a statistical significant difference between the silica coated classes and the airborne-particle abrasion. The silica coated classes gave the highest surface roughness. This may be due to the fact that silica particles may cause deeper penetration on the surface of the C.P. Ti and Ti–6Al–4V alloy and also some of these particles may be leached out during cleaning, leaving a deeper micro-pores. These results contradict those of Bienias et al. [7] who reported that sandblasting recorded higher surface treatment than silica coating. This difference may be due to the method of silica coating used. In the present study, silica coating was performed by the tribochemical method while Bienias et al. used a sol–gel coating. The results showed that by comparing C.P. Ti and Ti–6Al–4V subjected to the same surface treatment, C.P. Ti recorded higher surface roughness than Ti–6Al–4V.These results are in accordance with Guilherme et al. [6] and Bienias et al. [7] who reported the same findings. Clinical failures of metal–ceramic restorations are often due to fractures or chipping of the veneering ceramic. A crucial point is the metal–ceramic bond strength [29]. Obtaining a good quality bond strength of metal–ceramic system depends on the properties of both the metal substrate and dental porcelain, their interactions in the capabilities of chemical and mechanical bonding formation [27]. Compared to the bonding in typical metal–porcelain systems, the adhesion between titanium and porcelain is relatively poor [14] attributed to the forming of a thick titanium layer, known as the ‘‘a-case’’ on its surface [21]. This oxide layer forms during the casting of the titanium and the high temperature sintering of the porcelain [13]. Low fusing ceramics were developed to improve the adhesion of titanium to porcelain. It is claimed that at the firing temperature of the low fusing titanium porcelain (720–750 8C) a dissociation of the superficial native titanium oxides takes place, followed by dissolution of elements within the titanium mass, accompanied by diffusion of the ceramic material elements [19]. According to Adachi et al. [10] at the firing temperature of low fusing porcelain (750 8C) an oxide layer of about 32 nm thickness was
created, whereas at higher temperatures the adherence of oxides was reduced. There are several mechanical tests that have been used to determine the debonding strength between metal and porcelain including a plethora of flexural and shear designs [30]. Generally the bond strength of metal–ceramic systems is evaluated with the three point bending according to ISO 9693:1999 [24] or with the shear testing. The latter although reliable, cannot be directly compared with the three point bending results. In some cases only the fracture mode has been used as a classification criterion [31] based on the former version of ISO 9693 standard, where metal–ceramic strips were bended over a rod to a 908 angle of the specimens ends. The specimens were then flattened and the percentage of adherent porcelain was determined along the predominant part of the middle third of the metallic substrate. However, it is obvious that such a test cannot provide qualitative results. The latest revision of ISO 9693 in 1999 based on the full adoption of the DIN 13927 [32] also commonly known as Schwickerath test [33]. According to this specification the metal–ceramic system comply with the requirements pass ISO 9693 when four out of six specimens have a debonding strength higher than 25 MPa [24]. All samples were subjected to thermal cycling: 6000 cycles to simulate a 5-year clinical service [22,26]. The results showed that the control classes (no surface treatment) recorded the least bond strength for both C.P. Ti (12.49 Mpa) and Ti–6Al–4V (13.34 Mpa) respectively. These results are in accordance with other investigators [4,14,16,17]. These results may be attributed to the fact that unmodified titanium surface, that is, a surface not receiving any surface treatment, produces a weak, non protective and non adherent oxide layer that is unsuitable for porcelain bonding [11]. The results showed that, the control classes (no surface treatment) recorded bond strength values below the acceptable limit of 25 MPa determined in ISO 9693, indicating the importance of using surface treatment for C.P. Ti and Ti–6Al–4V. The mode of failure for control classes showed mainly adhesive fracture at Ti–ceramic interface due to separation of the ceramic material from the substrate. This mode of failure may be attributed to the poor bond strength between the Ti and the ceramic for these subgroups. Poor bond strength may be attributed to the low surface roughness of these subgroups. Several investigators reported that surface treatment using airborne-particle abrasion enhanced the bond strength of commercially pure titanium or Ti alloys to low-fusing porcelain
208
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209
[7,16,17]. In this study, airborne-particle abrasion surface treatment was performed using 250 mm, as Derand and Hero [34] observed that the use of 250 mm alumina particles, compared with 50 mm particles, significantly improved the bond between titanium and ceramic. The bond strength recorded for the airborne-particle abrasion were above the acceptable limit of 25 MPa determined in ISO 9693, indicating that this type of surface treatment may be useful for improving the bond between Ti and ceramic due to the increase of surface area available for mechanical interlocking. At the same time the mode of failure of these subclasses was a combination of cohesive and adhesive bond failures, i.e. traces of porcelain were observed on specimens which may be due to the relatively high bond strength recorded between Ti and ceramic in those subclasses. This performance may be attributed to the fact that airborne-particle abrasion likely improves the bond strength by removing loosely attached furrows, overlaps, and flakes of metal created by grinding procedures, provides mechanical interlocking, increases surface area, and increases wettability [4]. On the other side, results obtained in this research contradict that of Gilbert et al. [14] who reported that airborneparticle abrasion could contaminate the surface of titanium with alumina particles, which could weaken the mechanical interlocking of the porcelain and titanium [35]. The results showed that there was a statistical significant difference between the airborne-particle abrasion subclasses and the silica coated subclasses for C.P. Ti and Ti–6Al–4V. Bienias et al. [7] reported that application of SiO2 as an intermediate coating, produced by the sol–gel method, to both C.P. Ti and Ti– 6Al–4Valloy significantly improves the bond strength of metal– porcelain systems in comparison to the metal substrate only after sandblasting, and may have clinical use. In this present study, the metal was treated by SiO2 in a different way than with Bienias et al. [7] as it was blasted. The silica coated subclasses, recorded higher bond strength value than the airborne-particle abrasion subclasses. This may be due to the fact that blasting the metal with Al2O3 + silica combines micromechanical retention produced by airborne-particle abrasion and chemical bonding resulting from silica coating of the Ti surface. Investigators supposed that during the firing of the ceramic layer, there occurs diffusion of the components and possibly a chemical reaction at the interface between the metal substrate-intermediate coatings– porcelain [7]. Investigators draw the attention to the importance of obtaining chemical bonding to enhance the bond strength between metal substrate and porcelain [15]. According to Fischer [27], the contribution of chemical bonds to metal–ceramic bond strength is much more important than mechanical interlocking, obtained by roughening the surface. The mode of failure of these classes was a combination of cohesive and adhesive bond failures, i.e. traces of porcelain were observed on specimens which may be due to the relatively high bond strength recorded between Ti and ceramic in those subclasses as for the airborneparticle abrasion classes. In this present research, the results showed that irrespective to the type of surface treatment tested, there was significant difference between the C.P. Ti and Ti–6Al–4V alloy, the latter recorded higher bond strength value. According to Adachi et al.
