Design

and

resistance

technique

variables

of metal-ceramic

affecting

fracture

restorations*

Walter S. Warpeha, Jr., D.D.S., M&D.,** Richard J. Goodkind, D.M.D., M.S.*** Uniuersity of Minnesota, School of Dentistry,

and Minneapolis,

Minn.

U

nexplained clinical failures still occur after 20 years of refining the metalCareful clinical observation has revealed ceramic system for dental restorations. several types of failures. Three of the most common types are cleavage through the porcelain-metal interface, fracture through the opaque or body porcelain, and crazing of the surface of the restoration. These fractures are associated with additional patient trauma, frustration of the dentist, and increased office time and expense. Therefore, there was a need to investigate some of the complex variables which may contribute to fracture of metal-ceramic restorations. In this study, several variables in metal design and porcelain manipulation were examined in relation to the fracture resistance of a clinical model. LITERATURE

REVIEW

The terms “porcelain fused to gold or metal,” “porcelain baked on gold or metal,” as well as “ceramo-metallic” have been used previously to describe this type of restoration. Phillips1 stated that the correct term should be “metal-ceramic” restoration which better describes this composite structure. Research on the metal-ceramic restoration has been overwhelmingly concerned with evaluating bond strength. Shell and Nielsen2 designed a test to measure the shear strength of the bond. Several other studies using somewhat similar testing procedures compared bond strength values of various metal preparations.3-5 Yet these studies did not relate bond strength values with clinical fracture resistance. Many dental articles have empirically suggested metal design and preparation based mostly on industry’s experience with enameling systems, yet there are few Read before the American

Prosthodontic

*Condensed from a thesis presented of Master of Science from the University “*Clinical ***Associate

Assistant

Professor,

Professor,

Director

Society,

Chicago,

in partial fulfillment of Minnesota.

Division

111. of the requirements

for the degree

of Prosthodontics.

of Graduate

Prosthodontics.

291

292

Warpeha

and Goodkind

Fig. 1. Direction crown form.

of force applied

J. Prosthet. Dent. March. 1976

to a maxillary

premolar

clinically applicable studies. Nally6 found that twice the impact strength of a porcelain jacket by dropping a metal ball from various heights. ceramic restorations combine the strength of the properties inherent in vacuum-fired porcelain. METHODS

AND

tooth and design rationale

for the

a metal-ceramic crown exhibited restoration. Impact was produced Clinically, it appears that metalnoble metals and the fine esthetic

MATERIALS

To assess clinical strength of a metal-ceramic crown, a mechanical model must be designed which approaches the clinical form of a tooth. The model must be loaded to duplicate those forces found in mastication, and this load should be applied at a site comparable to an occlusal contact. The action of a maxillary against a mandibular premolar was analyzed in working movement, and the force was diagramed (Fig. 1). A die was designed so that porcelain could be added to the metal substructure to complete the fabrication of the maxillary premolar cusp form. In addition, this die would serve as a base to hold the crown form at a 30 degree angle so that the cusp incline would be perpendicular to the testing load. The dies were solid castings and were reproduced by use of wax patterns obtained from a rubber mold. Nally, Farah, and Craig7 found up to ten times more tensile stress in uncemented samples subjected to compressive testing. By use of a solid die, the variability of such factors as coping fit and metal deformation was reduced. The castings were made of Ceramco O* alloy using Ceramigold” investment burned out at 1,300° F. for one hour. Three designs were fabricated from three separate molds. The castings were identical in all dimensions except for the acuteness of the metal under the cusp tip (Fig. 2). The body and opaque porcelain was applied and fired after the metal castings were machined to 0.1 mm. tolerance, and the surfaces were prepared in various ways according to group. Some groups were *Ceramco,

Inc., Long Island City, N. Y.

.

Factors

Table

I. Average

fracture

fracture

5

5 5 4 3 2

293

(in pounds)

1

2

3

Group description

347 401 400 395

359 356

249 328

Fine stone, no metal coating Fine stone, metal coating Coarse stone, metal coating Coarse stone, no metal coating Fine stone, nonoxidized Reiected samples

153 150

II. Analysis

of restorations

Design

No. of samples for each design

Table

strength

affecting

of variance

Source of variability

1

d!

1

SS

Problem I. Three designs with and without metal coating-30

Factor A (coated or uncoated) Factor B (three designs) A-B interaction Error Total

1 2 2 24 29

14083 41145 8832 66440 130500

MS observations

14083 20573 4416 2768

Problem 2. Design No. 1 with two surface roughnesses, coated or uncoated-19

Factor A (coated or uncoated) Factor B (coarse or line) A-B interaction Error Total *Not significant.

