f surface
roughness
and cement space on crown
retention Niwut
Juntavee,
DDS, MSD,a and Philip
L. Millstein,
DMD, MSb
Khon Kaen Dental School, Khon Kaen, Thailand, and Goldman School of Graduate Dentistry, Boston University, Boston, Mass. The effects of varying luting agent space and internal surface roughness with different types of cores and cements were studied. One hundred eighty amalgam and 180 composite cores were cemented into standardized stainless steel retainers. Cores and retainers were divided into 12 groups according to core type, core diameter, and retainer roughness. Each group was further subdivided according to cement, A: zinc phosphate (ZOP); B: resin; and C: glass ionomer cement (GIG). Subgroups were divided into thermal-cycled and nonthermal-cycled groups. Thermal cycling was at 5O to 55O C, repeated 500 times. Cores were separated from their retainers with a compression rod in an Instron testing machine at a crosshead speed of 0.02 cm/minute. Results were as follows: Amalgam cores were most retentive. Resin and ZOP cements were equally retentive with amalgam cores, but GIC was less retentive. Resin cores cemented with resin cement were more than twice as retentive than those cemented with ZOP or GIC cements. Retainers with rough internal surfaces were most retentive. A reduced cement space between core and retainer was most retentive. Thermal cycling reduced retention. (J PROSTHETDENT~~~~;~&~S~-~.)
c
ast metal restorations continue to be commonly used to restore severely damaged teeth and to reinforce endodontically treated teeth. Often so little tooth structure remains that a cast restoration cannot be adequately supported and retained without an intermediate buildup with a suitable dowel and core used to provide a supporting substructure for a cast restoration. Important factors relating to the success of a cast restoration are the design of the preparation of the supporting core and the accuracy of the casting. Other factors include the nature of the core material on which the casting is cemented, the luting agent and its biomechanical characteristics, and the degree of bond strength between the cement and the core material.le3 Dental amalgam, widely used as a restorative material, is the choice of many dentists for core buildup. It has a high compressive strength and adapts well to tooth structure. Composite resin has been promoted as a core material because it sets quickly and allows a crown preparation to be started immediately after placement. Dental cements are expected not only to lute cast restorations to prepared tooth structure but also to provide a marginal seal that will be impervious to the deleterious effects of oral fluids. It has been postulated that cement dis-
solution and consequent loss of retention may be hastened by thermal expansion and contraction. The difference in the coefficients of linear thermal expansion between cast restorations, dental cements, and the underlying tooth structure may be responsible for the cement dissolutionP Thermal cycling is the laboratory model used to mimic oral temperature changes that range from 10’ to 50” C. The bond strength of a luting agent to dentin is an important consideration in the successof cast restorations. It is equally important that the bond strength of the luting agent to various core materials be within the range of clinical acceptability. The retention of a restoration to various core materials may be influenced by the physical properties of the core material or the luting agent, as well as by the stress imposed by thermal cycling. Few studies examine the bond strength of luting agents to dentin substitutes.5,6 Even fewer deal with the effect of internal surface roughness of a casting, 7-gthe amount of luting agent space, or the effect of thermal cycling on the cast retention.‘O, I1 This study explores crown retention as it relates to: (I) core material, (2) luting agents; (3) thicknesses of cement, (4) internal surface roughness of castings, and (5) the effects of thermal stress.
aIn~tructor, Department of Prosthodontics, Khon Kaen Dental School. bAssociate Clinical Professor, Department of Biomaterials, Goldman School of Graduate Dentistry, Boston University. 10/1139143
Three hundred sixty cylindrical cores (180 of each type) were used in this study: (1) amalgam (Valiant, L. D. Caulk, Milford, Del.) and (2) composite resin (Core Paste, Den Mat Corp., Santa Maria, Calif.). Sixty cores of each typed
482
MATERIALS AND Core fabrications
METHODS
SEPTEMBER1992
VOLUME68
NUMBER3
CEMENT
SPACE
AND
SURFACE
ROUGHNESS
cap
Side View
Top View
Core Cement
Ring Fig. 1. Schematic of core fabrication.
measuring 8.7,8.8, and 8.9 mm in diameter and 6.0 mm in length were made in a Teflon-coated aluminum mold (Du Pont Corp., Wilmington, Del.) (Fig. 1). All materials were mixed and manipulated according to the manufacturer’s instructions. After setting, the cores were stored in 100% humidity at 37’ C for 1 week.