[10] the thicknesses of the oxide layers on C.P. Ti and Ti–6Al– 4V alloy heated to 750 8C were 32 and 11 nm, respectively, but at 1000 8C the thickness was approximately 1 mm for C.P. Ti and Ti–6Al–4V alloy. The thicker the interfacial oxide layer becomes, the weaker the bond strength of metal–porcelain [10]. These results are in accordance with of Bienias et al. [7] who reported that bond strength between ceramic and sandblasted C.P. Ti 23.04 MPa, while it was 28.24 MPa for Ti–6Al–4V. While these results contradicted that of Suansuwan and Swain [15] who reported a significant difference between C.P. Ti and Ti alloy which recorded lesser bond strength value. Comparing the results of surface roughness and of bond strength, revealed that surface roughness did not necessarily resulted in an increase in bond strength. This result is in accordance with previous investigators Reyes et al. [4]. The results showed that for the same type of surface treatment, there was a significant difference in bond strength to ceramic between C.P. Ti and Ti–6Al–4V with the latter recording higher values, while as regard surface roughness a significant difference existed with the C.P. Ti recording higher values. This may be due to the fact that bond between metal and ceramics depends upon other factors (chemical, thermal), this assumption is in accordance with other researchers [7]. As regard thermal cycling, the results showed a decrease in bond strength between C.P. Ti and Ti–6Al–4V to ceramic. This results is in accordance with Fisher et al. (2009) [36] who reported the same findings. The decrease in bond strength due to thermal cycling may be due to the fact that thermal cycling leads to stress at the metal–ceramic interface. Destruction of the bonds starts at the edges of the veneer at the alloy–ceramic interface. As a result, the bonding area is reduced leading to lower bond strength [36]. Also, ceramics show strong tendency towards stress corrosion, a phenomenon which originates in an increasing hydrolysis of the ceramic [37]. 6. Conclusions 1) Silica coating recorded higher surface roughness than airborne-particle abrasion. 2) Both tested treatments recorded bond strength values above the acceptable limit of 25 MPa determined in ISO 9693. 3) C.P. Ti gave higher value of surface roughness than the Ti– 6Al–4V alloy. 4) Ti–6Al–4V recorded higher bond strength to ceramic than C.P. Ti. 5) Increases surface roughness of Ti and Ti–6Al–4V alloy did not obligatory resulted in an increase in their bond strength to ceramics. 6) Aging by thermal cycling decreased Ti and Ti–6Al–4V alloy bond strength to ceramic. References [1] Li J-X, Zhang Y-M, Han, Zhao Y-M. Effects of micro-arc oxidation on bond strength of titanium to porcelain. Surf Coat Technol 2010;204:1252–8. [2] Oshida Y, Munoz CA, Winkler MM, Hashem A, Itoh M. Fractal dimension analysis of aluminum oxide particle for sandblasting dental use. Biomed Mater Eng 1993;3:117–26.
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209 [3] Wagner WC, Asgar K, Bigelow WC, Flinn RA. Effect of interfacial variables on metal–porcelain bonding. J Biomed Mater Res 1993;27:531–7. [4] Reyes MJ, Oshida Y, Andres CJ, Barco T, Hovijitra S, Brown D. Titanium–porcelain system. Part III. Effects of surface modification on bond strengths. Biomed Mater Eng 2001;11:117–36. [5] Kelly M, Asgar K, O’Brien WJ. Tensile strength determination of the interface between porcelain fused to gold. J Biomed Mater Res 1969;3:403–8. [6] Guilherme AS, Henriques GEP, Zavanelli RA, Mesquita MF. Surface roughness and fatigue performance of commercially pure titanium and Ti– 6Al–4V alloy after different polishing protocols. J Prosthet Dent 2005; 93:378–85. [7] Bienias J, Surowska B, Stoch A, Matraszek H, Walczak M. The influence of SiO2 and SiO2–TiO2 intermediate coatings on bond strength of titanium and Ti6Al4V alloy to dental porcelain. Dent Mater 2009;25:1128–35. [8] Moffa JP, Lugassy AA, Guckes AD, Gettleman L. An evaluation of non precious alloys for use with porcelain veneers. Part II: industrial safety and biocompatibility. J Prosthet Dent 1973;30:424–31. [9] Williams TRP, Johnson CE, Winchell PG, Philips RW. Be, Li and Na redistribution near a porcelain Ni alloy interface shown by ion microprobe mass analysis. J Dent Res 1978;57:233–6. [10] Adachi M, Mackert Jr JR, Parry EE, Fairhurst CW. Oxide adherence and porcelain bonding to titanium and Ti–6Al–4V alloy. J Dent Res 1990;69:1230–5. [11] Wang RR, Fung KK. Oxidation behavior of surface-modified titanium for titanium–ceramic restorations. J Prosthet Dent 1997;77:423–34. [12] Kern M, Thompson VP. Effects of sandblasting and silica coating procedures on pure titanium. J Dent 1994;22:300–6. [13] Wang RR, Welsch GE, Monteiro O. Silicon nitride coating on titanium to enable titanium–ceramic bonding. J Biomed Mater Res 1999;46:262–70. [14] Gilbert JL, Covey DA, Lautenschlager EP. Bond characteristics of porcelain fused to milled titanium. Dent Mater 1994;10:134–40. [15] Suansuwan N, Swain MV. Adhesion of porcelain to titanium and a titanium alloy. J Dent 2003;31:509–18. [16] Kim JT, Cho SA. The effects of laser etching on shear bond strength at the titanium ceramic interface. J Prosthet Dent 2009;101:101–6. [17] Al Hussaini I, Al Wazzan KA. Effect of surface treatment on bond strength of low-fusing porcelain to commercially pure titanium. J Prosthet Dent 2005;94:350–6. [18] Zinelis S, Barmpagadaki X, Vergos V, Chakmakchi M, Eliade G. Bond strength and interfacial characterization of eight low fusing porcelains to cp Ti. Dent Mater 2010;26:264–73. [19] Papadopoulos TD, Spyropoulos KD. The effect of a ceramic coating on the CP Ti-porcelain bond strength. Dent Mater 2009;25:247–53. [20] White SN, Ho L, Caputo AA, Goo E. Strength of porcelain fused to titanium beams. J Prosthet Dent 1996;75:640–8.
