You are on page 1of 5

Inuence of recasting palladium-silver alloy on the t of crowns with different marginal congurations

Murilo B. Lopes, DDS, MS,a Simonides Consani, DDS, MS, PhD,b Mario A. C. Sinhoreti, DDS, MS, PhD,c and Lourenco Correr-Sobrinho, DDS, MS, PhDd Department of Operative Dentistry, Dentistry School of Piracicaba, State University of Campinas, Piracicaba, SP, Brazil Statement of problem. Dental laboratories may occasionally recast previously cast dental alloy to produce a
prosthesis, but this process may have a negative inuence on marginal adaptation.

Purpose. The aim of this study was to evaluate the cervical and internal t of complete metal crowns that were
cast and recast using palladium-silver alloy and 3 different marginal congurations: straight shoulder, 20-degree bevel shoulder, and 45-degree chamfer. Material and methods. Thirty aluminum dies were made, 10 for each marginal conguration. Each group of 10 was further divided into 2 subgroups (n=5), according to the alloy (Pors-on 4) casting history. The dies were waxed using a cylindrical metal matrix to standardize the thickness of the walls. The wax patterns were invested in a phosphate-bonded investment (Deguvest) and cast after heating in an oven (EDG), according to the manufacturers instruction. The alloy was melted with an oxygen-gas heat source and cooled to room temperature after casting. Afterwards, the specimens were seated on the dies with a static load of 9 kgf for 1 minute. The marginal discrepancy was measured with a metric microscope, using a digital micrometer. All the measurements were done by the same operator. Each crown-die interface was measured 3 times on each of the 4 diametrically opposite points, for a total of 12 measurements for each specimen. Specimens were longitudinally sectioned to obtain 2 hemi-sections for internal discrepancy measurements, measured at 3 points: 1 in the center of the occlusal wall and 2 at each center of each axial wall. Three measurements were made for each point, for a total of 9 measurements per specimen. The results were submitted to parametric statistical 3-way analysis of variance followed by Tukey-Kramer HSD post hoc analysis (a=.05). Results. The new alloy provided signicantly better adaptation (P,.001) than the recast alloy for both marginal and internal discrepancy measurements. Marginal designs did not shown any statistical differences when the new metal was used. For recast metal, the straight shoulder showed signicantly better adaptation (P,.001) than the 20-degree bevel shoulder and the 45-degree chamfered congurations, which were statistically similar to each other. Conclusion. The results of this study suggest that recasting procedures for crown fabrication should not be used with the palladium-silver alloy tested. (J Prosthet Dent 2005;94:430-4.)

Since the oral environment may reduce the longevity of restorations, and recast alloy can decrease the marginal t of cast restorations, the recasting procedure should not be used for restorations made with the palladium-silver alloy tested.

alladium-silver based alloys were introduced for dental use in the 1930s,1 but their use in combination with dental ceramics was not documented until 1974.2 Investigators have conrmed that the physical properties of these alloys are acceptable for dental treatment3-6 and that the cost is economically practical for xed

Postgraduate student. Professor and Chairman. c Professor. d Professor.


prosthodontics.7 Since palladium-silver alloy was introduced to the profession, its use has increased signicantly.8 Casting procedures require use of more metal than is needed to produce a restoration.9,10 Dental laboratories often reuse the casting surplus (sprue and metal remaining in the crucible former) to produce castings when high-cost alloys are used.3 Marginal adaptation is one of the most important and weakest links in the success of a cast restoration. There is some indication that the quality of marginal adaptation may be as important to gingival health as margin location.11 Factors that affect




cast restoration treatment success include restorative form, axial wall convergence angle, marginal conguration, axial wall roughness, and cement thickness.12,13 There is controversy in the literature regarding margin preparation and conguration for a cast restoration. Furthermore, the rationale for these choices is unclear.14-17 Alloy reproducibility is an important factor to consider when choosing the alloy material due to its oxidation,18 component volatilization by the heatsource,19 and porosity development.20 Thus, a recast alloy may not have the same reproducibility as a new alloy cast for the rst time. The position of a well-tting crown with good margins before cementation can be changed due to the cement layer.21,22 Therefore, internal relief is necessary to obtain space for the luting material. This relief can be achieved by a number of techniques, including chemical etching, electrochemical stripping, overexpansion of investment in casting procedures, and die spacing.23 Since physical properties of new and recast alloys are not always the same, evaluation of whether the marginal design can compensate for any recast alloy deciency is important. Thus, the aim of this study was to evaluate the cervical and internal t of complete metal crowns cast and recast with the same palladium-silver alloy, using 3 different marginal congurations: straight shoulder, 20-degree bevel shoulder, and 45-degree chamfer.