1 1 I 15 18

4521 2298 2825 31850 41874

p value

F

4521 2298 2825 2123

5.09 7.43 1.59

0.025

o.10* p>o.10* p > 0.10*

prepared with a fine aluminum oxide stone,* while others were prepared with a coarse silicon carbide stone.? All samples were degassed except for the nonoxidized group (Table I). Half of the samples were coated with a metal conditi0ner.S ‘All samples received identical applications of baked porcelain which were ground to the same tolerances as the metal substructure. This included machining of the top of the sample to assure a smooth, flat surface perpendicular to the testing apparatus. The samples were glazed after being examined for flaws in the porcelain under 3x magnification. TESTING

OF THE SAMPLES

Forty-four samples were compressed in an Instron loading machines until gross fracture occurred (Fig. 3). The loading stylus had a beveled point 1 mm. in diameter and was rigidly attached to the load cell. The base of the sample was securely seated into a plate which was attached to the crossbar of the Instron machine. “Dura-White REl, Shofu Dental Corp., Menlo Park, Calif. tDura-Green IC2, Shofu Dental Corp., Menlo Park, Calif. SBritecote, Ceramco, Inc., Long Island City, N. Y. BInstron type II, Instron Engineering Corp., Canton, Mass.

Fig. 2. Solid castings representing Fig. 3. Test sample fracturing

the three designs selected for testing

under loading

stylus of Instron

(top and side views).

machine.

The stylus was lined up to contact each sample in the same place as indicated by a marked point which was centrally located 5 mm. from the posterior metal buttress (Fig. 4). This point simulated a working contact on an upper buccal cusp which is a common fracture site observed in clinical restorations. The load was applied at the rate of 0.05 inch per minute, and a load deflection curve was recorded on the machine. Ultimate fracture was recorded as a major deflection of the graph, and in nearly all cases, it coincided with visible gross fracture.

RESULTS Ultimate fracture strength varied between 140 and 460 pounds of pressure among the samples tested, even though the fracture patterns were similar for all designs (Fig. 5). Average fracture strengths are listed by group in Table I. The results were statistically evaluated by use of an analysis of variance (Table II) . The first problem analyzed in this table was the strength of the three designs. Half of the samples were prepared with a surface metal conditioner, and the remaining samples were not treated. The second analysis involved design No. 1 dies which were prepared for porcelain application with either a fine or a coarse stone. Half of each group again received a metal coating. Of the three metal designs, irrespective of metal coating, No. 3 (the most acute) fractured with significantly less pressure (p < 0.005). The roundest (design No. 1) gave the highest values. Therefore, design No. 1 was further tested to evaluate the effect of surface roughness with and without the use of a surface conditioner. Application of a metal coating agent to a dull matte finish before firing the porcelain significantly altered the strength in several groups. Over-all, those groups of all designs prepared with a fine stone and coated with the coating agent showed increased strength. The dies of design No. 1 which were prepared with a coarse stone showed no significant difference whether coated or not. The most acute design, design No. 3, was particularly affected by the metal coating (increased strength), while design Nos. 1 and 2 did not show such clear distinction. A thick application of the metal coating which produced a “24 karat gold” surface severely reduced the strength of one sample.

Factors

affecting

fractw

Fig. 4. Completed die marked with black dot indicating point of loading. Fig. 5. Typical similar fracture patterns observed in design Nos. 1, 2, and 3.

Different degrees of surface roughness produced by a fine (aluminum oxide) stone or a coarse (silicon carbide) stone did not significantly alter the fracture strength. However, because of its contamination, the silicon carbide produced bubbles in the porcelain. In one sample, when the bubbling occurred directly below the loading stylus, the strength of the test sample was severely decreased. No significant differences were found among the various preparations of design No. 1 (roughness-coating interaction) with the sample sizes used in the investigation. Ultimate fracture strength was also severely reduced in the group with design No. 2 where the samples were not “degassed” and, therefore, not oxidized before porcelain application.