209
[21] Pang IC, Gilbert JL, Chai J, Lautenschlager EP. Bonding characteristics of low-fusing porcelain bonded to pure titanium and palladium–copper alloy. J Prosthet Dent 1996;73:17–25. [22] Probster L, Maiwald U, Weber H. Three-point binding strength of ceramics fused to cast titanium. Eur J Oral Sci 1996;104:313–9. [23] Atsu S, Berksun S. Bond strength of three porcelains to two forms of titanium using two firing atmospheres. J Prosthet Dent 2000;84: 567–74. [24] ISO 9693. Metal–ceramic dental restorative systems. Geneva: International Organization for Standardization; 1999. [25] Watanabe I, Watanabe E, Yoshida K, Okabe T. Effect of surface contamination on adhesive bonding of cast pure titanium and Ti–6Al–4V alloy. J Prosthet Dent 1999;81:270–6. [26] Leibrock H, Degenhart M, Behr M, Rosentritt M, Handel G. In vitro study on the effect of thermo-and load cycling on the bond strength of porcelain repair systems. J Oral Rehabil 1999;26:130–7. [27] Fischer J. Ceramic bonding to a dental gold–titanium alloy. Biomaterials 2002;23:1303–11. ¨ zcan M. Effect of alloy type and [28] Nergiz I, Schmage P, Herrmann W, O surface conditioning on roughness and bond strength of metal brackets. Am J Orthod Dentofacial Orthop 2004;125:42–50. [29] Kellerhoff RF, Fischer J. In vitro fracture strength and thermal shock resistance of metal–ceramic crowns with cast and machined AuTi frameworks. J Prosthet Dent 2007;97:209–15. [30] Hammad IA, Talic YF. Designs of bond strength tests for metal– ceramic complexes: review of the literature. J Prosthet Dent 1996;75: 602–8. [31] Lee KM, Cai Z, Griggs JA, Guiatas L, Lee DJ, Okabe T, SEM/EDS. evaluation of porcelain adherence to gold-coated cast titanium. J Biomed Mater Res B Appl Biomater 2004;68:165–73. [32] DIN 13927. Metal–ceramic system: German Standard Institution Berlin. Beuth Verlag; 1990. [33] Lenz J, Schwarz S, Schwickerath H, Sperner F, Schafer A. Bond strength of metal–ceramic systems in three-point flexure bond test. J Appl Biomater 1995;6:55–64. [34] Derand T, Hero H. Bond strength of porcelain on cast vs. wrought titanium. Scand J Dent Res 1992;100:184–8. [35] Miyakawa O, Watanabe K, Okawa S, Kanatani M, Nakano S, Kobayashi M. Surface contamination of titanium by abrading treatment. Dent Mater J 1996;15:11–21. [36] Fischer J, Zba¨ren C, Stawarczyk B, Ha¨mmerle CHF. The effect of thermal cycling on metal–ceramic bond strength. J Dent 2009;37:549–53. [37] Scherrer S, Denry I, Wiskoct A, Besler U. Effect of water exposure on the fracture toughness and flexure strength of a dental glass. Dent Mater 2001;17:367–71.
