Bond strength of orthodontic composite cement treated porcelain

to

Devin Cochran, DDS, MS, a Kathy L. O'Keefe, DDS, b David T. Turner, DDS, MS, c and John M. Powers, PhD a San Antonio and Houston, Texas

A porcelain-fused-to-metal ceramic was prepared for bonding by five treatments: sandblasting, sandblasting and silanating, hydrofluoric acid etching, hydrofluoric acid etching and silanating, and 600-grit polishing and silanating. Two commercial, all-purpose bonding agents were used to bond a composite cement to the porcelain samples. In vitro tensile bond strengths were compared with samples for which no bonding agent was used. Highest bond strengths (22 to 41 MPa) were obtained, with one exception, when the porcelain surface was silanated; however, the use of silane increased the occurrence of porcelain fracture on debonding. Composite cement bonded without bonding agent to nonsilanated porcelain prepared by sandblasting or etching with hydrofluoric acid had bond strengths of 6.5 MPa and 18 MPa, respectively, with all bond failures at the bracket/composite interface. The use of all-purpose bonding agents and silanating agents may not be necessary for adequate orthodontic direct bonding. (Am J Orthod Dentofac Orthop 1997;111: 297-300.)

T h e number o f patients seeking o r thodon- tic t rea tment has expanded to include more adults than ever before. Direct bonding of or thodont ic a t tachments has removed some of the esthetic con- cerns many adults previously had when cOnsidering or thodont ic therapy.

With an increase in adult t rea tment comes the challenge of direct bonding to restored surfaces, such as porcelain. Porcelain does not respond to the etching process (application of 37% phosphor ic acid) used for mechanical re tent ion on an enamel surface. To gain the mechanical retent ion necessary for bonding, the surface must be roughened chem- ically by etching with 9.6% hydrofluoric acid, or mechanically by d iamond bur or sandblasting. In addition, the use of silane as a pr imer has been shown to increase the bond strength of composi te to porcelain. 1-a

The purpose of this study was to determine the in vitro tensile bond strengths and failure locations of an or thodont ic composi te cement bonded in conjunct ion with an all-purpose bonding agent to a

From the University of Texas-Houston Science Center, Dental Branch. aIn private practice, San Antonio, Texas. 1'Associate professor, Department of Prosthodontics. CAssistant professor, Department of Craniofacial Growth and Develop- ment. aProfessor, Department of Basic Sciences; Director, Houston Biomaterials Research Center. Reprint requests to: Dr. Jobn M. Powers, Biomaterials Research Center, UT-Houston Dental Branch, 6516 John Freeman, Houston, TX 77030- 3402. Copyright © 1997 by the American Association of Orthodontists. 0889-5406/97/$5.00 + 0 8/1/70216

porcelain substrate p repared by five treatments: sandblasting, sandblasting and silanating, hydroflu- oric acid etching, hydrofluoric acid etching and silanating, and 600-grit polishing and silanating.

MATERIALS AND METHODS

Two all-purpose bonding agents with no bonding agent as a control were tested for their effect on in vitro tensile bond strength of a direct bonding composite cement to porcelain with five surface preparations (600- grit polishing and silanating, sandblasting, sandblasting and silanating, etching, and etching and silanating). The combination that gave the highest bond strength was then tested by using metal and ceramic brackets.

Ninety porcelain (Ceramco II, Body Shade A2, #9312019, Ceramco Inc.) disks, approximately 5 mm thick and 8 mm in diameter, were prepared from a polytetra- fluoroethylene die and fired in a vacuum oven (Jelenko Auto LTIIVPF, Jelrus Technical Products Corp.). The firing cycle was as follows: dry, 5 minutes; preheat, 5 minutes; low set temperature, 593 ° C; high set tempera- ture, 951 ° C; and rate of temperature increase, 10 ° C per minute under a vacuum of 71 cmHg. Samples were inspected under magnification for surface defects and embedded in mounting plastic with the porcelain surface exposed. The porcelain samples were then ground smooth and flat on a metallurgical wheel using 240-, 320-, 400-, and 600-grit silicon carbide paper (Buehler Ltd.). The final thickness of the samples was approximately 4 ram.