Pors-on 4 (Degussa-Huls, Hanau, Germany), a palladium-silver alloy (57.8% palladium, 30% silver, 6% tin, 4% indium, 2% zinc, and 0.20% ruthenium), was used in this study. Thirty dies were made of cylindrical aluminum bars simulating a complete crown preparation. These were prepared on a lathe (TR 600; Metalurgica Riosulense S.A., Rio do Sul, Brazil) with a square steel bit (Tool Master 12 HSS TM-024; Grupo Albafer, Sao Paulo, Brazil) at 780 rpm by 1 technician. The dies had a cylindrical base (9 mm in diameter and 5 mm in height) followed by a conical section (7 mm diameter on the base and 6 mm on the top of the cone, with a 5.7-mm height and 5-degree angulation on each side). Specimens were divided into 3 groups (n=10), and margins were prepared on the lathe according to the following marginal conguration: 1-mmwide, 45-degree straight shoulder, 20-degree beveled shoulder, or 45-degree straight chamfer margin, each one with 10 degrees of occlusal convergence. Thirty complete crown patterns were made with wax (Plastodent Art Line; Degussa-Huls) on the metal dies. To standardize the thickness of the lateral walls of the crowns, a cylindrical metal matrix was adapted. To facilitate the waxing, the die and the matrix were heated to 80C 6 5C. The liqueed wax was dropped in excess into the matrix-die system. A glass slab was then put

over the matrix in place and loaded with a 500-g weight for 5 minutes. After matrix removal, the nish line was visually inspected and a 2.5-mm-diameter wax sprue was xed at 45 degrees on the occlusal surface of the crown. Three wax patterns, 1 for each marginal conguration, were xed on the crucible former. A strain-reducing agent (Waxit; Degussa SA, Guarulhos, SP, Brazil) was applied on the pattern surfaces. Patterns were invested (Deguvest Investment; Degussa-Huls) into a silicone casting ring, and no liner was used. After the investment set, the investment cylinders were put into an electric oven (Edgcom 5p; EDG Equipments Ltd, Sao Carlos, SP, Brazil) for wax elimination, following the manufacturers instructions. The investment cylinders were equally subdivided into 2 additional groups: 5 cylinders with new alloy and 5 with 100% recast alloy, for a total of 5 castings for each marginal conguration and alloy. The alloys were cast using a gas-oxygen torch (Draeger; Labor Dental, Sao Paulo, Brazil) and a centrifugal casting machine (Motorcast compact; Degussa AG, Hanau, Germany). Castings were bench-cooled and removed from the investment. The specimens were relieved internally using airborne-particle abrasion (100-mm aluminum oxide) for 3 minutes with 4 kgf pressure. For marginal t measurement, the crowns were seated on their respective dies and an occlusal load was applied with a pneumatic press regulated to 9 kgf for 1 minute.23 The crown and die system was analyzed with a metric microscope (Ernst Leitz; Wetzlar, Germany) at 330 magnication using a digital micrometer (543 IDF series; Mitutoyo Sul Americana Ltd, Sao Paulo, Brazil) with 0.001 mm of measurement accuracy. The specimens were horizontally positioned on a support to avoid movement. All measurements were done by the same operator. Each crown-die interface was measured 3 times on each of 4 diametrically opposite points (90 degrees between each point ) that were previously marked with a ne permanent pen, for a total of 12 measurements for each specimen. For the internal t measurements, the crowns were seated on their respective dies, and an occlusal load was applied with a pneumatic press regulated to 9 kgf for 1 minute.23 The crowns were then joined to the metal die with a fast-setting cyanoacrylate adhesive (Superbonder; Henkel Ltd, Sao Paulo, Brazil) on the external surface. Afterwards, the crown-die systems were embedded in polyester resin (Resapol; Fiber Center, Sao Paulo, Brazil), and the specimens were sectioned longitudinally and mesio-distally. Three measurement points were marked, at the center of the occlusal wall and at the center of each axial wall. All the measurements were made by the same operator. Three measurements were made for each of the 3 points, for a total of 9 measurements per specimen.