Retainer
Fig. 2. Schematic of luted retainer ring with end caps.
Compressive
fabrication
Rod
Sixty cylindrical retainers 12 mm in diameter and 6 mm in length were prepared from a rod of base metal alloy (Rexillium III, Jeneric Gold Co., Wallingford, Conn.). A cylindrical space 9 mm in diameter was machined into the center of each cylinder, yielding a 1.5 mm thick outer surface. The inner surfaces of the cylindrical retainers were sandblasted with a fine-grit aluminum oxide abrasive. All retainers were cleaned and sandblasted with either a fineor coarse-grit aluminum oxide abrasive before reuse.
Core retainer
preparation
Before cementation all excess core material was carefully removed from the specimens by using a laboratory handpiece with an abrasive cutting instrument. The cores were cleaned with an air/water spray and were dried with compressed air for 10 seconds. Circumferential retainers were cleaned with a detergent in an ultrasonic bath and were washed in acetone.
Core retainer
cementation
Cores and retainers were divided into 12 groups of 30 specimens each according to type of core material, diameter of core (cement space), and internal surface roughness of retainer. Each group was further subdivided into three subgroups of 10 samples according to the luting agent: (1) zinc phosphate cement (Mission, White Dental, Lakewood, N.J.); (2) resin cement (Panavia Ex, J. Morita, Tustin, Calif.); and (3) glass ionomer cement (Ketac Cement, ESPEPremier, Norristown, Pa.). All cements were mixed according to the manufacturers’ instructions. The mixed cements were applied to tbe inner surfaces of the retainers and the outer surfaces of the cores. Each core was seated into a retainer and pushed. in place until it was level with the outer
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Core Cement Metal Casting Seating Base Fig. 3. Schematic
of retainer
in machined
base before
push-out test.
surface of the retainer. The open-ended design of the retainer assured complete seating of the core into the retainer. Excess cement was removed, and upon setting all exposed areas were coated with varnish. Samples were stored in 100% humidity at 37’ C. Each subgroup was divided into two groups consisting of a control group and an experimental group of five samples each. Samples in the experimental group were thermal cycled 1 week after cementation.
Thermal
cycling
procedure
Before testing, 1.5 mm thick machined caps (Rexillium) were cemented with silicone adhesive to both ends of the retainers to seal the core material from excess moisture and to provide a uniformly thick metal jacket that could be heated or cooled upon thermal cycling (Fig. 2). The samples were cycled 500 times between a 60’ C hot bath and a 4O C cold bath. The retainers were placed in each bath for 60 seconds, with &second intervals between immersions. Total immersion time was 1000 minutes. At the end of that
483
JUNTAVEE
Mean
separation
force
”
GIC
Resin Resin
Core
15O”nl tine. T
Fig. 4. Histogram shows maximum and minimum mean retentive strength of resin cores. Resin luting agent with cement space of 50 grn, rough-coarse internal surface, nonthermal-cycled; GIC luting agent with cement space of 150 pm, fine internal surface, thermal-cycled.
Mean
separation
force
fibs)
1500:
-
I
1400’ 1300, [ 1200
0
SD
EssMwm
t
1100 1000, L 900’ 800
tr
700 1 500 500 400 300 200 100 n GIG
ZOQ
Resin
Amalgam
Core
Fig. 5. Histogram shows retentive strength of amalgam cores. Resin luting agent with cement space of 100 pm, coarse internal surface, nonthermal-cycled, zinc phosphate (ZOP) luting agent with cement space of 100 Km, coarse internal surface, nonthermal-cycled, and glass ionomer (GICj luting agent, cement space of 100 pm, coarse internal surface, nonthermal-cycled.
time the caps were removed and the samples were tested for retention. Core retainer
separation
Forces required to dislodge the samples were measured by a constant displacement rate testing machine (Instron Corp., Canton, Mass.). Each core retainer assembly was seated in a custom-machined steel base that was prepared with a shoulder and slot centered to firmly seat the retainers before testing. Cores were forced out of the retainers by a custom-machined steel compression rod at 0.02 cm/min crosshead speed (Fig. 31, and the separation loads were recorded.