Journal of Prosthodontic Research 56 (2012) 204–209 www.elsevier.com/locate/jpor
Original article
Effect of surface roughness and thermal cycling on bond strength of C.P. titanium and Ti–6Al–4V alloy to ceramic Cherif Adel Mohsen B.D.S., M.D.Sc., D.D.Sc.* Faculty of Dentistry, University of Minia, Minia, Egypt Received 13 January 2011; received in revised form 10 September 2011; accepted 15 October 2011 Available online 10 May 2012
Abstract Purpose: Studying the effect of surface roughness and thermal cycling on titanium–ceramic bonding. Methods: One hundred fourteen samples in the form of bar for the C.P. titanium and Ti–6Al–4V alloy were used. They were divided into two groups according to the type of bar. Each group was then subdivided according to the type of surface treatment to three subgroups, control, airborne-particle abrasion and silica coated. Each subgroup was subdivided into two classes according to the type of test (surface roughness and bond strength). Samples used for the bond strength test were veneered. These samples were subdivided into two subclasses according to thermal cycling; whether without thermal cycling or after 6000 thermal cycles. Results: The surface roughness test results showed that silica coating recorded the highest surface roughness. Also C.P. titanium gave higher value of surface roughness than Ti–6Al–4V alloy. As regard the bond strength, the airborne-particle abrasion classes and the silica coated classes recorded bond strength values above the acceptable limit of 25 MPa determined in ISO 9693. As regard thermal cycling, the results showed that aging by thermal cycling decreased the metal–ceramic bond strength. Conclusions: The airborne-particle abrasion and the silica coating are acceptable treatments for titanium–ceramic restorations. Increasing surface roughness of C.P. titanium and Ti–6Al–4V alloy not necessarily results in an increase in their bond strength to ceramics. Aging affects the metal– ceramic bond strength. # 2011 Japan Prosthodontic Society. Published by Elsevier Ireland. All rights reserved. Keywords: Titanium; Bond strength; Surface roughness; Thermal cycling; Surface treatment
1. Introduction The formation of a metal–ceramic bond at porcelain firing temperatures should be strong enough to resist both transient and residual thermal stresses and mechanical forces in function. The requirements which are to provide such a bond are discussed as chemical, thermal, and mechanical [1,2]. Surface roughness of the titanium can also drastically affect the titanium–porcelain adhesion. Wagner et al. [3] found a direct correlation between surface roughness and bond strength; the greater the roughness, the higher the bond strength. The porous surface would be in favor of mechanical interlocking between the oxide layer and the porcelain [1]. On the contrary, Reyes et al. [4] concluded that increasing surface roughness does not
* Corresponding author at: Fixed Prosthodontics Department, Faculty of Dentistry, Minia University, 4 El Tharrir St., Dokki, Guiza, Egypt. Fax: +20 0233372078. E-mail address: [email protected].
necessarily result in an increase in bond strength. Kelly et al. [5] reported that surface roughness may increase the stress concentration at the metal–ceramic interface, and generate steep re-entrant angles that may prevent complete wetting and result in voids at metal–ceramic interfaces. A comparison of surface roughness between C.P. Ti and Ti– 6Al–4V specimens was carried out by Guilherme et al. They revealed C.P. Ti to be significantly rougher than Ti–6Al–4V alloy [6]. Bienias et al. [7] reported that the surface roughness Ra after sandblasting only reached 0.69 0.02 mm for C.P. titanium and 0.55 0.02 mm for Ti–6A1–4V alloy. The success of the porcelain-fused-to-alloy restorations depends acutely on the success of the strong bonding between porcelain and the metal substructure [8,9]. Bonding porcelain to titanium for porcelain fused to metal (PFM) crowns is still an outstanding problem for the current use of metal–ceramic restorations. The highly oxidative nature of the titanium surface has been postulated as the cause of the weak bond between porcelain and titanium [10,11]. Many surface treatments of
1883-1958/$ – see front matter # 2011 Japan Prosthodontic Society. Published by Elsevier Ireland. All rights reserved. doi:10.1016/j.jpor.2011.10.001
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209
titanium have been proposed to overcome this problem such as sandblasting, acid treatment, coating using silicon nitride or SiO2 or SiO2–TiO2, bonding agents and laser etching [4,7,12– 19]. These studies investigated appropriate treatments to achieve a better bonding between porcelain and titanium as quantified by the resultant bond strength [20–22]. Some authors have shown that Ti–ceramic bonding was acceptable [21,22] but variations exist between different titanium and ceramic systems [23]. The purpose of this present study is (i) to study the effect of C.P. titanium and Ti–6Al–4V alloy surface treatment with airborne-particle abrasion or silica coating on the surface roughness and on the titanium–ceramic bond strength. (ii) To investigate the relation between surface roughness of and bond strength. (iii) To study the effect of thermal cycling on Ti– ceramic bond strength.
205
at 2.8-bar pressure from a distance of 10 mm for 20 s. All specimens were subjected to ultrasonic cleaning using distilled water for 10 min and then rinsed in distilled water. Specimens were left to dry before the application of porcelain. 2.4. Ceramic application (for the bond strength test samples) Next, in the middle of the specimens (8 mm 3 mm 1 mm) Ceramics2in1 (Biodenta, Berneck, Switzerland) lowfusing dental porcelain was fired on C.P. Ti and Ti–6Al–4V alloy according to the manufacturer’s recommendations, as listed in Table 1A custom-made jig was used to ensure the precision of porcelain thickness. 3. Testing procedure
2. Materials and methods 3.1. Surface roughness 2.1. Samples construction One Hundred and fourteen samples of C.P. Ti (grade 2) and Ti alloy Ti–6Al–4V (Titanium Int., USA) were used in this study. The samples were cut using an electric saw to form a bar with the following dimension (25.0 mm 3.0 mm 0.5 mm) as recommended with the ISO 9693 [24]. The samples were then ground on all sides with 600 grit SiC paper under continuous water cooling. Finally the samples were ultrasonically rinsed in deionized water for 30 min. 2.2. Grouping The samples were divided into two groups (57 samples each) according to the type of bar (C.P. Ti or Ti–6Al–4V alloy). Each group was then subdivided according to the type of surface treatment to three subgroups (19 samples each). The first subgroup was not subjected to any surface treatment and served as a control. The second subgroup was subjected to airborneparticle abrasion, while the third subgroup was silica coated. Each subgroup was further subdivided into two classes according to the type of test (five samples for the surface roughness test and 14 samples for the bond strength test). Then the bond strength classes were subdivided into two subclasses according to thermal cycling; whether without thermal cycling or after 6000 thermal cycles. 2.3. Surface treatment - Airborne-particle abrasion: It was accomplished using 250 mm aluminum oxide particles at 2–3 bars air pressure from a distance of 10 mm for 30 s [17,25]. - Silica coating: The samples were first blasted with 250 aluminum oxide particles at 2–3 bars air pressure from a distance of 10 mm for 30 s.The samples were then ultrasonically cleaned for 10 min in a water bath, then dried. Finally, the samples were blasted with a mixture of 50 mm Al2O3 and 50 mm silica particles (Rolloblast 50 mm, Renfert, Germany)