The 90 samples were pumiced lightly, rinsed 10 sec- onds, dried, and divided into five groups of 18 samples each. Two groups were sandblasted (Micro-Etcher, Dan-

297

2 9 8 Cochran el al. American Journal of Orthodontics and Dentofacial Orthopedics March 1997

40

35

30

}2s =* cn

o

15

6[~rit/Silone SQndb~c~ted Sondblo, stedlSilane Etched Etched/Sik:ine

Tmalmenl

Fig. 1. Bond strength (MPa) of composite cement bonded to porcelain with five surface treatments of porcelain and two all-purpose bonding agents, Optibond (OP), Scotchbond MP (SB), and no bonding agent (none).

Table I. Bond strength (MPa) of composite cement to treated porcelain using five surface treatments, two bonding agents, and a control in which no bonding agent was used

t Adhesive Agent

Treatment None SB

600-grit finish and silanated 28.3 (7.6)* 13.5 (4.3) 28.1 (6.3) Sandblasted 6.5 (1.4) 4.0 (1.9) 4.5 (1.7) Sandblasted and silanated 39.1 (4.8) 21.8 (4.5) 41.3 (8.7) Etched 17.8 (2.2) 10.2 (3.0) 16.1 (1.5) Etched and silanated 35.5 (7.2) 23.2 (3.8) 34.5 (9.7)

*Means of six replications with standard deviations in parentheses. Tukey- Kramer intervals calculated at the 0.05 significance level for comparisons among five treatments and three adhesive agents were 5.0 and 3.3 MPa, respectively.

ville Engineering) to a frosty surface with 50 tzm alumi- num oxide particles at a pressure of 0.48 MPa at 9 L per minute for 3 seconds, rinsed 15 seconds, and dried. After sandblasting, the samples were cleansed for 60 seconds with 40% phosphoric acid (3M Dental Products) and then rinsed for 30 seconds with an air/water spray. Two other groups of samples were etched with 9.6% hydrofluoric acid gel (Ceram-Etch, #3H073, Gresco Products, Inc.) for 150 seconds, rinsed 15 seconds, and dried. The fifth group, along with one each of the sandblasted and the etched groups, were pretreated over the entire surface with a silanating agent (Scotchprime, #3CR, 3M Dental Prod- ucts).

After surface preparation, each of the groups was further divided into three subgroups of six samples each. Two subgroups were pretreated over the entire surface and blown to a thin film with oil-free compressed air, with one of two commercial all-purpose bonding agents (OP, Optibond, #24638/754605, Kerr Manufacturing; or SB, Scotchbond MP, 3BU, 3M Dental Products). Each was light-cured according to manufacturers' directions using a visible light-curing unit (Optilux 400, Demetron Research Corp.). The final subgroup was not treated with a bonding agent.

A two-paste composite cement (Concise Direct Bond- ing Cement, paste A-1961A, paste B-1961B, resin A-1922A, resin B-1922B, 3M Dental Products) was then bonded to the porcelain samples following the manufac- turer's directions. The use of the unfilled resin was limited to those samples that were not treated with either all- purpose bonding agent. The bonded sample was an in- verted, truncated cone with a bonding diameter of 3 mm at the bond interface and a diameter of 5 mm at a height of 5 mm. 5 The composite cones were made using a polytetrafluoroethylene mold on a bonding jig.

After storage in water at 37°C for 24 hours, the bonded cones were debonded in tension using a jig 5 on a testing machine (Model 8501, Instron Corp.) at a cross- head speed of 0.05 cm per minute. The bond strength in megapascals (MPa) was determined by dividing the load at debonding by the area of bonding.

The site of fracture was determined by visual inspec-

American Journal of Orthodontics and Dentofacial Orthopedics Cochran et al. 299 Volume 111, No. 3

Table II. Fa i l u r e s i tes o f c o m p o s i t e c e m e n t b o n d e d to t r e a t e d p o r c e l a i n us ing five su r f ace t r e a t m e n t s a n d t h r e e b o n d i n g a g e n t s

Treatment C

600-grit finish and silanated 3 0 Sandblasted 0 6 Sandblasted and silanated 5 0 Etched 0 6 Etched and silanated 6 0

None

CP

Adhesive Agent

3 2 4 0 3 1 2 0 0 6 0 0 6 0 1 5 0 1 6 0 0 0 0 6 0 0 6 0 0 6 0 0 5 0 1

*C is failure within cement, CP is failure at the cement/porcelain interface, and P is failure within porcelain.

tion and recorded. Fracture sites were defined as occur- ring cohesively within the cement (C), at the cement/ porcelain interface (CP), or cohesively within the porcelain (P).