484
MILLSTEIN
RESULTS
(Ibs)
1500;
50em r”“ph, NT
AND
The retentive capacity of crown retainers as a function of the type of cement, the cement film thickness, the type of core material, and the internal surface roughness of retainers was measured by determining the mean separation force for core and cement before and after thermal cycling. The values obtained were subjected to a five-way analysis of variance (ANOVA). This statistical treatment revealed a significant (p < 0.001) five-factor interaction involving the type of core material, the type of cement, cement film thickness, the internal surface roughness of the retainer, and the effect of thermal cycling upon the retentive bond strength of the cemented cores. Further statistical analysis of variance, using the Newman-Keuls test, described more fully the effect of thermal cycling upon bond strength, which was significant (p < 0.001) for combinations of cement a,nd core mateirals. It also evaluated the significance of cement film thickness upon bond strength. Because there were so many statistically significant interactions, only those interactions with clinical application are referred to in the text. The highest mean bond strength (1374.40 lb) was obtained with resin cores cemented with resin cement to the coarse internal surfaces of the cast retainers, with a cement space of 50 pm. The lowest mean bond strength (149.32 lb) was obtained with resin cores cemented with glass ionomer cement to the fine internal surfaces of retainers, with a cement space of 150 Km (Fig. 4). There was no significant difference between the bond strength of an amalgam core/resin cement combination and that of an amalgam core/zinc phosphate cement combination. These two values did, however, differ significantly from that of the amalgam/glass ionomer bond strength (Fig. 5). There was a highly significant difference between resin core/resin cement bond strength and both resin core/zinc phosphate and resin core/glass ionomer combinations (Fig. 6). There was also a significant difference in retention between retainers with fine and coarse internal surfaces. Coarse internal surfaces provided greater retention than fine internal surfaces. This applied to all cements (Fig. 7). There was a significant difference on the retentive bond strength (p < 0.001) as cement film thickness was increased. Retentive bond strength decreased as the cement film thickness increased to 150 pm. Thermal cycling significantly reduced the bond strength of all cement-core samples (Fig. 8). DISCUSSION The main objective of this study was to assesscast restoration cementation as it related to some of the many variables that affect retention. Five variables: (1) cements; (2) cement film thickness; (3) core materials; (4) internal surface roughness of the retainer; and (5) thermal stress were selected for this study. The results substantiated the SEPTEMBER
1992
VOLUME
68
NUMBER
3
CEMENT
SPACE
Mean
AND
seuaration
SURFACE
!orce
ROUGHNESS
lean
00s)
1400 1300
1200 1100 1000
900
900
800 700
600 700
t
mMMean
400
I
300 200 100 0 ftesin
ZOP
GIC
Resin Core
Fig. 6. Histogram shows retentive strength of resin cores. Resin luting agent with cement space of 100 pm, coarse internal surface, nonthermal-cycled; zinc phosphate (ZOP) luting agent with cement space of 100 pm, coarse internal surface, nonthermal-cycled; and glass ionomer {GIG) luting agent, cement space of 100 pm, coarse internal surface, nonthermal-cycled. differences in retention attributable to the choice of core materials (amalgam versus composite resin), the cement film thickness (50, 100, and 150 pm), the internal surface roughness of retainers (fine versus coarse), and thermal stress (thermal cycled versus nonthermal cycled). Nonthermal cycled amalgam cores exhibited a higher bond strength with all the cements than all nonthermal cycled resin cores. These findings compare favorably with those of Dilts et al.,5 who reported that amalgam was the most retentive core material with all cements. Their study also found that the strongest bond occurred between a composite core and resin cement. Millstein and Nathanson reported that a composite resin core, when used with resin cement, provided the most retentive core-cement combination These conclusions are strongly supported in this study with resin cores cemented with resin cement. The consistently high bond strength of cemented amalgam cores found in this study and others is attributed for the most part to the dimensional stability of amalgam. Because amalgam neither shrinks nor swells, the thin, fragile film of set cement is less likely to suffer any damage. In addition, surface roughness probably also contributes to retention. Unpolished set amalgam, unlike the smooth resin cores, provided a surface roughness as a by-product of crystallization. Nonthermal cycled resin cores yielded retentive bond strengths significantly smaller than those found with amalgam cores, with one major exception: resin cores cemented with resin cement. The unusually high results obtained with this combination may be explained by the similarity of the chemical composition between the resin core material and the resin luting agent. The compatibility may consist of an extended chemical reactivity between the two materials or, perhaps the cohesion of physical surfaces THE
SD
500
400
100 0
0
600
500
200
(Ibs)
1300
1100 1000
300
force
1400
t
1200
600
separation
15001
1600
JOURNAL
OF PROSTHETIC
DENTISTRY
internal
Surface
7. Histogram shows differences inretentive strengths of amalgam cores with 50 wrn luting agent space, nonthermal-cycled, comparing fine and coarse internal surfaces of retainers,
Fig.
