- Email: [email protected]
Resin shear bond strength to porcelain and a base metal alloy using two polymerization schemes
Resin shear bond strength to porcelain and a base metal alloy using two
polymerization schemes Ibrahim Fevzi Tulunoglu, DDS, MS, PhD,a and Bedri Beydemir, DDS, MS, PhD b
Hacettepe University Faculty of Dentistry, Ankara, Turkey Statement of problem. Fractures in ceramometal restorations can occur and need to be repaired because replacements are not an economic solution. Purpose. This study evaluated the shear bond strengths of 4 porcelain repair systems (Metabond C&B [ME], Silistor [SI], Clearfil Lustre [CL], and Scotchbond Multipurpose Plus [SQ]) to a base metal alloy and porcelain in relation with the polymerization shrinkage of a visible light-cured composite superstructure and compared with the ceramometal bond strength (Vita VMK 68). Material a n d m e t h o d s . Thirty-two samples were prepared for each bonding system: 16 for resin-metal bond strength test, and 16 for resin porcelain bond strength test. For each group, bonding agent was applied to 8 substructures and the resin superstructure was polymerized onto the bonding agent; and for the remaining 8 specimens, prepolymerized resin superstructures were bonded with bonding agent. All specimens were subjected to 500 cycles between 5°C and 55°C with 20 seconds dwell time. Tests were performed in a mechanical testing machine with a 0.5 ram/rain crosshead speed. Results. All materials showed an increase in shear bond strength when prepolymerized resin superstructures were used. However, the effect of polymerization shrinkage of resin superstructure was statistically significant only for CL group (/'<.05). The highest metal-resin bond was obtained from ME group with prepolymerized resin superstructures (35.27 _+2.40 MPa), and the lowest value was obtained for the SI group in which resin superstructures were polymerized in situ (8.71 _+1.03 MPa). The highest porcelain-resin bond was obtained from SC group with prepolymerized resin superstructures (20.71 _+1.13 MPa) and the lowest was obtained from SI group (9.99 _+ 1.52 MPa). Conclusion. Higher bond strength values were obtained with prepolymerized resin superstructures compared to in situ polymerized superstructures. Metabond C&B provided the best results for both prepolymerized and in situ polymerized resin superstructure preparation techniques at the failures where metal was exposed. The best results in situations in which the fracture is limited into porcelain were obtained with the use of Scotchbond Multipurpose Plus material. However, a variety of in vivo and in vitro tests are required before a final judgment is made. (J Prosthet Dent 2000;83:181-6.)
Ceramic materials with excellent biocompatibility are used to achieve highly esthetic characteristics in metal-ceramic restorations. However, failures can occur and replacements are n o t an economic solution. A sound repair is the best solution in most o f cases. Two types o f b o n d , metal-resin and porcelain-resin, are involved in the repair process o f ceramometal restorations. Surface configuration, reactivity o f the bonding surfaces, and use o f adhesive resins are important for metal-resin bond. 1-6 To achieve a satisfactory bond between porcelainresin and metal-resin, several mechanical and chemical aAssistant Professor, Department of Prosthodontics. bprofessor, Department of Prosthodontics, Gulhane Military Medical Academy, Center of Dental Sciences. FEBRUARY2000
retention systems were developed. Mechanical roughening o f porcelain surfaces with a coarse diamond has improved repair strengths. 7 Air abrasion (sandblasting) 8 and acid etching with hydrofluoric acid, 9-11 acidulated phosphate fluoride, 12 a m m o n i u m bifluoride, 13 or phosphoric acid are other c o m m o n l y used methods to achieve retentive porcelain surface texture. It has been reported that the use o f sandblasting causes a metal surface with higher chemical reactivity than acid etching and mechanical roughening with rotary instruments, s G o o d bond strengths were reported with the use o f hydrofluoric acid alone and with its combinations with other surface roughening methods. 14-18 However, intraoral use o f hydrofluoric acid may eventually be harmful to oral tissues because o f the very aggressive nature o f this acid in required concentrations THE JOURNAL OF PROSTHETICDENTISTRY 181
THE JOURNAL OF PROSTHETIC DENTISTRY
TULUNOGLU AND BEYDEMIR
~
Sbe,aa'~bl~e
F F
D ~
of ~ foro~ ~plied
Fig. 1. Schematic representation of test system.
Table I. Distribution of the specimens through the groups
ME
MATERIAL AND METHOD
ON
SON
P
8
8
M
8
8
SI
P
8
8
CL
M P
8 8
8 8
M
8
8
P
8
8
M
8
8
SC
ON: Resin superstructures were prepared onto the adhesive surfaces; SON: performed resin superstructures were luted with adhesives to the metal or porcelain surfaces; ME: Metabond C&B group; SI: Silistor group; CL: Clearfil Lustre group; SC: Scotchbond Multipurpose Plus group; P: porcelain-resin group; M: metal-resin group.