Table I. Three-way ANOVA

Variation causes df Sum of square Mean square F P

Material (A) Margin (B) A3B Error Repetition Adaptation (C) A3C B3C A3B3C Error Total

1 2 2 24 29 2 2 4 4 48 89

93781.834 16108.665 23212.603 12836.959 245940.062 116424.376 8697.332 51222.155 40820.619 50503.084 513607.630

193781.834 8054.332 11606.301 534.873 58212.188 4348.666 12805.538 10205.154 1052.147

362.294 15.058 21.699

.00001 .00016 .00003

55.327 4.133 12.170 9.699

.00001 .02152 .00001 .00004

General mean=77.861; variation coefcient (A)=17.149%; variation coefcient (B)=41.660%.

Table II. Mean values (mm) of marginal discrepancies relative to margin design
Location Alloy Beveled shoulder 5% Chamfer 5% Straight shoulder 5%

Cervical Occlusal Axial

New Recast New Recast New Recast

0.0 236.1 79.5 188.3 0.0 69.1

6 6 6 6 6 6

0.0 15.7 19.3 31.2 0.0 19.9

aA bB aB bB aA aA

0.0 69.98 78.4 129.9 0.0 100.7

6 6 6 6 6 6

0.0 26.3 14.8 51.3 0.0 31.3

aA aA aB aB aA aAB

0.0 41.8 125.1 168.7 0.0 113.5

6 6 6 6 6 6

0.0 22.0 8.1 26.2 0.0 31.8

aA aA aB abC aA aB

Mean values followed by different lowercase letters in same row or different capital letters within same column indicate signicant difference for Tukey test (a=.05).

The data were submitted to statistical analysis using a parametric method with repeated-measures analysis of variance (ANOVA). The data were submitted to statistical evaluation by 3-way ANOVA, considering the interaction of margin design, alloy type, and measurement location. A split-plot design was used, supported by repeated measurements made from the same experimental group at different measurement locations. Since the interaction of factors was signicant, differences were submitted to multiple-comparison testing (Tukey-Kramer HSD post hoc analysis) set at a 5% probability level.

Table I shows the 3-way ANOVA for all the materials tested. The marginal adaptation, when new alloy was used, presented a statistically signicant difference (P,.001) with higher mean results compared to the recast alloy (Table II). There was not a signicant difference among the margin congurations tested when new alloy was used. However, for the recast alloy, the beveled shoulder margin conguration displayed a signicant difference (P,.001) with higher marginal discrepancies compared to other margin designs, which were all signicantly different (P,.001) from each other. When the occlusal region was analyzed, the new alloy did not show a signicant difference among the marginal congurations tested. However, for the recast

alloy, the crowns with the beveled shoulder margin showed a signicant difference (P,.001) with higher marginal discrepancies than the 45-degree chamfered specimens. The straight shoulder specimens were not signicantly different from the beveled shoulder and the 45-degree chamfer specimens with recast alloy. When the axial adaptation was analyzed, there were no statistical differences among the marginal congurations tested (Table III). For the beveled shoulder factor for the new alloy, the occlusal region presented signicantly (P,.001) higher mean values of marginal discrepancy relative to the cervical and axial regions that were not signicantly different from each other. For the recast alloy, the occlusal and cervical regions showed higher marginal discrepancy values than the axial region. When the 45-degree chamfered shoulder was analyzed for the new alloy, the occlusal region displayed signicantly (P,.001) higher mean values of marginal discrepancy than the cervical and axial regions, which were not signicantly different from each other. For the recast alloy, the occlusal region presented signicantly (P,.001) higher mean values than the axial region. The axial region did not differ from the other regions (Table III). Analyzing the straight shoulder design for the new alloy, the occlusal region showed signicantly (P,.001) higher mean values of marginal discrepancy than the cervical and axial regions, and there was no signicant



Table III. Mean values (mm) of discrepancies relative to internal location

Location/alloy New 5% Recast 5%

Total perimeter Cervical Occlusal Axial

31.4 0.0 94.3 0.0

6 6 6 6

51.4 0.0 26.2 0.0

A a b a

124.2 116.0 162.3 94.4

6 6 6 6

66.8 93.3 40.4 27.5

B a b a

Mean values followed by different lowercase letters in same column or different capital letters in row indicate signicant difference for Tukey test (P,.05).

difference between the cervical and axial regions. For the recast alloy, the occlusal region presented signicantly (P,.001) higher mean values than the cervical and axial regions. The marginal discrepancy values for the axial region were not signicantly different from those in the cervical region (Table III). When measurement location was considered with alloy type, the occlusal region showed the highest level of marginal discrepancy (P,.001) between the cervical and axial regions (Table II).