Mean
separation
force
(Ibs)
~~~-7;:
so0 700 600 500 400 300 200 100 0 50 micron
100 micron
Cement
150 micron
Space
8. Histogram depicts statistically significant differences in retentive strength as related to luting agent space and thermal cycling (hatched bars) when resin cores are luted to retainers with a fine internal surface. Fig.
that approximate one another so closely in microscopic detail. In all but one instance zinc phosphate and resin luting agents were the most retentive, regardless of the nature of the core materials. These findings again support the work by Dilts et al.,5 in which resin and zinc phosphate cements were most retentive when luted to composite resin and amalgam cores. The one exception was that of a resin luting agent and a resin core, which yielded the highest mean retentive value recorded in this study. This finding corresponds to that of Al-Quoud et al.,lr who reported that composite resin cores luted with polyurethane resin luting agents were the most retentive. The glass ionomer luting agent was generally less retentive than other agents, but revealed a greater retentive capacity with amalgam cores than with resin cores. It is con485
JUNTAVEE
ceivable that the explanation for this unusual observation may be found in the displacement of hydrogen bonds, inherent in glass ionomer luting agents, by accessible atoms of tin (Sn++) from the surface of the amalgam core.12 The general superiority of a zinc phosphate luting agent with respect to retentive bond strength has been ascribed to its ability to wet the surface to which it has been applied, (that is, its intrinsically low surface tension). This characteristic is a function of molecular cohesiveness that must be of a relatively low order in the freshly mixed zinc phosphate luting agent, and is expressed by its reduced viscosity and excellent flow. It is this ability to flow into or about each microscopic crevice or roughness before setting that enhances the retentive property of the hardened luting agent. During luting, phosphoric acid may contribute to overall retention by its “cleansing” activity in that it might etch the core or casting surface, thereby increasing the effective surface area of retention. Resin luting agents yielded retentive bond strengths in the same range as zinc phosphate luting agents with one exception when they were used with resin cores. The effect of luting agents’ film thickness and type, as they relate to the bond strength of cast retainers, was significant. An increase of luting agent film thickness reduced the bond strength of castings with all types of luting agents and with the two types of core material. The film thicknesses of 50 and 100 pm were not statistically significantly different in retentive bond strength; however, there was a significant reduction in retentive bond strength of the retainers when the luting agent film thickness was increased to 150 pm. Variation of the internal surface roughness of the retainers was sufficiently significant to affect the retentive bond strength. Retainers with a coarse internal surface exhibited a higher bond strength with all types of luting agents used than the retainers with finer internal surfaces. This finding correlates well with that of Phillips,13 who reported that castings with roughened surfaces provided greater retention than castings with smooth surfaces. CONCLUSIONS 1. Amalgam was superior to other core materials regardless of the luting agent used. Resin cores yielded a lower bond strength than amalgam except when used with resin luting agents.
486
AND
MILLSTEIN
2. Zinc phosphate and resin luting agents were more retentive than glass ionomer luting agents for all core materials. 3. Thermal cycling reduced the retentive bond strength for all samples. 4. Luting agent film thicknesses of 50 and 100 pm were more retentive than aluting agent film thickness of 150 pm. 5. Retainers with coarse internal surfaces exhibited a higher bond strength than those with smooth internal surfaces. We thank L. Corman,DMD, PhD, for sharing his knowledgein the chemistry of cementationand Ms. Linda Rosefor her help in analyzing the statistical data. REFERENCES 1.