Means and standard deviations were determined from six replications for each condition. Data were analyzed statistically by two-way analysis of variance (Super- ANOVA, Abacus Concepts) with a factorial design. Means were compared by Tukey-Kramer intervals (Super- ANOVA) calculated from the analysis of variance at the 0.05 level of significance. Differences between two means that were greater than the Tukey-Kramer interval were considered statistically different.

Ten porcelain (Lumina Twin, 4B24, Ormco Corp.) and 10 stainless steel (Mini-Diamond Twin, 4A140A) brackets were then bonded to sandblasted and silanated porcelain surfaces treated with SB. The brackets were for a maxillary left central incisor with slot dimensions of 0.018 x 0.026 inches. A 25 mm length of 0.018 × 0.025-inch stainless steel wire was inserted into the bracket slot and tied with two 0.012-inch stainless steel ligature wires. These wires were used to pull the bracket from the base in tension by using the testing machine. The bond strength was determined from the load at failure and the measured area of the bracket. The site of fracture was recorded and classified as occurring cohesively within the porcelain, at the cement/porcelain interface, cohesively within the cement, at the cement/bracket interface, or cohesively within the bracket.

RESULTS

The data for bond strengths and failure sites are listed in Tables I and II and shown in Fig. 1. Analysis of variance showed significant differences in bond strength with surface treatment (p = 0.0001) being the most important factor, followed by the adhesive (p = 0.0001) used. Tukey-Kramer intervals at the 0.05 significance level for comparisons among five porcelain treatments and among three bonding agents were 5.0 and 3.3 MPa, respectively.

The highest bond strengths (22 to 41 MPa) were obtained with one exception (OP, 600-grit

finish and silanated surface, 13.5 MPa) when the porcelain surface was silanated. Sandblasting without the use of silane showed the weakest bond strengths (4.0 to 6.5 MPa) with or without bond- ing agent. Samples prepared with hydrofluoric acid etching had intermediate values of bond strength (10.2 to 17.8 MPa).

Cement/porcelain interracial failures were ob- served in all the hydrofluoric acid and sandblasted groups where silane was not used. When silane was used for these two surface treatments, the bond failure occurred within the cement 92% of the time. The other 8% fractured cohesively within the porcelain. Failure sites in the 600-grit and silanated group were more variable and in- cluded the highest percentage of cohesive porce- lain failures (28%).

Metal and porcelain brackets were bonded with bonding agent SB to sandblasted and silanated porcelain samples. Means and standard deviations of bond strengths of metal and porcelain brackets to porcelain were 5.0 (0.8) and 9.9 (2.6) MPa, respec- tively. Bond failure locations were 100% at the cement/bracket interface for metal brackets and 90% within the bracket and 10% at the cement/ bracket interface for porcelain brackets.

DISCUSSION

This in vitro investigation showed significant differences in tensile bond strengths of two all- purpose bonding agents to five treated porcelain surfaces. Although samples treated with bonding agent SB had higher bond strengths than OP for the five surface preparations tested, the SB samples were statistically equal to the samples that received no adhesive. These results coincide with the r e s u L of a previous study 6 that showed high tensile bond strengths (23 MPa) of a composite to sandblasted and silanated porcelain. The bonding agent OP

Cochran el al. American Journal of Orthodontics and Dentofacial Orthopedics March 1997

decreased the bond strength otherwise gained by sandblasting or hydrofluoric acid etching. Perhaps OP would have performed better if the porcelain primer (silane) designed for OP had been used.