Successful crowns must have appropriate marginal sealing and design because these factors are essential to protect the luting agent from dissolution. In this study, the data regarding cervical marginal discrepancy showed that the new alloy had signicantly lower mean values than the recast alloy (Tables I and II). Reisbick and Brantley4 reported a decrease in yield strength, percentage elongation, and an increase in porosity with recast gold alloy. Rasmussen and Doukoudakis21 observed that casting porosity size and frequency increased in alloys with more than 85% recast material. According to Papazoglou et al,19 dimensional changes that may occur with palladium-based alloys at high temperatures are unpredictable, because they may be caused by internal oxidation near the surface, creep, or strain release. However, dental laboratories having experience with these alloys should generally be able to compensate for such dimensional changes and achieve a clinically acceptable casting t.12 Tuccillo et al20 compared the effects of 3 casting techniques on dimensional stability and hardness of a gold alloy and veried that noble metal did not change its composition, but that lighter nonprecious metal components were lost. According to the results of the current study, some alloy components probably volatilized due to their lower melting points (tin = 231.9C, indium = 156.2C, and zinc = 420C). Consequently, this likely resulted in an increased molten alloy viscosity due to the increasing of the alloy melting point. This could result in a poorer adaptation, considering that the loss of these constituents did not promote signicant change in the linear

coefcient of thermal expansion of the alloy10 and should be identied when analyzing marginal conguration. The results of this study showed that the 20-degree beveled shoulder conguration had a signicantly higher marginal discrepancy than the others, when using recast alloy. This probably occurred due to the difculty of the high-viscosity alloy to reproduce details, such as the thin margin of the 20-degree beveled shoulder. Syu et al14 and Byrne18 found no difference among the beveled shoulder, 45-degree chamfer, and straight shoulder when new alloy was used. In the present study, when new alloy was used, the results were in accordance with these cited studies. However, Schillinburg et al15 and Faucher and Nicholls16 demonstrated that the straight shoulder had the best marginal adaptation. Dedmon12 analyzed castings from different laboratories and showed that the beveled shoulder resulted in the worst marginal adaptation. However, this report did not mention either the heat source or the type of alloy used in castings. The nonstandardization technique of the laboratories could have been inuenced by different alloy compositions during the casting procedures, producing varying levels of reproducible details among the marginal congurations. The results of Dedmon12 are in accordance with the present study when recast alloys were used. According to Pilo and Cardash,13 the occlusal marginal discrepancy for cement thickness on the axial walls was a sine function of half of the convergence angle of the crown preparation. Thus, to estimate the occlusal marginal discrepancy, the following formula was used:
Occlusal marginal discrepancy Cement thickness on axial walls Sine of half of convergent angle

In the present study, it should be considered that the axial marginal discrepancy was equal to the space that the cement would occupy after crown cementation. The total convergence of axial walls in this study was 10 degrees; half of this is 5 degrees, and the sine of 5 degrees is 0.087. According to the formula, if the axial marginal discrepancy were 1 mm, the occlusal marginal discrepancy would be 11.5 mm. Thus, it is understandable why a small increase in axial wall marginal discrepancy would cause an increasingly decient occlusal and cervical adaptation. Independent of other factors (Table III), the recast alloy showed signicantly greater marginal discrepancy compared to the new alloy. Even with internal relief, there was attrition between the axial walls and the crown when recast alloy was used, which interfered with the crowns correct seating. This was probably due to the higher alloy contraction with the recast alloy. This was also demonstrated by Rasmussen and Doukoudakis21 even when new alloy was incorporated with recast alloy.



The experimental ndings in the present study using Pors-on 4 alloy might be generalized to other Pd-Ag alloys, since these alloys have a relatively narrow composition range. The clinical consensus is to produce metal restorations with the highest adaptation. Since castings must endure intraoral conditions, such as masticatory forces, solubilization, and microorganism interaction, these marginal discrepancies may negatively inuence the longevity of cast restorations. The less accurate the casting t and the greater the amount of cement exposed, the more likely the restorations failure because of cement degradation.10 This study evaluated the adaptation of cast metal crowns to dies. It is probable that the adaptation level veried in this study would be changed after the cementation procedure.