2. 3. 4. 5. 6.
I. 8. 9. 10. 11. 12. 13.
Ghan KC, Azarbal P, Kerber PE. Bond strength of cements to crown bases. J PROSTHETDENT 1981;46:297-9. Oldman D, Swartz M, Phillips R. Retentive properties of dental cement on crown retention. J PROSTHETDENT 1964;14:760-8. Worley JL, Hamm RC, van Fraunbofer JA. Effects of cement on crown retention. J PROSTHETDENT 1982;48:289-91. O’Brien WJ. Dental materials: Properties and selection. Chicago: Quintessence Publishing Co, 1989527-8. Dilts WE, Duncanson MG, Miranda J, Brackett SE. Relative shear bond strengths of luting media with various core materials. J PROSTHETDENT 1985;53:505-8. Millstein P, Nathanson D. Temporary cement effect on composite core retention with permanent cement in vitro [Abstract]. J Dent Res 1988;67(special issue):344. Hormati AA, Denehy GE. Retention of cast crowns cemented to amalgam and composite resin cores. J PROSTHETDENT 1981;45:525-8. Ady AB, Fairhust CW. Bond strength of two types of cement to gold casting alloy. J PROSTHETDENT 1973:29:217-20. Jorgensen KD. The relationship between the retention and convergence angle in cemented veneer crowns. Acta Odontol Stand 1955;13:35-45. Norman RD, Swartz ML, Phillips RW. Study on film thickness, solubility and marginal leakage on dental cements. J Dent Res 1963;42:950-8. Al-Quoud OA, Millstein P, Nathanson D. Effects of thermalcycling on retention of different cement-core material combinations [Abstract]. J Dent Res 1989;68(special issue):955. Hotz P, McLean JW, Seed I, Wilson AD. The bonding of glass ionomer cements to metal and tooth substrates. Br Dent J 1977;142:41-7. Phillips RW. Skinner’s Science of dental materials. 8th ed., Philadelphia: WE Saunders, 1982463.
Reprintrequeststo: DR. PHILIP L. MILLSTEIN 15 LANCDONST. CAMBRIDGE,MA 02138
SEPTEMBER
1992
VOLUME
68
NUMBER
3
roughness
and cement space on crown
retention Niwut
Juntavee,
DDS, MSD,a and Philip
L. Millstein,
DMD, MSb
Khon Kaen Dental School, Khon Kaen, Thailand, and Goldman School of Graduate Dentistry, Boston University, Boston, Mass. The effects of varying luting agent space and internal surface roughness with different types of cores and cements were studied. One hundred eighty amalgam and 180 composite cores were cemented into standardized stainless steel retainers. Cores and retainers were divided into 12 groups according to core type, core diameter, and retainer roughness. Each group was further subdivided according to cement, A: zinc phosphate (ZOP); B: resin; and C: glass ionomer cement (GIG). Subgroups were divided into thermal-cycled and nonthermal-cycled groups. Thermal cycling was at 5O to 55O C, repeated 500 times. Cores were separated from their retainers with a compression rod in an Instron testing machine at a crosshead speed of 0.02 cm/minute. Results were as follows: Amalgam cores were most retentive. Resin and ZOP cements were equally retentive with amalgam cores, but GIC was less retentive. Resin cores cemented with resin cement were more than twice as retentive than those cemented with ZOP or GIC cements. Retainers with rough internal surfaces were most retentive. A reduced cement space between core and retainer was most retentive. Thermal cycling reduced retention. (J PROSTHETDENT~~~~;~&~S~-~.)