The surface roughness (Ra) was measured in microns using a mechanical profiler (Form Talysurf i60, Taylor Hobson Precision, England). The surface profile included the variation on the surface topography which filtered to obtain surface roughness profile only. The filtration was done using Gaussian filter with cut-off wavelength = 0.25 mm and measuring length = 5 mm. 3.2. Thermal cycling Thermal cycling was performed with bath temperatures of 5 2 8C and 55 2 8C, respectively. The dwell time in each bath was 20 s, the transfer time between the two baths was 10 s, resulting in a frequency of 1 cycle/min. Each sample was subjected to 6000 thermal cycles to simulate a 5-year clinical service [22,26]. 3.3. Bond strength The bond strength testing was performed with a 3-point bending test on universal testing machine (LLOYD Instruments, UK). The specimens were positioned on a specially fabricated metal support with the ceramic positioned on the side opposite that contacted by the applied force. The distance between the supports was 20 mm. A compressive load was applied at the midpoint of the metal strip with a rounded-tip loading rod at a crosshead speed of 0.5 mm/min until a sudden drop in load occurred in the load-deflection curve, indicating the bond failure. The failure load was recorded and calculated by the following equation given in ISO 9693 [24] s ¼kF where F is the maximum force applied in Newton before debonding (failure load), k is a constant determined from a graph in ISO 9693 with units of mm2.The value of k depends
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209
206
Table 1 Firing schedules of Ceramics2in1 veneering material. Ceramics2in1 layers
Standby t (8C)
Closing time (min)
Temp. (8C rise/min)
1st Firing temp. (8C)
2nd Firing temp. (8C)
Holding time
Vacuum start (8C)
Vacuum end (8C)
Special
Opaquer Dentin Glaze
400 400 400
6 4 4
55 45 45
800 740 710
800 730 –
2 1 1
450 450 450
799 739/729 709
– – Slower cooling approx 6 min
Table 2 The surface roughness of the tested groups (mm). Upper: mean, lower: SD. Groups
Control
Airborne-particle abrasion
Silica coating
C.P. Ti
0.25 (0.03)
1.49 (0.11)
1.76 (0.08)
Ti–6Al–4V
0.09 (0.02)
1.02 (0.07)
1.25 (0.08)
Fig. 1. Comparison between the surface roughnesses of the tested groups (mm).
Table 3 The bond strength of the tested groups (MPa). Upper: mean, lower: SD.
on the thickness and the elastic modulus of the tested titanium samples. All data obtained in this research were calculated, tabulated and statistically analyzed. One-way ANOVA and Tukey tests were used for the surface roughness data while a Two-way ANOVA and Fisher’s LSD tests was used for the bond strength data.
Groups
Thermal cycling
Control
Airborne-particle abrasion
Silica coating
C.P. Ti
Without
12.49 (2.70) 9.37 (0.56)
29.40 (3.73) 25.10 (5.52)
33.20 (5.40) 26.45 (5.89)
13.34 (2.13) 10.02 (1.11)
33.14 (6.04) 27.56 (4.39)
37.80 (5.39) 30.79 (7.12)
4. Results The means and standard deviations of the surface roughness for the two tested groups are presented in Fig. 1 and Table 2. A one-way ANOVA test was used to determine significant differences between the tested groups, subgroups and classes ( p < 0.05). The Tukey test for multiple comparisons of means at ( p < 0.05) was done following the one-way analysis of variance. The results showed a statistical difference between the treated surfaces and the control subgroups (no surface treatments). The results showed that there was a statistical difference between the two tested surface treatments. Silica coating recorded the highest surface roughness 1.76 and 1.25 mm for C.P. Ti and Ti–6Al–4V alloy, respectively. While airborne-particle abrasion recorded: 1.49 and 1.02 mm for C.P. Ti and Ti–6Al–4V alloy, respectively. The results also showed that there was a significant difference between cp Ti and Ti– 6Al–4V alloy for the same surface treatment as well as for the control classes: C.P. Ti gave higher value of surface roughness than the Ti–6Al–4V alloy. The means and standard deviations of the bond strength for the two tested groups are presented in Fig. 2 and Table 3. A two way ANOVA Test was used to determine significant differences between the tested groups, subgroups and subclasses ( p < 0.05). Fisher’s LCD method for multiple comparisons of means at ( p < 0.05) was done following the two-way analysis of variance. The results showed a statistical difference
After Ti–6Al–4V
Without After
between the treated surfaces and the control subgroups (no surface treatments). The results showed that for both groups tested, there was a statistical difference between the airborneparticle abrasion subclasses and the silica coated subclasses. At the same time, the results showed that the airborne-particle abrasion subclasses and the silica coated subclasses recorded bond strength values above the acceptable limit of 25 MPa determined in ISO 9693 [24] for the bond strength between titanium or its alloys and different low-fusing dental porcelains. For the same type of surface treatment tested, there was a significant difference between the C.P. Ti and Ti–6Al–4V alloy which recorded higher bond strength values. Thermal cycling resulted in a significant decrease in bond strength for all the tested subclasses. As regard the mode of failure, the C.P. Ti and Ti–6Al–4V alloy surface after debonding demonstrated a combination of cohesive and adhesive bond failures, i.e. traces of porcelain were observed on specimens that were treated with the of airborne-particle abrasion and that silica coated. 5. Discussion In general, the roughness of the surface can be connected to the enhancement of the bond strength of the metal–ceramic
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209
207
Bond strength (MPa) 29.4 12.49
9.37
without aging
aer aging
Control
25.1
without aging
aer aging
Airborne-parcle abrasion
33.2
without aging
26.45
aer aging
Silica coang
c.p.Ti
13.34
10.02
without aging
aer aging
Control
33.14
27.56
without aging
aer aging
Airborne-parcle abrasion
37.8
without aging
30.79
aer aging
Silica coang
Ti-6Al-4V
Fig. 2. Comparison between the bond strength of the tested groups (MPa).