for etching porcelain surfaces. 19 Thermal and load cycling are important focuses of dental material testing in simulating oral conditions.l,6,11,14,20-25 Silanes are chemical products that can bond to organic materials with their organic endings and to silica materials with their SiO endings.26,27 They are often used for bonding composite resins to porcelain.28, 29 Light-cured composites have better water absorption, wear resistance properties, low polymerization shrinkage, and their manipulation is easier.3°, 31 However, Lai et a132 stated that they found 2% polymerization shrinkage for light-cured composites. In this study, the shear bond strengths of 4 used porcelain repair systems: Metabond C&B (ME), Silistor (SI), Clearfil Lustre (CL), and Scotchbond Multipurpose Plus (SQ) to metal and porcelain were investigated related to the polymerization shrinkage of a resin composite and compared with the ceramometal bond strength (Vita VMK 68 and Remanium CS). 182
One hundred and twenty-eight specimens were prepared for 4 groups, each representing a different repair material. Each repair material was tested fbr bonding resin to a nickel-chromium base metal alloy containing Ni 61%, Cr 26%, SI 1.5%, Mo 11%, S-Fe-Ce-AI lower than 1% (Remanium CS, Dentaurum, Pfbrzheim, Germany); and porcelain (VMK-68, VITA, Schaan, Liechtenstein) and in each subgroup, preformed resin superstructures were luted with adhesive resins to half of the substructures, and resin superstructures were polymerized onto the adhesive painted metal or porcelain substructures fbr the other half of the specimens (Fig. 1 and Table I). Standard cylindrical wax patterns (9 mm diameter [O] and 1 m m thick) were prepared, invested, and cast in Remanium CS alloy. The cast metal substructures were then embedded in autopolymerizing acrylic resin in standard jigs (10 mm diameter and 20 m m in height). Porcelain subgroups were prepared and fired in standard jigs (9 mm diameter and 2 mm height) and after firing, were embedded in autopolymerizing resin in the same jig as were for the metal specimens. Metal surfaces and the porcelain surfaces were sandblasted with 50 g aluminum oxide for 10 seconds at a constant pressure of 0.4 MPa. Metabond C&B (Parkell, Farmingdale, N.Y.) was applied onto the prepared specimen surfhces according to the manufacturer's recommendations: Four drops of liquid were mixed with 1 drop of catalyst, and then mixed with powder and applied onto the specimen surf:aces and air dried for 10 minutes. For the preparation of specimens of Silistor (Kulzer, Wehrheim, Germany) groups, Silicer was applied onto the substructures and air dried for 2 minutes, then a 1:1 mix of Silibond and Dentacolor Opaque was applied onto the metal surVOLUME 83 NUMBER 2
T U L U N O G L U A N D BEYDEMIR
THE J O U R N A L OF PROSTHETIC DENTISTRY
Table II. The base statistical values obtained from shear strength testing ON
ME SI CL SC
P M P M P M P M
SON
Mean (MPa)
SE
SD
Minimum
Maximum
Mean (MPa)
SE
SD
Minimum
Maximum
13.85 34.55 9.99 8.71 12.79 20.57 16.26 9.94
0.82 2.03 1.52 1.03 0.61 0.82 1.55 1.15
2.31 5.74 4.31 2.91 1.72 2.33 4.37 3.04
10.27 24.18 2.45 1.85 9.77 17.05 6.85 6.84
17.21 40.64 14.04 11.10 14.91 23.62 20.03 14.45
14.93 35.27 10.53 9.33 15.60 30.30 20.71 11.74
0.85 2.40 0.86 0.34 1.00 1.19 1.13 0.97
2.40 6.78 2.42 0.97 2.64 3.33 3.18 2.75
12.02 20.02 6.11 7.98 10.59 23.89 14.94 5.42
19.25 40.64 13.42 10.59 18.21 34.28 24.23 13.89
ON: Resin superstructures were prepared onto the adhesive surfaces; SON: performed resin superstructures were luted with adhesives to the metal or porcelain surfaces; M£: Metabond C&B group; SI: Silistor group; CL: Clearfil Lustre group; SC: Scotchbond Multipurpose Plus group; P: porcelainlresin group; M: metalresin group.
Table III. Results of 3-way ANOVA test ((z=.05)
Source
Corrected model Intercept Material groups (ME, SI, CL, SC) Bonding surfaces (P, M) Polymerization schemes (ON, SON) Material groups-bonding surfaces Material group-polymerization schemes Bonding surfaces-polymerization schemes Material groups x bonding surfaces-polymerization schemes Error Tota I Corrected total
Type III sum of squares
9358.948* 37129.7 4029.224 1020.688 227.002 3766.528 161.495 7.137 100.216 1359.957 48225.6 1O718.9
df
Mean square
F
Significance
15 1 3 1 1 3 3 1 3 110 126 125
623.930 37129.7 1343.075 1020.688 227.002 1255.509 53.832 7.137 33.405 12.363
50.467 3003.235 108.634 82.568 18.361 101.552 4.354 0.577 2.702
<.0005 <.0005 <.0005 <.0005 <.0005 <.0005 .006 .449 .049
R2 - 0.873 (adjusted R2 - 0.856/. ON: Resin superstructures were prepared onto the adhesive surfaces; SON: performed resin superstructures were luted with adhesives to the metal or porcelain surfaces; ME: Metabond C&B group; SI: Silistor group; CL: Clearfil Lustre group; SC: Scotchbond Multipurpose Plus group; P: porcelainlresin group; M: metalresin group.
faces and Silibond was applied onto the porcelain surfaces and light polymerizcd for 40 seconds. The porcelain substructure surfaces of Clearfil Lustre (Cavcx, Haarlem, Holland) group were acid etched with phosphoric acid (K-etchant, Kuraray Co Ltd, Osaka, lapan), cleaned with tap water and a mix of Porcelain Activator, Photobond catalyst and Universal (Kuraray Co. Ltd) was applied and air dried for 30 seconds and finally opaque was applied and light cured Ibr 60 seconds. Scotchbond Ceramic Primer was applied on the porcelain surfaces of Scotchbond Multipurpose Plus (3M Dental Products, St Paul, Minn.) group specimens and after air drying, ScotchBond Multipurpose Plus adhesive activator was applied and light cured tbr 60 seconds. The same procedure was followed for the preparation of metal samples, except the Scotchbond ceramic prirncr application. FEBRUARY 2000
All resin superstructures were prepared with Clearfil Lustre composite (Cavex, Holland). Specimens were stored in distilled water at room temperature for 48 hours and then subjected to 500 thermocycles, between 5°C and 55°C with a 20-second dwell time. Shear strength tests were perfbrmed in Lloyd mechanical testing machine (model LR30K, Lloyd Instruments, Farnham, U. K.) with a crosshcad speed of 0.5 ram/rain. Statistical evaluation was per~brmed with 3-way analysis of variance (ANOVA) and Tukey tests. Data were evaluated with 3-way ANOVA after the demonstration of the hypothesis that the error variances of the dependent variable is equal across groups with Levcne's test RESULTS The results of the shear bond strength tests are presented in Table II. The 3-way ANOVA test revealed that 183
THE JOURNAL OF PROSTHETIC DENTISTRY
= 10
;
S~
T U L U N O G L U A N D BEYDEMIR
:!
ME
Ck Tell group=
$C
Fig. 2. Mean shear bond strengths of porcelain surface
groups.
Sl
$C
CL T u t q~oul~
ME
Fig. 4. Mean shear bond strengths of in situ polymerized resin superstructure groups.
8
i SI
SC
CL Telt groups
ME
SI
SC
CL
ME
T u t Group
Fig. 3. Mean shear bond strengths of metal surface groups.