Within the limitations of this study, the following conclusions were drawn: 1. New palladium-silver alloy (Pors-on 4) produced signicantly (P,.02) better marginal and internal adaptation than recast alloy. 2. There were no signicant differences among the different marginal congurations tested when new alloy was used. 3. For the recast alloy, the straight chamfered and 45-degree chamfered congurations produced signicantly improved (P,.001) cervical adaptation than the beveled shoulder, except when analyzing the axial marginal discrepancy.
1. Craig RG, Peyton FA. Restorative dental materials. 5th ed. St. Louis: Mosby; 1975. p. 318. 2. Tuccillo JJ. Composition and functional characteristics of precious metal alloys for dental restorations. In: Valega TM, editor. Alternatives to gold alloys in dentistry. Bethesda (MD): US Department of Health, Education, and Welfare, Publication No. (NIH) 77-1227; 1977. p. 40-67. 3. Hong JM, Razzoog ME, Lang BR. The effect of recasting on the oxidation layer of a palladium-silver porcelain alloy. J Prosthet Dent 1988;59: 420-5. 4. Reisbick MH, Brantley WA. Mechanical property and microstructural variations for recast low-gold alloy. Int J Prosthodont 1995;8:346-50. 5. Lewis AJ. The effects of remelting on the mechanical properties of a nickel base partial denture casting alloy. Aust Dent J 1975;20:89-93.

6. Nelson DR, Palik JF, Morris HF, Comella MC. Recasting a nickel-chromium alloy. J Prosthet Dent 1986;55:122-7. 7. Viennot S, Dalard F, Lissac M, Grosgogeat B. Corrosion resistance of cobalt-chromium and palladium-silver alloys used in xed prosthetic restorations. Eur J Oral Sci 2005;113:90-5. 8. Goodacre CJ. Palladium-silver alloys: a review of the literature. J Prosthet Dent 1989;62:34-7. 9. Presswood RG. Multiple recast of a nickel-chromium-beryllium alloy. J Prosthet Dent 1983;50:198-9. 10. Anusavice KJ. Phillips science of dental materials. 10th ed. Philadelphia: W. B. Saunders; 1996. p. 315-26, 491-2. 11. Gardner FM. Margins of complete crownsliterature review. J Prosthet Dent 1982;48:396-400. 12. Dedmon HW. The relationship between open margins and margin designs on full cast crowns made by commercial dental laboratories. J Prosthet Dent 1985;53:463-6. 13. Pilo R, Cardash HS. In vivo retrospective study of cement thickness under crowns. J Prosthet Dent 1998;79:621-5. 14. Syu JZ, Byrne G, Laub LW, Land MF. Inuence of nish-line geometry on the t of crowns. Int J Prosthodont 1993;6:25-30. 15. Shillingburg HT Jr, Hobo S, Fisher DW. Preparation design and margin distortion in porcelain-fused-to-metal restorations. J Prosthet Dent 1973; 29:276-84. 16. Faucher RR, Nicholls JI. Distortion related to margin design in porcelainfused-to-metal restorations. J Prosthet Dent 1980;43:149-55. 17. Fusayama T. Factors and technique of precision casting. Part I and II. J Prosthet Dent 1959;9:468-86. 18. Byrne G. Inuence of nish-line form on crown cementation. Int J Prosthodont 1992;5:41-5. 19. Papazoglou E, Brantley WA, Johnston WM. Evaluation of hightemperature distortion of high-palladium metal-ceramic crowns. J Prosthet Dent 2001;85:133-40. 20. Tuccillo JJ, Lichtenberger H, Nielsen JP. Composition stability of gold base dental alloys for different melting techniques. J Dent Res 1974;53: 1127-31. 21. Rasmussen ST, Doukoudakis AA. The effect of using recast metal on the bond between porcelain and a gold-palladium alloy. J Prosthet Dent 1986;55:447-53. 22. Wilson PR. The effect of die spacing on crown deformation and seating time. Int J Prosthodont 1993;6:397-401. 23. Milan FM, Consani S, Correr-Sobrinho L, Sinhoreti MA, Sousa-Neto MD, Knowles JC. Inuence of casting methods on marginal and internal discrepancies of complete cast crowns. Braz Dent J 2004;15:127-32. Reprint requests to: DR MURILO BAENA LOPES RUA GUILHERME LOURENCAO, 35 ANHANGABAU 13208-113, JUNDIAI SAO PAULO BRAZIL FAX: 55-19-3421-0144 E-MAIL: 0022-3913/$30.00 Copyright 2005 by The Editorial Council of The Journal of Prosthetic Dentistry. doi:10.1016/j.prosdent.2005.08.018