system, which significantly influences the achievement of sufficient bonding, observed in previous studies [27,28]. Roughness can create an increase in the surface area of a metal substrate for porcelain and enhance the metal–ceramic bond [7]. The results of the surface roughness measurements in this study showed that for the same tested group (C.P. Ti/Ti– 6Al–4V), a statistical difference between the treated surfaces and the control subgroups (no surface treatments), also there was a statistical significant difference between the silica coated classes and the airborne-particle abrasion. The silica coated classes gave the highest surface roughness. This may be due to the fact that silica particles may cause deeper penetration on the surface of the C.P. Ti and Ti–6Al–4V alloy and also some of these particles may be leached out during cleaning, leaving a deeper micro-pores. These results contradict those of Bienias et al. [7] who reported that sandblasting recorded higher surface treatment than silica coating. This difference may be due to the method of silica coating used. In the present study, silica coating was performed by the tribochemical method while Bienias et al. used a sol–gel coating. The results showed that by comparing C.P. Ti and Ti–6Al–4V subjected to the same surface treatment, C.P. Ti recorded higher surface roughness than Ti–6Al–4V.These results are in accordance with Guilherme et al. [6] and Bienias et al. [7] who reported the same findings. Clinical failures of metal–ceramic restorations are often due to fractures or chipping of the veneering ceramic. A crucial point is the metal–ceramic bond strength [29]. Obtaining a good quality bond strength of metal–ceramic system depends on the properties of both the metal substrate and dental porcelain, their interactions in the capabilities of chemical and mechanical bonding formation [27]. Compared to the bonding in typical metal–porcelain systems, the adhesion between titanium and porcelain is relatively poor [14] attributed to the forming of a thick titanium layer, known as the ‘‘a-case’’ on its surface [21]. This oxide layer forms during the casting of the titanium and the high temperature sintering of the porcelain [13]. Low fusing ceramics were developed to improve the adhesion of titanium to porcelain. It is claimed that at the firing temperature of the low fusing titanium porcelain (720–750 8C) a dissociation of the superficial native titanium oxides takes place, followed by dissolution of elements within the titanium mass, accompanied by diffusion of the ceramic material elements [19]. According to Adachi et al. [10] at the firing temperature of low fusing porcelain (750 8C) an oxide layer of about 32 nm thickness was
created, whereas at higher temperatures the adherence of oxides was reduced. There are several mechanical tests that have been used to determine the debonding strength between metal and porcelain including a plethora of flexural and shear designs [30]. Generally the bond strength of metal–ceramic systems is evaluated with the three point bending according to ISO 9693:1999 [24] or with the shear testing. The latter although reliable, cannot be directly compared with the three point bending results. In some cases only the fracture mode has been used as a classification criterion [31] based on the former version of ISO 9693 standard, where metal–ceramic strips were bended over a rod to a 908 angle of the specimens ends. The specimens were then flattened and the percentage of adherent porcelain was determined along the predominant part of the middle third of the metallic substrate. However, it is obvious that such a test cannot provide qualitative results. The latest revision of ISO 9693 in 1999 based on the full adoption of the DIN 13927 [32] also commonly known as Schwickerath test [33]. According to this specification the metal–ceramic system comply with the requirements pass ISO 9693 when four out of six specimens have a debonding strength higher than 25 MPa [24]. All samples were subjected to thermal cycling: 6000 cycles to simulate a 5-year clinical service [22,26]. The results showed that the control classes (no surface treatment) recorded the least bond strength for both C.P. Ti (12.49 Mpa) and Ti–6Al–4V (13.34 Mpa) respectively. These results are in accordance with other investigators [4,14,16,17]. These results may be attributed to the fact that unmodified titanium surface, that is, a surface not receiving any surface treatment, produces a weak, non protective and non adherent oxide layer that is unsuitable for porcelain bonding [11]. The results showed that, the control classes (no surface treatment) recorded bond strength values below the acceptable limit of 25 MPa determined in ISO 9693, indicating the importance of using surface treatment for C.P. Ti and Ti–6Al–4V. The mode of failure for control classes showed mainly adhesive fracture at Ti–ceramic interface due to separation of the ceramic material from the substrate. This mode of failure may be attributed to the poor bond strength between the Ti and the ceramic for these subgroups. Poor bond strength may be attributed to the low surface roughness of these subgroups. Several investigators reported that surface treatment using airborne-particle abrasion enhanced the bond strength of commercially pure titanium or Ti alloys to low-fusing porcelain
208
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209