Fig. 5. Mean shear bond strengths of prepolymerized resin superstructure groups.
the differences between bonding system groups (F=108.634, P=.0001), between porcelain and metal groups (F=82.558, P=.0001), and betwecn in situ polymerizcd and prepolymcrized resin superstructure groups werc significant (F= 18.361, P=.0001) (Table III). Interactions between repair system groups and porcelain-and-metal groups (F= 101.552, P=.0001), and fbr interactions between repair system groups and in situ polymerized-and-prepolymerized resin superstructure groups were significant (F=4.354, P=.0001). However, interactions between porcelain-and-metal groups and in situ polymerizcd-and-prcpolymerized resin superstructure groups wcre not significant (F=0.577, P>.0S),and the interactions between repair system groups, porcelain-and-metal groups, and in situ polymerized-and-prcpolymerized resin superstructure groups were significant (F=2.702, P=.05). When Clearfil Lustre was used on metal surfaces, the differences between the in situ polymerized-andprepolymerized resin superstructure subgroups were statistically significant (P=.0117). Also, when Scotchbond Multipurpose Plus was used on porcelain surfaces, the differences between the in situ polymerizedand-prepolymerized resin superstructure subgroups were statistically significant (P=.0251 ). The differences between the in situ polymerized-and-prepolymerized
resin superstructure subgroups of the other groups were not statistically significant (P>.05) (Figs. 2 through 5).
184
DISCUSSION The renewal of the fracturcd ccramomctal restorations causes a waste of time and economical problems. Recently, many ceramomctal restoration repair systems and materials were developed to overcomc this important problem. A noble metal alloy could be included to this study. However, this seemed to overcrowd the parameters and complicate the study too much to interpret. Nickel-chromium alloys are widely used in the fabrication of ceramometal restoration framework. Adhesive resins show better bonding properties to base metal alloys because of the metal oxides they can easily form at their surfaces. 1,2 Adhesive materials containing 4-META had been shown to havc good bonding properties to base metal alloys. 3-5 It has also been shown that metal oxides, especially chromium oxide plays an important role at the bond between 4-META containing adhesive materials and metal alloys. 3,4 The best results among the metal surface subgroups of the 4 materials tested were obtained from Metabond C&B group, and this was in accordance with the findings of these previous studies. VOLUME 83 NUMBER 2
T U L U N O G L U A N D BEYDEMIR
The porcelain resin bond strengths obtained from Metabond C&B group (in situ polymerized, 13.85 _+2.31 MPa; prepolymerized, 14.93 _+21.93 MPa) are differing from the findings of Kato et al. 1 This may be because they have used a different experimental design in which they bonded 2 identical porcelain disks to each other with adhesive resins. Appeldoorn et al27 evaluated bond strength of composite to porcelain after 3 months of storage in distilled water, followed by 2500 thermocycles; they demonstrated that relatively high bond strengths were achieved with sandblasting specimens. Kato et al 1 reported that porcelain-to-porcelain bond achieved by use of porcelain liner M and Superbond C&B porcelain liner exhibited bond strengths greater than 20 MPa after thermocycling. Unfortunately, they did not included any of the other systems investigated in this study. Porcelain Liner M contains a silane coupler and a carboxylic compound (4-META), the combination can activate the silane. In their study, primers that exhibited durable bond strengths were not a prehydrolyzed type but a 2 or 3 component type. This type of primer does not use an aqueous solution during silane treatment and has a prolonged shelf-life. Acid activators have hydrophobic structures in the compound and also have curing ability. It is probable that these factors reciprocally enhanced the bond to porcelain. Furthermore, the best results in our study were obtained with such materials. Suliman et al28 investigated the effects of hydrofluoric acid-etching, roughening with diamond burs and hydrofluoric acid-etching, and roughening with diam o n d burs and sandblasting, and reported that the most effective surface treatment for porcelain-resin bond was mechanical roughening with diamond burs, followed by chemical etching with hydrofluoric acid. They used an intraoral sandblasting device and 50 gm aluminum oxide and 15-second application time. Although they did not report the sandblasting pressure, their findings and conclusions were different from the findings of Kato et al, 1 who reported that light sandblasting provided the best results. Appeldoorn et a127 evaluated the bond strength of composite to porcelain after 2 months water storage, fbllowed by 2500 thermocycles. Their results demonstrated that relatively higher bond strengths were achieved with sandblasted specimens. Their results differed from those of Bertolotti et al,8 who reported higher bond strengths with air abrasion than with diamond roughening. Llobell et alll preferred to use 8% hydrofluoric acid application versus sandblasting because even at light pressure, mitraoral air-abrasive spraying procedures may damage the remaining glazed porcelain surfaces unless careful isolation is added to the recommended guidelines. However, intraoral use of hydrofluoric acid may eventually be harmful to oral tissues because of the FEBRUARY 2000
THE JOURNAL OF PROSTHETIC DENTISTRY