Silanating agents appeared to have the greatest effect on bond strength in all groups, with an increase of 450% to 820% in sandblasted groups and 100% to 130% in etched groups when silane was used. These findings are in agreement with previous studies 7'8 that found increased bond strengths when silanating agents were used. However, fracture sites within the porcelain occurred 15% of the time when silanating agent was used. Of these, 63% were in the 600-grit finish and silanated group. Increased frac- ture of the chemically and mechanically treated porcelain may be the result of surface weakening. The use of silane without bonding agent appeared to weaken the porcelain more. There is also some evidence to suggest that some dental porcelains fracture more easily than others. 9

In comparison of sandblasted surfaces to sur- faces etched with hydrofluoric acid when no silane was used, the etched samples performed signifi- cantly better. This result may be due to deep etching of the porcelain creating greater mechanical reten- tion for the unfilled resin. Although sandblasting was the quickest and easiest method of roughening the porcelain in vitro, this procedure may be difficult to perform intraorally. Hydrofluoric acid etching could be accomplished along with phosphoric acid etching of enamel for other teeth to be bonded, but would require a rubber dam because of the potential harmful effects of hydrofluoric acid intraorally.

Mean bond strengths for bonded brackets to natural teeth are well below all values for the composite-porcelain bonding combinations where silane was used. The majority of bond failure oc- curred at the bracket/composite interface. Previous studies are in agreement with these results to the extent that where the bond to porcelain is stronger than the bond to the bracket, the failure sites occurred as stated in these studies. 1°-12

Clinically, bond strengths in the 7 to 10 MPa range or higher with no porcelain fracture on debonding would be acceptable. This range includes all groups tested except the sandblasted without

silane group. Consequently, etching of porcelain restorations with hydrofluoric acid avoids the ex- pense of an all-purpose adhesive and a silanating agent. This group also had no porcelain fracture. A previous study 6 reported that increased fracture of the porcelain occurred with hydrofluoric acid with an etching time of 5 minutes. For this reason, the etching time in this study was reduced to 21/2 min- utes.

CONCLUSION

It is evident from this study that bonding and silanat- ing agents achieve very high in vitro bond strengths for orthodontic composite cements to porcelain. However, with this technique, there is some risk for porcelain fracture. At this time, it does not appear to be necessary or desirable to use either an all-purpose adhesive or silanating agent to get clinically acceptable orthodontic bonds to porcelain when using the composite cement tested. Assuming that the porcelain restoration does not fracture at the debonding appointment, the safest method for composite removal would be to remove the gross excess with a diamond or carbide bur, followed by con- touring and polishing the remaining composite with sand- paper disks.

We thank the manufacturers for providing the com- mercial products.

REFERENCES

1. Newman GV. Bonding to porcelain. J Clin Orthod 1983;17:53-5. 2. Matsumura H, Kawahara M, Tanaka T, Atsuta M, A new porcelain repair system

with a silane coupler, ferric chloride, and adhesive opaque resin. J Dent Res

1989;68:813-8. 3. Bertolotti RL, Lacy AM, Watanabe LG. Adhesive monomers for porcelain repair.

Int J Prosthod 1989;2:483-9. 4. Pratt RC, Burgess JO, Schwartz RS, Smith JH. Evaluation of bond strength of six

porcelain repair systems. J Prosthet Dent 1989;62:11-3. 5. Barakat MM, Powers JM. In-vitro bond strength of cements to treated teeth. Aust

Dent J 1986;31:415-9. 6. Wolf DM, Powers JM, O'Keefe ILL. Bond strength of composite to porcelain

treated with new porcelain repair agents. Dent Mater 1992;8:158-61. 7. Newman SM, Dressler KB, Grenadier MR. Direct bonding of orthodontic brackets

to esthetic restorative materials using a silane. Am J Orthod 1984;86:503-6. 8. Smith GA~ Mclnnes-Ledouw WR, Weinberh R. Orthodontic bonding to porcelain-

bond strength and refinishing. Am J Orthod Dentofac Orthop 1988;94:245-52. 9. Kao EC, Johnston WM. Fracture incidence on debonding of orthodontic brackets

from porcelain veneer laminates. J Prosthet Dent 1991;66:631-7. 10. Dickinson PT, Powers JM. Evaluation of fourteen direct-bonding orthodontic

bases. Am J Orthod 1980;78:630-9. 11. Buzzitta VA, Hallgren SE, Powers JM. Bond strength of orthodontic direct-

bonding cement-bracket systems as studied in-vitro. Am J Orthod 1982;81:87-92. 12. Evans LB, Powers JM. Factors affecting in-vitro bond strength of no-mix orthodon-

tic cements. Am J Orthod 1985;87:508-12.