c
ast metal restorations continue to be commonly used to restore severely damaged teeth and to reinforce endodontically treated teeth. Often so little tooth structure remains that a cast restoration cannot be adequately supported and retained without an intermediate buildup with a suitable dowel and core used to provide a supporting substructure for a cast restoration. Important factors relating to the success of a cast restoration are the design of the preparation of the supporting core and the accuracy of the casting. Other factors include the nature of the core material on which the casting is cemented, the luting agent and its biomechanical characteristics, and the degree of bond strength between the cement and the core material.le3 Dental amalgam, widely used as a restorative material, is the choice of many dentists for core buildup. It has a high compressive strength and adapts well to tooth structure. Composite resin has been promoted as a core material because it sets quickly and allows a crown preparation to be started immediately after placement. Dental cements are expected not only to lute cast restorations to prepared tooth structure but also to provide a marginal seal that will be impervious to the deleterious effects of oral fluids. It has been postulated that cement dis-
solution and consequent loss of retention may be hastened by thermal expansion and contraction. The difference in the coefficients of linear thermal expansion between cast restorations, dental cements, and the underlying tooth structure may be responsible for the cement dissolutionP Thermal cycling is the laboratory model used to mimic oral temperature changes that range from 10’ to 50” C. The bond strength of a luting agent to dentin is an important consideration in the successof cast restorations. It is equally important that the bond strength of the luting agent to various core materials be within the range of clinical acceptability. The retention of a restoration to various core materials may be influenced by the physical properties of the core material or the luting agent, as well as by the stress imposed by thermal cycling. Few studies examine the bond strength of luting agents to dentin substitutes.5,6 Even fewer deal with the effect of internal surface roughness of a casting, 7-gthe amount of luting agent space, or the effect of thermal cycling on the cast retention.‘O, I1 This study explores crown retention as it relates to: (I) core material, (2) luting agents; (3) thicknesses of cement, (4) internal surface roughness of castings, and (5) the effects of thermal stress.
aIn~tructor, Department of Prosthodontics, Khon Kaen Dental School. bAssociate Clinical Professor, Department of Biomaterials, Goldman School of Graduate Dentistry, Boston University. 10/1139143
Three hundred sixty cylindrical cores (180 of each type) were used in this study: (1) amalgam (Valiant, L. D. Caulk, Milford, Del.) and (2) composite resin (Core Paste, Den Mat Corp., Santa Maria, Calif.). Sixty cores of each typed
482
MATERIALS AND Core fabrications
METHODS
SEPTEMBER1992
VOLUME68
NUMBER3
CEMENT
SPACE
AND
SURFACE
ROUGHNESS
cap
Side View
Top View
Core Cement
Ring Fig. 1. Schematic of core fabrication.
measuring 8.7,8.8, and 8.9 mm in diameter and 6.0 mm in length were made in a Teflon-coated aluminum mold (Du Pont Corp., Wilmington, Del.) (Fig. 1). All materials were mixed and manipulated according to the manufacturer’s instructions. After setting, the cores were stored in 100% humidity at 37’ C for 1 week.
Retainer
Fig. 2. Schematic of luted retainer ring with end caps.
Compressive
fabrication
Rod
Sixty cylindrical retainers 12 mm in diameter and 6 mm in length were prepared from a rod of base metal alloy (Rexillium III, Jeneric Gold Co., Wallingford, Conn.). A cylindrical space 9 mm in diameter was machined into the center of each cylinder, yielding a 1.5 mm thick outer surface. The inner surfaces of the cylindrical retainers were sandblasted with a fine-grit aluminum oxide abrasive. All retainers were cleaned and sandblasted with either a fineor coarse-grit aluminum oxide abrasive before reuse.
Core retainer
preparation
Before cementation all excess core material was carefully removed from the specimens by using a laboratory handpiece with an abrasive cutting instrument. The cores were cleaned with an air/water spray and were dried with compressed air for 10 seconds. Circumferential retainers were cleaned with a detergent in an ultrasonic bath and were washed in acetone.
Core retainer
cementation
Cores and retainers were divided into 12 groups of 30 specimens each according to type of core material, diameter of core (cement space), and internal surface roughness of retainer. Each group was further subdivided into three subgroups of 10 samples according to the luting agent: (1) zinc phosphate cement (Mission, White Dental, Lakewood, N.J.); (2) resin cement (Panavia Ex, J. Morita, Tustin, Calif.); and (3) glass ionomer cement (Ketac Cement, ESPEPremier, Norristown, Pa.). All cements were mixed according to the manufacturers’ instructions. The mixed cements were applied to tbe inner surfaces of the retainers and the outer surfaces of the cores. Each core was seated into a retainer and pushed. in place until it was level with the outer
THE
JOURNAL
OF PROSTHETIC
DENTISTRY
Core Cement Metal Casting Seating Base Fig. 3. Schematic
of retainer
in machined
base before
push-out test.
surface of the retainer. The open-ended design of the retainer assured complete seating of the core into the retainer. Excess cement was removed, and upon setting all exposed areas were coated with varnish. Samples were stored in 100% humidity at 37’ C. Each subgroup was divided into two groups consisting of a control group and an experimental group of five samples each. Samples in the experimental group were thermal cycled 1 week after cementation.