aggressive naturc of this acid in requircd concentrations for etching porcelain surfaccs. 19 Load and thermal cycling arc important focuses of dental material testing in simulating oral conditions. Opinions vary regarding the effects of thermal cycling on the bond strength of porcelain repair systems. Most authors have suggested that bond strengths of all the tested materials were negatively affected from thermal cycling,l,6,14, 20-22 and some stated that therc were no significant differcnces in bond strengths of some of the materials they testcd.18, 26 The detrimcntal effect of thermocycling on the porcelain-resin or metal-rcsin bond is based on the differcnce of thcrmal expansion coefficients. 23-2~ In our study, as the effects of different amounts ofthermocyclcs were not cvaluated, an application of 500 thermocyclcs was chosen as this amount seemed to represent a mean value of the thermocycling applications in previous studies. However, there is a need of well-established scales, indicating the amount of thermocycles corresponding to a certain period of intraoral use of different materials and their combinations. Another important application in simulation of oral conditions is fatigue test and load fatigue may be defined as a phenomcnon, whercby a sample that has been repeatedly subjected to load well below the level that causes fracture in static tensile tests eventually fails after being subjected to this comparatively small cyclic load. This fatiguc failure is preceded by a combination of crack initiation and crack propagation, followed by a failure that occurred in the form of a fracture. Llobell et al ll state that fatigue tests can be a more significant test as it can simulate long-term clinical use. In our study, the effect of polymerization shrinkage of the composite resulted in significant differences fbr Clearfil Lustre-metal (P=.00117) and Scotchbond Multipurpose Plus-porcelain (P=.0251) groups. These results indicate that better results can be obtained with the indirect preparation of the resin superstructure when these repair materials are used. There are no reports on the minimum shear bond strength value required for the ceramometal restoration repair materials. However, one can easily assume that this value must be near or more than the metal-porcelain bond strength. In our study, higher values than the control group were obtained from in situ polymerized resin superstructures onto metal surfaces coated with Metabond C&B, prepolymerized resin superstructures luted onto metal surfaces with the use of Metabond C&B, in situ polymerized resin superstructures onto metal surfaces coated with Clearfil Lustre, prepolymerized resin superstructures luted onto metal surfaces with the use of Clearfil Lustre, and prepolymerized resin superstructures luted onto metal surfhces with Scotchbond Multipurpose Plus groups. These results indicate that the use of Metabond C&B where metal is exposed provided the best results for both resin super185
THE JOURNAL OF PROSTHETIC DENTISTRY
structure preparation techniques and that the best results could be obtained with the use of Scotchbond Multipurpose Plus in situations where fracture is limited to porcelain. CONCLUSIONS Within the limits of this study, the following conclusions were drawn: 1. Higher bond strength values were obtained with prepolymerized superstructures compared to in situ polymerized superstructures. 2. Metabond C&B provided the best results tbr both prepolymerized and in situ polymerized resin superstructure preparation techniques at the failures where metal is exposed. 3. The best results when the fracture was limited into porcelain were obtained with Scotchbond Multipurpose Plus porcelain liner. However, in vivo and in vitro tests are required before a final judgment is made. REFERENCES 1. Kato H, Matsumura H, Tanaka T, Atsuta M. Bond strength and durability of porcelain bonding systems. J Prosthet Dent 1996;75:163-8 2. Crim GA, Garcia-Godoy F. Microleakage: the effect of storage and cycling duration. J Prosthet Dent 1987;57:574-6. 3. Taira Y, Yoshida K, Matsumura El, Tanaka T, Atsuta M. Adhesive metal primers for a light cured opaque resin. In: Advanced Prosthodontics Worldwide, Proceeding on World Congress on Prosthodontics held in Hiroshima, 1991. Hiroshima, Japan: Publication Committee; September 1991. p. 217. 4. Barzilay I, Myers ML, Cooper LB, Graser GN. Mechanical and chemical retention of laboratory cured composite to metal sudaces. J Prosthet Dent 1988;59:131-7. 5. Cooley RL, Tseng EY, EvansJG. Evaluation of 4-META porcelain repair system. J Esther Dent 1991 ;3:11-3. 6. Beck DA, Janus CE, Douglas HB. Shear bond strength of composite resin porcelain repair materials bonded to metal and porcelain. J Prosthet Dent 1990;64:529-33. 7. }ochen DG, Caputo AA. Composite resin repair of porcelain denture teeth. J Prosthet Dent 1977;38:673-9. 8. Bertolotti RL, Lacy AM, Watanabe LG. Adhesive monomers for porcelain repair. Int J Prosthodont 1989;2:483-9. 9. Simonsen RJ, Calamia JR. Shear bond strength of composite bond to etched porcelain. J Dent Res 1983;62:297 (abstract). 10. Stangel I, Nathanson D, Hsu CS. Shear bond strength of composite bond to etched porcelain. J Dent Res 1987;66:1460 5. 11. Llobe]l A, Nicho[ls JI, Kois JC, Da]y CH. Fatigue life of porcelain repair systems. Int J Prosthodont 1992;5:205-13. 12. Lacy AM, Laluz J, Watanabe LG, Dellinges M. Effect of porcelain surface treatment on the bond to composite. J Prosthet Dent 1988;60:288-91. 13. Kern M, Thompson VP. Bonding to glass infiltrated alumina ceramic: adhesive methods and their durability. J Prosthet Dent 1995;73:240-9.
186
TULUNOGLU AND BEYDEMIR
14. Thurmond JW, Barkmeier WW, Wilwerding T. Effect of porcelain surface treatments on bond strengths of composite resin bonded to porcelain. J Prosthet Dent 1994;72:355-9. 15. Roulet JF, S6derholm KJ, Longmate I. Effects of treatment and storage conclitions on ceramic/composite bond strength. J Dent Res 1995;74:381-7. 16. Ozden AN, Akaltan E, Can G. Effect of surface treatments of porcelain on the shear bond strength of applied dual-cured cement. J Prosthet Dent 1994;72:85-8. 1 7. Aida M, Hayakawa T, Mizukawa K. Adhesion of composite to porcelain with various surface conditions. J Prosthet Dent 1995;73:464-70. 18. Kupiec KA, Wuertz KM, Barkmeier WW, Wilwerding TM. Evaluation of porcelain surface treatments and agents for composite-to-porcelain repair. J Prosthet Dent 1996;76:119-24. 19. Fan PL. Porcelain repair materials. Council on Dental Materials, Instruments, and Equipment. J Am Dent Assoc 1991 ;122:124-30. 20. Creugers NH, Snoek PA, Kayser AF. An experimental porcelain repair system evaluated under controlled clinical conditions. J Prosthet Dent 1992;68:727-7. 21. Gregory WA, Hagen CA, Powers JM. Composite resin repair of porcelain using different bonding materials. Oper Dent 1988;13:114-8. 22. Pratt RC, Burgess JO, Schwartz RS, Smith JH. Evaluation of bond strength of six porcelain repair systems. J Prosthet Dent 1989;62:11-3. 23. Newburg R, Pameijer CH. Composite resins bonded to porcelain with silane solution. J Am Dent Assoc 1978;96:288-9. 