[7,16,17]. In this study, airborne-particle abrasion surface treatment was performed using 250 mm, as Derand and Hero [34] observed that the use of 250 mm alumina particles, compared with 50 mm particles, significantly improved the bond between titanium and ceramic. The bond strength recorded for the airborne-particle abrasion were above the acceptable limit of 25 MPa determined in ISO 9693, indicating that this type of surface treatment may be useful for improving the bond between Ti and ceramic due to the increase of surface area available for mechanical interlocking. At the same time the mode of failure of these subclasses was a combination of cohesive and adhesive bond failures, i.e. traces of porcelain were observed on specimens which may be due to the relatively high bond strength recorded between Ti and ceramic in those subclasses. This performance may be attributed to the fact that airborne-particle abrasion likely improves the bond strength by removing loosely attached furrows, overlaps, and flakes of metal created by grinding procedures, provides mechanical interlocking, increases surface area, and increases wettability [4]. On the other side, results obtained in this research contradict that of Gilbert et al. [14] who reported that airborneparticle abrasion could contaminate the surface of titanium with alumina particles, which could weaken the mechanical interlocking of the porcelain and titanium [35]. The results showed that there was a statistical significant difference between the airborne-particle abrasion subclasses and the silica coated subclasses for C.P. Ti and Ti–6Al–4V. Bienias et al. [7] reported that application of SiO2 as an intermediate coating, produced by the sol–gel method, to both C.P. Ti and Ti– 6Al–4Valloy significantly improves the bond strength of metal– porcelain systems in comparison to the metal substrate only after sandblasting, and may have clinical use. In this present study, the metal was treated by SiO2 in a different way than with Bienias et al. [7] as it was blasted. The silica coated subclasses, recorded higher bond strength value than the airborne-particle abrasion subclasses. This may be due to the fact that blasting the metal with Al2O3 + silica combines micromechanical retention produced by airborne-particle abrasion and chemical bonding resulting from silica coating of the Ti surface. Investigators supposed that during the firing of the ceramic layer, there occurs diffusion of the components and possibly a chemical reaction at the interface between the metal substrate-intermediate coatings– porcelain [7]. Investigators draw the attention to the importance of obtaining chemical bonding to enhance the bond strength between metal substrate and porcelain [15]. According to Fischer [27], the contribution of chemical bonds to metal–ceramic bond strength is much more important than mechanical interlocking, obtained by roughening the surface. The mode of failure of these classes was a combination of cohesive and adhesive bond failures, i.e. traces of porcelain were observed on specimens which may be due to the relatively high bond strength recorded between Ti and ceramic in those subclasses as for the airborneparticle abrasion classes. In this present research, the results showed that irrespective to the type of surface treatment tested, there was significant difference between the C.P. Ti and Ti–6Al–4V alloy, the latter recorded higher bond strength value. According to Adachi et al.
[10] the thicknesses of the oxide layers on C.P. Ti and Ti–6Al– 4V alloy heated to 750 8C were 32 and 11 nm, respectively, but at 1000 8C the thickness was approximately 1 mm for C.P. Ti and Ti–6Al–4V alloy. The thicker the interfacial oxide layer becomes, the weaker the bond strength of metal–porcelain [10]. These results are in accordance with of Bienias et al. [7] who reported that bond strength between ceramic and sandblasted C.P. Ti 23.04 MPa, while it was 28.24 MPa for Ti–6Al–4V. While these results contradicted that of Suansuwan and Swain [15] who reported a significant difference between C.P. Ti and Ti alloy which recorded lesser bond strength value. Comparing the results of surface roughness and of bond strength, revealed that surface roughness did not necessarily resulted in an increase in bond strength. This result is in accordance with previous investigators Reyes et al. [4]. The results showed that for the same type of surface treatment, there was a significant difference in bond strength to ceramic between C.P. Ti and Ti–6Al–4V with the latter recording higher values, while as regard surface roughness a significant difference existed with the C.P. Ti recording higher values. This may be due to the fact that bond between metal and ceramics depends upon other factors (chemical, thermal), this assumption is in accordance with other researchers [7]. As regard thermal cycling, the results showed a decrease in bond strength between C.P. Ti and Ti–6Al–4V to ceramic. This results is in accordance with Fisher et al. (2009) [36] who reported the same findings. The decrease in bond strength due to thermal cycling may be due to the fact that thermal cycling leads to stress at the metal–ceramic interface. Destruction of the bonds starts at the edges of the veneer at the alloy–ceramic interface. As a result, the bonding area is reduced leading to lower bond strength [36]. Also, ceramics show strong tendency towards stress corrosion, a phenomenon which originates in an increasing hydrolysis of the ceramic [37]. 6. Conclusions 1) Silica coating recorded higher surface roughness than airborne-particle abrasion. 2) Both tested treatments recorded bond strength values above the acceptable limit of 25 MPa determined in ISO 9693. 3) C.P. Ti gave higher value of surface roughness than the Ti– 6Al–4V alloy. 4) Ti–6Al–4V recorded higher bond strength to ceramic than C.P. Ti. 5) Increases surface roughness of Ti and Ti–6Al–4V alloy did not obligatory resulted in an increase in their bond strength to ceramics. 6) Aging by thermal cycling decreased Ti and Ti–6Al–4V alloy bond strength to ceramic. References [1] Li J-X, Zhang Y-M, Han, Zhao Y-M. Effects of micro-arc oxidation on bond strength of titanium to porcelain. Surf Coat Technol 2010;204:1252–8. [2] Oshida Y, Munoz CA, Winkler MM, Hashem A, Itoh M. Fractal dimension analysis of aluminum oxide particle for sandblasting dental use. Biomed Mater Eng 1993;3:117–26.