DISCUSSION The fracture strengths exhibited in this study seemed to be consistent with maximum biting pressures as reported by K1affenbach.s He found that only a small percentage of people can register more than 275 pounds of pressure on one posterior tooth. In an individual who could place this pressure on a metal-ceramic restoration, nine of the 44 samples tested in this study would have failed. Many of these failures would have corresponded to faulty laboratory techniques, i.e., no degassing and/or shiny Britecote. Designs. The most acute metal design failed at significantly less pressure than the two more rounded designs. This is consistent with the general principle of avoiding sharp angles under the porcelain. Normally, the supporting metal acts as a stress distributor.’ The decreased strength might have resulted from an increased stress concentration around the acute angle and smaller porcelain cross-sectional area of design No. 3. Clinically, this would encourage care in tooth preparation and metal design so that no acute angles are formed which might contribute to restoration fracture. Perhaps evenness of porcelain application contributes to over-all strength and should be given consideration in pprcelain fabrication. Design No. 3 (weakest) had the least uniform layer of porcelain over the metal structure, and design No. 1 (strongest) had the most uniform layer.

296

Warpeha

and Goodkind

J. Prosthet. Dmt. March, 1976

Metal preparations. Writers disagree as to whether bond strength is significantly altered by increasing surface roughness. *I 3, 5 Shell and Nielsen2 believe that the metalceramic bond is two-thirds chemical and one-third VanderWall’s force. Therefore, little importance is attributed to the surface roughness of the prepared metal. They minimized the importance of mechanical bonding. A finely roughened surface, however, may be “wetted” more easily, therefore, possibly increasing the bond strength. Kelly and associates” stated that very rough surfaces may increase stress concentration at the bond, thus weakening the bond. Whether or not the bond was altered, no significant difference in fracture resistance was found between the rough- and finestone samples of design No. 1 in this investigation. Porcelain contamination. The strength of the porcelain itself contributed a large part of the crown’s fracture resistance. The use of coarse silicon carbide stone, which left carbonaceous material imbedded in the metal, produced bubbling of the porcelain. This bubbling was effectively sealed off by use of the coating agent. Yet the fracture strength was not affected significantly in the group when minor bubbling occurred. In one instance, where the bubbling was severe and located under the loading stylus, fracture occurred at relatively the same load as in the samples which had reduced bond strength. This decrease in strength could have resulted from the weakened cross section of the porcelain or an absence of bonding due to the inclusion of voids. Metal coating and oxide layer. There are disagreements as to whether bond strength is altered by use of a metal coating agent.2, 3, 5 Yet this study showed an inconsistent effect on the fracture strength. The groups prepared with a fine stone, which included all three designs, showed significantly higher fracture strengths with the use of the metal conditioner. The group consisting only of design No. 1 prepared with a coarse stone showed no significant difference. An explanation for this occurrence is not evident. The only statement that can be made is that the use of a coating agent does not decrease fracture resistance. Anthony and associate9 found that the bond strength was reduced by 30 per cent if the alloy surface was depleted of oxide and reduced by 84 per cent when the alloy was coated with 24 karat gold. When the metal coating agent was applied too thick producing a “24 karat gold” finish or when “degassing” was deleted before porcelain application, clean separation at the metal was evidence of a reduced bond strength. Perhaps the trace metallic oxides in the metal were not available (as evidenced by a shiny surface) for chemical bonding with the porcelain. In these instances, the fracture strength was also severely reduced. RELIABILITY

OF METHOD

Relating the laboratory results obtained from testing a composite handmade object and directly applying these data to clinical situations are extremely difficult. A solid understructure was used to eliminate variables of coping fit, distortion, and metal rigidity and strength. The solid die used in this study cooled at a different rate than the customary thin coping underneath a metal-ceramic restoration, and perhaps the quality of the porcelain could be altered by this difference. The rate of loading was 0.02 mm. per second on the Instron testing machine in order that an expanded graph would be produced during testing. This is far

Volume 3.5 Number 3

Factors affecting fracture of restorations

297

slower than the rate of chewing of 7.4 mm. per second. However, the teeth would seldom strike at this rate of acceleration without food being interposed. In sustained muscle contraction, such as clenching or bruxing, forces might be applied at a rate much closer to the actual testing conditions described in this study. The construction of the porcelain crown forms is subject to variation. The size, density, and residual stress in each porcelain sample were similar, yet not identical. The point was made In the study of Shell and Nielsen* that the fracture strength of a brittle solid is statistical in nature, depending on the probability that a flaw capable of initiating fracture at a specific applied stress is present. For these reasons, this study had considerable variability which prevented good statistical evidence concerning certain questions. More definitive results could be obtained with similar studies using larger sample sizes. CONCLUSIONS An attempt was made in this study to make a clinically meaningful measure of fracture resistance in metal-ceramic restorations. Forty-four solid metal-ceramic crown forms were fabricated and subjected to compressive load testings. Variables included the presence or absence of a metal coating agent, the type of metal preparation (using stones of different abrasives), and three designs of the underlying metal. The following conclusions were arrived at : (I) The design of the underlying metal structure had a significant relation to the ultimate fracture strength. (2) A design with a definite acuteness of the underlying metal structure failed at significantly lower ultimate fracture strengths. (3) A metal conditioning agent did not decrease fracture resistance if applied properly. (4) Fracture strength was severely decreased when (a) improper thickness of the coating agent was used and (b) porcelain was fused to an unoxidized metal surface. (5) Bond strength, although a contributing factor, may not be as important as metal design and proper manipulation of materials during fabrication of the restoration. The authors thank Jacob E. Bearman, Ph.D., Professor, Division of Biometry, University of Minnesota, School of Public Health, Minneapolis, Minn., for his assistance in the statistica’ interpretation of the data.