24. Eames WB, Rogers LB, Feller PR, Price WR. Bonding agents for repairing porcelain and gold: an evaluation. Oper Dent 1977;2:118-24. 25. Nowlin TP, Barghi N, Norling BK. Evaluation of the bonding of three porcelain repair systems. J Prosthet Dent 1981 ;46:516-8. 26. Diaz-Arno[d AM, Schneider RL, Aqui[ino SA. Bond strengths of intraoral porcelain repair materials. J Prosthet Dent 1989;61:305-9. 27. Appeldoorn RE, Wilwerding TM, Barkmeier WW. Bond strength of composite resin to porcelain with newer generation porcelain repair materials. J Prosthet Dent 1993;70:6-11. 28. Suliman AH, Swift EJJr, Perdigao J. Effects of surface treatments and bonding agents on bond strength of composite resin to porcelain. J Prosthet Dent 1993;70:118-120. 29. Bailey JH. Porcelain-to-composite bond strengths using four organosilane materials. J Prosthet Dent 1989;61:174-7. 30. Dennison JB. Status report on microfilled composite restorative resin. Council on Dental Materials, Instruments, and Equipment. J Am Dent Assoc 1982;105:488-92. 31. Lutz F, Phillips RW. A classification and evaluation of composite resin systems. J Prosthet Dent 1983;50:480-8. 32. Lai JH, Johnson AE. Measuring polymerization shrinkage of photo-activated restorative materials by a water-filled dilatometer. Dent Mater 1993; 9:139-43. Reprint requests to: DR [BRAHIM F. TULUNOGLU BAHCELIEVLER15 SOK. 41/1 06490 ANKARA TURKEY Fax: (312)3113741 Copyright © 2000 by The Editorial Council of The Journal of Prosthetic Dentisto4 0022-3913/2000/$12.00 + 0. 10/1/103750
VOLUME 83 NUMBER 2
polymerization schemes Ibrahim Fevzi Tulunoglu, DDS, MS, PhD,a and Bedri Beydemir, DDS, MS, PhD b
Hacettepe University Faculty of Dentistry, Ankara, Turkey Statement of problem. Fractures in ceramometal restorations can occur and need to be repaired because replacements are not an economic solution. Purpose. This study evaluated the shear bond strengths of 4 porcelain repair systems (Metabond C&B [ME], Silistor [SI], Clearfil Lustre [CL], and Scotchbond Multipurpose Plus [SQ]) to a base metal alloy and porcelain in relation with the polymerization shrinkage of a visible light-cured composite superstructure and compared with the ceramometal bond strength (Vita VMK 68). Material a n d m e t h o d s . Thirty-two samples were prepared for each bonding system: 16 for resin-metal bond strength test, and 16 for resin porcelain bond strength test. For each group, bonding agent was applied to 8 substructures and the resin superstructure was polymerized onto the bonding agent; and for the remaining 8 specimens, prepolymerized resin superstructures were bonded with bonding agent. All specimens were subjected to 500 cycles between 5°C and 55°C with 20 seconds dwell time. Tests were performed in a mechanical testing machine with a 0.5 ram/rain crosshead speed. Results. All materials showed an increase in shear bond strength when prepolymerized resin superstructures were used. However, the effect of polymerization shrinkage of resin superstructure was statistically significant only for CL group (/'<.05). The highest metal-resin bond was obtained from ME group with prepolymerized resin superstructures (35.27 _+2.40 MPa), and the lowest value was obtained for the SI group in which resin superstructures were polymerized in situ (8.71 _+1.03 MPa). The highest porcelain-resin bond was obtained from SC group with prepolymerized resin superstructures (20.71 _+1.13 MPa) and the lowest was obtained from SI group (9.99 _+ 1.52 MPa). Conclusion. Higher bond strength values were obtained with prepolymerized resin superstructures compared to in situ polymerized superstructures. Metabond C&B provided the best results for both prepolymerized and in situ polymerized resin superstructure preparation techniques at the failures where metal was exposed. The best results in situations in which the fracture is limited into porcelain were obtained with the use of Scotchbond Multipurpose Plus material. However, a variety of in vivo and in vitro tests are required before a final judgment is made. (J Prosthet Dent 2000;83:181-6.)
Ceramic materials with excellent biocompatibility are used to achieve highly esthetic characteristics in metal-ceramic restorations. However, failures can occur and replacements are n o t an economic solution. A sound repair is the best solution in most o f cases. Two types o f b o n d , metal-resin and porcelain-resin, are involved in the repair process o f ceramometal restorations. Surface configuration, reactivity o f the bonding surfaces, and use o f adhesive resins are important for metal-resin bond. 1-6 To achieve a satisfactory bond between porcelainresin and metal-resin, several mechanical and chemical aAssistant Professor, Department of Prosthodontics. bprofessor, Department of Prosthodontics, Gulhane Military Medical Academy, Center of Dental Sciences. FEBRUARY2000
retention systems were developed. Mechanical roughening o f porcelain surfaces with a coarse diamond has improved repair strengths. 7 Air abrasion (sandblasting) 8 and acid etching with hydrofluoric acid, 9-11 acidulated phosphate fluoride, 12 a m m o n i u m bifluoride, 13 or phosphoric acid are other c o m m o n l y used methods to achieve retentive porcelain surface texture. It has been reported that the use o f sandblasting causes a metal surface with higher chemical reactivity than acid etching and mechanical roughening with rotary instruments, s G o o d bond strengths were reported with the use o f hydrofluoric acid alone and with its combinations with other surface roughening methods. 14-18 However, intraoral use o f hydrofluoric acid may eventually be harmful to oral tissues because o f the very aggressive nature o f this acid in required concentrations THE JOURNAL OF PROSTHETICDENTISTRY 181
THE JOURNAL OF PROSTHETIC DENTISTRY
TULUNOGLU AND BEYDEMIR
~
Sbe,aa'~bl~e
F F
D ~
of ~ foro~ ~plied
Fig. 1. Schematic representation of test system.
Table I. Distribution of the specimens through the groups
ME
MATERIAL AND METHOD
ON
SON
P
8
8
M
8
8
SI
P
8
8
CL
M P
8 8
8 8
M
8
8
P
8
8
M
8
8
SC
ON: Resin superstructures were prepared onto the adhesive surfaces; SON: performed resin superstructures were luted with adhesives to the metal or porcelain surfaces; ME: Metabond C&B group; SI: Silistor group; CL: Clearfil Lustre group; SC: Scotchbond Multipurpose Plus group; P: porcelain-resin group; M: metal-resin group.