C.A. Mohsen / Journal of Prosthodontic Research 56 (2012) 204–209 [3] Wagner WC, Asgar K, Bigelow WC, Flinn RA. Effect of interfacial variables on metal–porcelain bonding. J Biomed Mater Res 1993;27:531–7. [4] Reyes MJ, Oshida Y, Andres CJ, Barco T, Hovijitra S, Brown D. Titanium–porcelain system. Part III. Effects of surface modification on bond strengths. Biomed Mater Eng 2001;11:117–36. [5] Kelly M, Asgar K, O’Brien WJ. Tensile strength determination of the interface between porcelain fused to gold. J Biomed Mater Res 1969;3:403–8. [6] Guilherme AS, Henriques GEP, Zavanelli RA, Mesquita MF. Surface roughness and fatigue performance of commercially pure titanium and Ti– 6Al–4V alloy after different polishing protocols. J Prosthet Dent 2005; 93:378–85. [7] Bienias J, Surowska B, Stoch A, Matraszek H, Walczak M. The influence of SiO2 and SiO2–TiO2 intermediate coatings on bond strength of titanium and Ti6Al4V alloy to dental porcelain. Dent Mater 2009;25:1128–35. [8] Moffa JP, Lugassy AA, Guckes AD, Gettleman L. An evaluation of non precious alloys for use with porcelain veneers. Part II: industrial safety and biocompatibility. J Prosthet Dent 1973;30:424–31. [9] Williams TRP, Johnson CE, Winchell PG, Philips RW. Be, Li and Na redistribution near a porcelain Ni alloy interface shown by ion microprobe mass analysis. J Dent Res 1978;57:233–6. [10] Adachi M, Mackert Jr JR, Parry EE, Fairhurst CW. Oxide adherence and porcelain bonding to titanium and Ti–6Al–4V alloy. J Dent Res 1990;69:1230–5. [11] Wang RR, Fung KK. Oxidation behavior of surface-modified titanium for titanium–ceramic restorations. J Prosthet Dent 1997;77:423–34. [12] Kern M, Thompson VP. Effects of sandblasting and silica coating procedures on pure titanium. J Dent 1994;22:300–6. [13] Wang RR, Welsch GE, Monteiro O. Silicon nitride coating on titanium to enable titanium–ceramic bonding. J Biomed Mater Res 1999;46:262–70. [14] Gilbert JL, Covey DA, Lautenschlager EP. Bond characteristics of porcelain fused to milled titanium. Dent Mater 1994;10:134–40. [15] Suansuwan N, Swain MV. Adhesion of porcelain to titanium and a titanium alloy. J Dent 2003;31:509–18. [16] Kim JT, Cho SA. The effects of laser etching on shear bond strength at the titanium ceramic interface. J Prosthet Dent 2009;101:101–6. [17] Al Hussaini I, Al Wazzan KA. Effect of surface treatment on bond strength of low-fusing porcelain to commercially pure titanium. J Prosthet Dent 2005;94:350–6. [18] Zinelis S, Barmpagadaki X, Vergos V, Chakmakchi M, Eliade G. Bond strength and interfacial characterization of eight low fusing porcelains to cp Ti. Dent Mater 2010;26:264–73. [19] Papadopoulos TD, Spyropoulos KD. The effect of a ceramic coating on the CP Ti-porcelain bond strength. Dent Mater 2009;25:247–53. [20] White SN, Ho L, Caputo AA, Goo E. Strength of porcelain fused to titanium beams. J Prosthet Dent 1996;75:640–8.
209
[21] Pang IC, Gilbert JL, Chai J, Lautenschlager EP. Bonding characteristics of low-fusing porcelain bonded to pure titanium and palladium–copper alloy. J Prosthet Dent 1996;73:17–25. [22] Probster L, Maiwald U, Weber H. Three-point binding strength of ceramics fused to cast titanium. Eur J Oral Sci 1996;104:313–9. [23] Atsu S, Berksun S. Bond strength of three porcelains to two forms of titanium using two firing atmospheres. J Prosthet Dent 2000;84: 567–74. [24] ISO 9693. Metal–ceramic dental restorative systems. Geneva: International Organization for Standardization; 1999. [25] Watanabe I, Watanabe E, Yoshida K, Okabe T. Effect of surface contamination on adhesive bonding of cast pure titanium and Ti–6Al–4V alloy. J Prosthet Dent 1999;81:270–6. [26] Leibrock H, Degenhart M, Behr M, Rosentritt M, Handel G. In vitro study on the effect of thermo-and load cycling on the bond strength of porcelain repair systems. J Oral Rehabil 1999;26:130–7. [27] Fischer J. Ceramic bonding to a dental gold–titanium alloy. Biomaterials 2002;23:1303–11. ¨ zcan M. Effect of alloy type and [28] Nergiz I, Schmage P, Herrmann W, O surface conditioning on roughness and bond strength of metal brackets. Am J Orthod Dentofacial Orthop 2004;125:42–50. [29] Kellerhoff RF, Fischer J. In vitro fracture strength and thermal shock resistance of metal–ceramic crowns with cast and machined AuTi frameworks. J Prosthet Dent 2007;97:209–15. [30] Hammad IA, Talic YF. Designs of bond strength tests for metal– ceramic complexes: review of the literature. J Prosthet Dent 1996;75: 602–8. [31] Lee KM, Cai Z, Griggs JA, Guiatas L, Lee DJ, Okabe T, SEM/EDS. evaluation of porcelain adherence to gold-coated cast titanium. J Biomed Mater Res B Appl Biomater 2004;68:165–73. [32] DIN 13927. Metal–ceramic system: German Standard Institution Berlin. Beuth Verlag; 1990. [33] Lenz J, Schwarz S, Schwickerath H, Sperner F, Schafer A. Bond strength of metal–ceramic systems in three-point flexure bond test. J Appl Biomater 1995;6:55–64. [34] Derand T, Hero H. Bond strength of porcelain on cast vs. wrought titanium. Scand J Dent Res 1992;100:184–8. [35] Miyakawa O, Watanabe K, Okawa S, Kanatani M, Nakano S, Kobayashi M. Surface contamination of titanium by abrading treatment. Dent Mater J 1996;15:11–21. [36] Fischer J, Zba¨ren C, Stawarczyk B, Ha¨mmerle CHF. The effect of thermal cycling on metal–ceramic bond strength. J Dent 2009;37:549–53. [37] Scherrer S, Denry I, Wiskoct A, Besler U. Effect of water exposure on the fracture toughness and flexure strength of a dental glass. Dent Mater 2001;17:367–71.