References 1. Phillips,

R. W.: Skinner’s

Science of Dental

Materials,

ed. 7, Philadelphia,

1973, W. B.

Saunders Company, p. 547. 2. Shell, J. S., and Nielsen, J. P.: Study of the Bond Between Gold Alloys and Porcelain, J. Dent. Res. 41: 1424-1437, 1962. 3. Goeller, I., Meyer, J. M., and Nally, J, N.: Comparative Study of Three Coating Agents and Their Influence on Bond Strength of Porcelain-Fused-to-Gold Alloys, J. PROSTHET. DENT.

28:

X14-511,

1972.

4. Anthony, D. H., Burnett, A. P., Smith, D. L., and Brooks, M. S.: Shear Test for Measuring Bonding in Cast Gold Alloy-Porcelain Composites, J. Dent. Res. 49: 27-33, 1970.

29%

Warpeha

J, Prosthrt. Dent. March, 1976

and Goodkind

5. Lavine, M. H., and Custer, F.: Variables Affecting the Strength of Bond Between Porcelain and Gold, J. Dent. Res. 45: 32-36, 1966. 6. Nally, J. N.: Chemico-physical Analysis and Mechanical Tests of the Ceramo-metallic Complex, Int. Dent. J. 18: 309-325, 1968. 7. Nally, J. N., Farah, J. W., and Craig, R. G.: Experimental Stress Analysis of Dental Restorations. Part IX. Two-Dimensional Photoelastic Stress Analysis of Porcelain Bonded to Gold Crowns, J. PROSTHET. DENT. 25: 307-316, 1971. 8. Klaffenbach, A. D.: Gnathodynamic, J. Am. Dent. Assoc. 23: 371-382, 1936. 9. Kelly, M., Asgar, K., and O’Brien, W. J.: Tensile Strength Determination of the Interface Between Porcelain Fused to Gold, J. Biomed. Mater. Res. 3: 403-408, 1969. UNIVERSITY OF MINNESOTA SCHOOL OF DENTISTRY MINNEAPOLIS, MINN. 55455

ARTICLES Open-eye

TO APPEAR IN FUTURE

impression

technique

Alice Katz, B.S., and Henry

D.D.S.,

pressures

Bone-titanium W. L. Kydd,

D.D.S.,

implant

against

M.S., Fredrick

Rehabilitation of grossly deficient implants and bone grafts F. James Kratochvil,

prostheses

0. Gold, D.D.S.

A method of measuring complete dentures Charles C. Kelsey, B.E.E., MS.

for orbital

Philip

mandibles

J. Boyne,

response

supporting

functioning

with

D.M.D.,

to mechanical

combined

subperiosteal

M.S., and R. L. Bump,

D.D.S.

stress

D.D.S., and C. H. Daly, Ph.D.

N. Landesman,

D.D.S.,

training

Jack Light,

Sidney I. Silverman,

D.D.S.,

M.S.D.,

aid in the speech

Registrations for relating the mandibular on Kennedy’s classification system 0. Lundquist,

prosthodontic

education

at the

M.Ed.

The use of an intraoral laryngectomy patients

Donald

tissues

D. Reid, B.S., M.S., and Joel A. Coplowitz,

The making of a prosthodontist: Advanced University of Southern California Howard

ISSUES

D.D.S., and Guy E. Fiebiger,

D.D.S.,

rehabilitation

and Lawrence

cast to the maxillary D.M.D.

of Garfinkel,

M.A.

cast based

Design and technique variables affecting fracture resistance of metal-ceramic restorations.

Design and resistance technique variables of metal-ceramic affecting fracture restorations* Walter S. Warpeha, Jr., D.D.S., M&D.,** Richard J...
1MB Sizes 0 Downloads 0 Views