for etching porcelain surfaces. 19 Thermal and load cycling are important focuses of dental material testing in simulating oral conditions.l,6,11,14,20-25 Silanes are chemical products that can bond to organic materials with their organic endings and to silica materials with their SiO endings.26,27 They are often used for bonding composite resins to porcelain.28, 29 Light-cured composites have better water absorption, wear resistance properties, low polymerization shrinkage, and their manipulation is easier.3°, 31 However, Lai et a132 stated that they found 2% polymerization shrinkage for light-cured composites. In this study, the shear bond strengths of 4 used porcelain repair systems: Metabond C&B (ME), Silistor (SI), Clearfil Lustre (CL), and Scotchbond Multipurpose Plus (SQ) to metal and porcelain were investigated related to the polymerization shrinkage of a resin composite and compared with the ceramometal bond strength (Vita VMK 68 and Remanium CS). 182
One hundred and twenty-eight specimens were prepared for 4 groups, each representing a different repair material. Each repair material was tested fbr bonding resin to a nickel-chromium base metal alloy containing Ni 61%, Cr 26%, SI 1.5%, Mo 11%, S-Fe-Ce-AI lower than 1% (Remanium CS, Dentaurum, Pfbrzheim, Germany); and porcelain (VMK-68, VITA, Schaan, Liechtenstein) and in each subgroup, preformed resin superstructures were luted with adhesive resins to half of the substructures, and resin superstructures were polymerized onto the adhesive painted metal or porcelain substructures fbr the other half of the specimens (Fig. 1 and Table I). Standard cylindrical wax patterns (9 mm diameter [O] and 1 m m thick) were prepared, invested, and cast in Remanium CS alloy. The cast metal substructures were then embedded in autopolymerizing acrylic resin in standard jigs (10 mm diameter and 20 m m in height). Porcelain subgroups were prepared and fired in standard jigs (9 mm diameter and 2 mm height) and after firing, were embedded in autopolymerizing resin in the same jig as were for the metal specimens. Metal surfaces and the porcelain surfaces were sandblasted with 50 g aluminum oxide for 10 seconds at a constant pressure of 0.4 MPa. Metabond C&B (Parkell, Farmingdale, N.Y.) was applied onto the prepared specimen surfhces according to the manufacturer's recommendations: Four drops of liquid were mixed with 1 drop of catalyst, and then mixed with powder and applied onto the specimen surf:aces and air dried for 10 minutes. For the preparation of specimens of Silistor (Kulzer, Wehrheim, Germany) groups, Silicer was applied onto the substructures and air dried for 2 minutes, then a 1:1 mix of Silibond and Dentacolor Opaque was applied onto the metal surVOLUME 83 NUMBER 2
T U L U N O G L U A N D BEYDEMIR
THE J O U R N A L OF PROSTHETIC DENTISTRY
Table II. The base statistical values obtained from shear strength testing ON
ME SI CL SC
P M P M P M P M
SON
Mean (MPa)
SE
SD
Minimum
Maximum
Mean (MPa)
SE
SD
Minimum
Maximum
13.85 34.55 9.99 8.71 12.79 20.57 16.26 9.94
0.82 2.03 1.52 1.03 0.61 0.82 1.55 1.15
2.31 5.74 4.31 2.91 1.72 2.33 4.37 3.04
10.27 24.18 2.45 1.85 9.77 17.05 6.85 6.84
17.21 40.64 14.04 11.10 14.91 23.62 20.03 14.45
14.93 35.27 10.53 9.33 15.60 30.30 20.71 11.74
0.85 2.40 0.86 0.34 1.00 1.19 1.13 0.97
2.40 6.78 2.42 0.97 2.64 3.33 3.18 2.75
12.02 20.02 6.11 7.98 10.59 23.89 14.94 5.42
19.25 40.64 13.42 10.59 18.21 34.28 24.23 13.89
ON: Resin superstructures were prepared onto the adhesive surfaces; SON: performed resin superstructures were luted with adhesives to the metal or porcelain surfaces; M£: Metabond C&B group; SI: Silistor group; CL: Clearfil Lustre group; SC: Scotchbond Multipurpose Plus group; P: porcelainlresin group; M: metalresin group.
Table III. Results of 3-way ANOVA test ((z=.05)
Source
Corrected model Intercept Material groups (ME, SI, CL, SC) Bonding surfaces (P, M) Polymerization schemes (ON, SON) Material groups-bonding surfaces Material group-polymerization schemes Bonding surfaces-polymerization schemes Material groups x bonding surfaces-polymerization schemes Error Tota I Corrected total
Type III sum of squares
9358.948* 37129.7 4029.224 1020.688 227.002 3766.528 161.495 7.137 100.216 1359.957 48225.6 1O718.9
df
Mean square
F
Significance
15 1 3 1 1 3 3 1 3 110 126 125
623.930 37129.7 1343.075 1020.688 227.002 1255.509 53.832 7.137 33.405 12.363
50.467 3003.235 108.634 82.568 18.361 101.552 4.354 0.577 2.702
<.0005 <.0005 <.0005 <.0005 <.0005 <.0005 .006 .449 .049
R2 - 0.873 (adjusted R2 - 0.856/. ON: Resin superstructures were prepared onto the adhesive surfaces; SON: performed resin superstructures were luted with adhesives to the metal or porcelain surfaces; ME: Metabond C&B group; SI: Silistor group; CL: Clearfil Lustre group; SC: Scotchbond Multipurpose Plus group; P: porcelainlresin group; M: metalresin group.
faces and Silibond was applied onto the porcelain surfaces and light polymerizcd for 40 seconds. The porcelain substructure surfaces of Clearfil Lustre (Cavcx, Haarlem, Holland) group were acid etched with phosphoric acid (K-etchant, Kuraray Co Ltd, Osaka, lapan), cleaned with tap water and a mix of Porcelain Activator, Photobond catalyst and Universal (Kuraray Co. Ltd) was applied and air dried for 30 seconds and finally opaque was applied and light cured Ibr 60 seconds. Scotchbond Ceramic Primer was applied on the porcelain surfaces of Scotchbond Multipurpose Plus (3M Dental Products, St Paul, Minn.) group specimens and after air drying, ScotchBond Multipurpose Plus adhesive activator was applied and light cured tbr 60 seconds. The same procedure was followed for the preparation of metal samples, except the Scotchbond ceramic prirncr application. FEBRUARY 2000
All resin superstructures were prepared with Clearfil Lustre composite (Cavex, Holland). Specimens were stored in distilled water at room temperature for 48 hours and then subjected to 500 thermocycles, between 5°C and 55°C with a 20-second dwell time. Shear strength tests were perfbrmed in Lloyd mechanical testing machine (model LR30K, Lloyd Instruments, Farnham, U. K.) with a crosshcad speed of 0.5 ram/rain. Statistical evaluation was per~brmed with 3-way analysis of variance (ANOVA) and Tukey tests. Data were evaluated with 3-way ANOVA after the demonstration of the hypothesis that the error variances of the dependent variable is equal across groups with Levcne's test RESULTS The results of the shear bond strength tests are presented in Table II. The 3-way ANOVA test revealed that 183
THE JOURNAL OF PROSTHETIC DENTISTRY
= 10
;
S~
T U L U N O G L U A N D BEYDEMIR
:!
ME
Ck Tell group=
$C
Fig. 2. Mean shear bond strengths of porcelain surface
groups.
Sl
$C
CL T u t q~oul~
ME
Fig. 4. Mean shear bond strengths of in situ polymerized resin superstructure groups.
8
i SI
SC
CL Telt groups
ME
SI
SC
CL
ME
T u t Group
Fig. 3. Mean shear bond strengths of metal surface groups.
Fig. 5. Mean shear bond strengths of prepolymerized resin superstructure groups.
the differences between bonding system groups (F=108.634, P=.0001), between porcelain and metal groups (F=82.558, P=.0001), and betwecn in situ polymerizcd and prepolymcrized resin superstructure groups werc significant (F= 18.361, P=.0001) (Table III). Interactions between repair system groups and porcelain-and-metal groups (F= 101.552, P=.0001), and fbr interactions between repair system groups and in situ polymerized-and-prepolymerized resin superstructure groups were significant (F=4.354, P=.0001). However, interactions between porcelain-and-metal groups and in situ polymerizcd-and-prcpolymerized resin superstructure groups wcre not significant (F=0.577, P>.0S),and the interactions between repair system groups, porcelain-and-metal groups, and in situ polymerized-and-prcpolymerized resin superstructure groups were significant (F=2.702, P=.05). When Clearfil Lustre was used on metal surfaces, the differences between the in situ polymerized-andprepolymerized resin superstructure subgroups were statistically significant (P=.0117). Also, when Scotchbond Multipurpose Plus was used on porcelain surfaces, the differences between the in situ polymerizedand-prepolymerized resin superstructure subgroups were statistically significant (P=.0251 ). The differences between the in situ polymerized-and-prepolymerized
resin superstructure subgroups of the other groups were not statistically significant (P>.05) (Figs. 2 through 5).
184
DISCUSSION The renewal of the fracturcd ccramomctal restorations causes a waste of time and economical problems. Recently, many ceramomctal restoration repair systems and materials were developed to overcomc this important problem. A noble metal alloy could be included to this study. However, this seemed to overcrowd the parameters and complicate the study too much to interpret. Nickel-chromium alloys are widely used in the fabrication of ceramometal restoration framework. Adhesive resins show better bonding properties to base metal alloys because of the metal oxides they can easily form at their surfaces. 1,2 Adhesive materials containing 4-META had been shown to havc good bonding properties to base metal alloys. 3-5 It has also been shown that metal oxides, especially chromium oxide plays an important role at the bond between 4-META containing adhesive materials and metal alloys. 3,4 The best results among the metal surface subgroups of the 4 materials tested were obtained from Metabond C&B group, and this was in accordance with the findings of these previous studies. VOLUME 83 NUMBER 2
T U L U N O G L U A N D BEYDEMIR
The porcelain resin bond strengths obtained from Metabond C&B group (in situ polymerized, 13.85 _+2.31 MPa; prepolymerized, 14.93 _+21.93 MPa) are differing from the findings of Kato et al. 1 This may be because they have used a different experimental design in which they bonded 2 identical porcelain disks to each other with adhesive resins. Appeldoorn et al27 evaluated bond strength of composite to porcelain after 3 months of storage in distilled water, followed by 2500 thermocycles; they demonstrated that relatively high bond strengths were achieved with sandblasting specimens. Kato et al 1 reported that porcelain-to-porcelain bond achieved by use of porcelain liner M and Superbond C&B porcelain liner exhibited bond strengths greater than 20 MPa after thermocycling. Unfortunately, they did not included any of the other systems investigated in this study. Porcelain Liner M contains a silane coupler and a carboxylic compound (4-META), the combination can activate the silane. In their study, primers that exhibited durable bond strengths were not a prehydrolyzed type but a 2 or 3 component type. This type of primer does not use an aqueous solution during silane treatment and has a prolonged shelf-life. Acid activators have hydrophobic structures in the compound and also have curing ability. It is probable that these factors reciprocally enhanced the bond to porcelain. Furthermore, the best results in our study were obtained with such materials. Suliman et al28 investigated the effects of hydrofluoric acid-etching, roughening with diamond burs and hydrofluoric acid-etching, and roughening with diam o n d burs and sandblasting, and reported that the most effective surface treatment for porcelain-resin bond was mechanical roughening with diamond burs, followed by chemical etching with hydrofluoric acid. They used an intraoral sandblasting device and 50 gm aluminum oxide and 15-second application time. Although they did not report the sandblasting pressure, their findings and conclusions were different from the findings of Kato et al, 1 who reported that light sandblasting provided the best results. Appeldoorn et a127 evaluated the bond strength of composite to porcelain after 2 months water storage, fbllowed by 2500 thermocycles. Their results demonstrated that relatively higher bond strengths were achieved with sandblasted specimens. Their results differed from those of Bertolotti et al,8 who reported higher bond strengths with air abrasion than with diamond roughening. Llobell et alll preferred to use 8% hydrofluoric acid application versus sandblasting because even at light pressure, mitraoral air-abrasive spraying procedures may damage the remaining glazed porcelain surfaces unless careful isolation is added to the recommended guidelines. However, intraoral use of hydrofluoric acid may eventually be harmful to oral tissues because of the FEBRUARY 2000
THE JOURNAL OF PROSTHETIC DENTISTRY
aggressive naturc of this acid in requircd concentrations for etching porcelain surfaccs. 19 Load and thermal cycling arc important focuses of dental material testing in simulating oral conditions. Opinions vary regarding the effects of thermal cycling on the bond strength of porcelain repair systems. Most authors have suggested that bond strengths of all the tested materials were negatively affected from thermal cycling,l,6,14, 20-22 and some stated that therc were no significant differcnces in bond strengths of some of the materials they testcd.18, 26 The detrimcntal effect of thermocycling on the porcelain-resin or metal-rcsin bond is based on the differcnce of thcrmal expansion coefficients. 23-2~ In our study, as the effects of different amounts ofthermocyclcs were not cvaluated, an application of 500 thermocyclcs was chosen as this amount seemed to represent a mean value of the thermocycling applications in previous studies. However, there is a need of well-established scales, indicating the amount of thermocycles corresponding to a certain period of intraoral use of different materials and their combinations. Another important application in simulation of oral conditions is fatigue test and load fatigue may be defined as a phenomcnon, whercby a sample that has been repeatedly subjected to load well below the level that causes fracture in static tensile tests eventually fails after being subjected to this comparatively small cyclic load. This fatiguc failure is preceded by a combination of crack initiation and crack propagation, followed by a failure that occurred in the form of a fracture. Llobell et al ll state that fatigue tests can be a more significant test as it can simulate long-term clinical use. In our study, the effect of polymerization shrinkage of the composite resulted in significant differences fbr Clearfil Lustre-metal (P=.00117) and Scotchbond Multipurpose Plus-porcelain (P=.0251) groups. These results indicate that better results can be obtained with the indirect preparation of the resin superstructure when these repair materials are used. There are no reports on the minimum shear bond strength value required for the ceramometal restoration repair materials. However, one can easily assume that this value must be near or more than the metal-porcelain bond strength. In our study, higher values than the control group were obtained from in situ polymerized resin superstructures onto metal surfaces coated with Metabond C&B, prepolymerized resin superstructures luted onto metal surfaces with the use of Metabond C&B, in situ polymerized resin superstructures onto metal surfaces coated with Clearfil Lustre, prepolymerized resin superstructures luted onto metal surfaces with the use of Clearfil Lustre, and prepolymerized resin superstructures luted onto metal surfhces with Scotchbond Multipurpose Plus groups. These results indicate that the use of Metabond C&B where metal is exposed provided the best results for both resin super185
THE JOURNAL OF PROSTHETIC DENTISTRY
structure preparation techniques and that the best results could be obtained with the use of Scotchbond Multipurpose Plus in situations where fracture is limited to porcelain. CONCLUSIONS Within the limits of this study, the following conclusions were drawn: 1. Higher bond strength values were obtained with prepolymerized superstructures compared to in situ polymerized superstructures. 2. Metabond C&B provided the best results tbr both prepolymerized and in situ polymerized resin superstructure preparation techniques at the failures where metal is exposed. 3. The best results when the fracture was limited into porcelain were obtained with Scotchbond Multipurpose Plus porcelain liner. However, in vivo and in vitro tests are required before a final judgment is made. REFERENCES 1. Kato H, Matsumura H, Tanaka T, Atsuta M. Bond strength and durability of porcelain bonding systems. J Prosthet Dent 1996;75:163-8 2. Crim GA, Garcia-Godoy F. Microleakage: the effect of storage and cycling duration. J Prosthet Dent 1987;57:574-6. 3. Taira Y, Yoshida K, Matsumura El, Tanaka T, Atsuta M. Adhesive metal primers for a light cured opaque resin. In: Advanced Prosthodontics Worldwide, Proceeding on World Congress on Prosthodontics held in Hiroshima, 1991. Hiroshima, Japan: Publication Committee; September 1991. p. 217. 4. Barzilay I, Myers ML, Cooper LB, Graser GN. Mechanical and chemical retention of laboratory cured composite to metal sudaces. J Prosthet Dent 1988;59:131-7. 5. Cooley RL, Tseng EY, EvansJG. Evaluation of 4-META porcelain repair system. J Esther Dent 1991 ;3:11-3. 6. Beck DA, Janus CE, Douglas HB. Shear bond strength of composite resin porcelain repair materials bonded to metal and porcelain. J Prosthet Dent 1990;64:529-33. 7. }ochen DG, Caputo AA. Composite resin repair of porcelain denture teeth. J Prosthet Dent 1977;38:673-9. 8. Bertolotti RL, Lacy AM, Watanabe LG. Adhesive monomers for porcelain repair. Int J Prosthodont 1989;2:483-9. 9. Simonsen RJ, Calamia JR. Shear bond strength of composite bond to etched porcelain. J Dent Res 1983;62:297 (abstract). 10. Stangel I, Nathanson D, Hsu CS. Shear bond strength of composite bond to etched porcelain. J Dent Res 1987;66:1460 5. 11. Llobe]l A, Nicho[ls JI, Kois JC, Da]y CH. Fatigue life of porcelain repair systems. Int J Prosthodont 1992;5:205-13. 12. Lacy AM, Laluz J, Watanabe LG, Dellinges M. Effect of porcelain surface treatment on the bond to composite. J Prosthet Dent 1988;60:288-91. 13. Kern M, Thompson VP. Bonding to glass infiltrated alumina ceramic: adhesive methods and their durability. J Prosthet Dent 1995;73:240-9.
186
TULUNOGLU AND BEYDEMIR
14. Thurmond JW, Barkmeier WW, Wilwerding T. Effect of porcelain surface treatments on bond strengths of composite resin bonded to porcelain. J Prosthet Dent 1994;72:355-9. 15. Roulet JF, S6derholm KJ, Longmate I. Effects of treatment and storage conclitions on ceramic/composite bond strength. J Dent Res 1995;74:381-7. 16. Ozden AN, Akaltan E, Can G. Effect of surface treatments of porcelain on the shear bond strength of applied dual-cured cement. J Prosthet Dent 1994;72:85-8. 1 7. Aida M, Hayakawa T, Mizukawa K. Adhesion of composite to porcelain with various surface conditions. J Prosthet Dent 1995;73:464-70. 18. Kupiec KA, Wuertz KM, Barkmeier WW, Wilwerding TM. Evaluation of porcelain surface treatments and agents for composite-to-porcelain repair. J Prosthet Dent 1996;76:119-24. 19. Fan PL. Porcelain repair materials. Council on Dental Materials, Instruments, and Equipment. J Am Dent Assoc 1991 ;122:124-30. 20. Creugers NH, Snoek PA, Kayser AF. An experimental porcelain repair system evaluated under controlled clinical conditions. J Prosthet Dent 1992;68:727-7. 21. Gregory WA, Hagen CA, Powers JM. Composite resin repair of porcelain using different bonding materials. Oper Dent 1988;13:114-8. 22. Pratt RC, Burgess JO, Schwartz RS, Smith JH. Evaluation of bond strength of six porcelain repair systems. J Prosthet Dent 1989;62:11-3. 23. Newburg R, Pameijer CH. Composite resins bonded to porcelain with silane solution. J Am Dent Assoc 1978;96:288-9. 24. Eames WB, Rogers LB, Feller PR, Price WR. Bonding agents for repairing porcelain and gold: an evaluation. Oper Dent 1977;2:118-24. 25. Nowlin TP, Barghi N, Norling BK. Evaluation of the bonding of three porcelain repair systems. J Prosthet Dent 1981 ;46:516-8. 26. Diaz-Arno[d AM, Schneider RL, Aqui[ino SA. Bond strengths of intraoral porcelain repair materials. J Prosthet Dent 1989;61:305-9. 27. Appeldoorn RE, Wilwerding TM, Barkmeier WW. Bond strength of composite resin to porcelain with newer generation porcelain repair materials. J Prosthet Dent 1993;70:6-11. 28. Suliman AH, Swift EJJr, Perdigao J. Effects of surface treatments and bonding agents on bond strength of composite resin to porcelain. J Prosthet Dent 1993;70:118-120. 29. Bailey JH. Porcelain-to-composite bond strengths using four organosilane materials. J Prosthet Dent 1989;61:174-7. 30. Dennison JB. Status report on microfilled composite restorative resin. Council on Dental Materials, Instruments, and Equipment. J Am Dent Assoc 1982;105:488-92. 31. Lutz F, Phillips RW. A classification and evaluation of composite resin systems. J Prosthet Dent 1983;50:480-8. 32. Lai JH, Johnson AE. Measuring polymerization shrinkage of photo-activated restorative materials by a water-filled dilatometer. Dent Mater 1993; 9:139-43. Reprint requests to: DR [BRAHIM F. TULUNOGLU BAHCELIEVLER15 SOK. 41/1 06490 ANKARA TURKEY Fax: (312)3113741 Copyright © 2000 by The Editorial Council of The Journal of Prosthetic Dentisto4 0022-3913/2000/$12.00 + 0. 10/1/103750
VOLUME 83 NUMBER 2