Thermal
cycling
procedure
Before testing, 1.5 mm thick machined caps (Rexillium) were cemented with silicone adhesive to both ends of the retainers to seal the core material from excess moisture and to provide a uniformly thick metal jacket that could be heated or cooled upon thermal cycling (Fig. 2). The samples were cycled 500 times between a 60’ C hot bath and a 4O C cold bath. The retainers were placed in each bath for 60 seconds, with &second intervals between immersions. Total immersion time was 1000 minutes. At the end of that
483
JUNTAVEE
Mean
separation
force
”
GIC
Resin Resin
Core
15O”nl tine. T
Fig. 4. Histogram shows maximum and minimum mean retentive strength of resin cores. Resin luting agent with cement space of 50 grn, rough-coarse internal surface, nonthermal-cycled; GIC luting agent with cement space of 150 pm, fine internal surface, thermal-cycled.
Mean
separation
force
fibs)
1500:
-
I
1400’ 1300, [ 1200
0
SD
EssMwm
t
1100 1000, L 900’ 800
tr
700 1 500 500 400 300 200 100 n GIG
ZOQ
Resin
Amalgam
Core
Fig. 5. Histogram shows retentive strength of amalgam cores. Resin luting agent with cement space of 100 pm, coarse internal surface, nonthermal-cycled, zinc phosphate (ZOP) luting agent with cement space of 100 Km, coarse internal surface, nonthermal-cycled, and glass ionomer (GICj luting agent, cement space of 100 pm, coarse internal surface, nonthermal-cycled.
time the caps were removed and the samples were tested for retention. Core retainer
separation
Forces required to dislodge the samples were measured by a constant displacement rate testing machine (Instron Corp., Canton, Mass.). Each core retainer assembly was seated in a custom-machined steel base that was prepared with a shoulder and slot centered to firmly seat the retainers before testing. Cores were forced out of the retainers by a custom-machined steel compression rod at 0.02 cm/min crosshead speed (Fig. 31, and the separation loads were recorded.
484
MILLSTEIN
RESULTS
(Ibs)
1500;
50em r”“ph, NT
AND
The retentive capacity of crown retainers as a function of the type of cement, the cement film thickness, the type of core material, and the internal surface roughness of retainers was measured by determining the mean separation force for core and cement before and after thermal cycling. The values obtained were subjected to a five-way analysis of variance (ANOVA). This statistical treatment revealed a significant (p < 0.001) five-factor interaction involving the type of core material, the type of cement, cement film thickness, the internal surface roughness of the retainer, and the effect of thermal cycling upon the retentive bond strength of the cemented cores. Further statistical analysis of variance, using the Newman-Keuls test, described more fully the effect of thermal cycling upon bond strength, which was significant (p < 0.001) for combinations of cement a,nd core mateirals. It also evaluated the significance of cement film thickness upon bond strength. Because there were so many statistically significant interactions, only those interactions with clinical application are referred to in the text. The highest mean bond strength (1374.40 lb) was obtained with resin cores cemented with resin cement to the coarse internal surfaces of the cast retainers, with a cement space of 50 pm. The lowest mean bond strength (149.32 lb) was obtained with resin cores cemented with glass ionomer cement to the fine internal surfaces of retainers, with a cement space of 150 Km (Fig. 4). There was no significant difference between the bond strength of an amalgam core/resin cement combination and that of an amalgam core/zinc phosphate cement combination. These two values did, however, differ significantly from that of the amalgam/glass ionomer bond strength (Fig. 5). There was a highly significant difference between resin core/resin cement bond strength and both resin core/zinc phosphate and resin core/glass ionomer combinations (Fig. 6). There was also a significant difference in retention between retainers with fine and coarse internal surfaces. Coarse internal surfaces provided greater retention than fine internal surfaces. This applied to all cements (Fig. 7). There was a significant difference on the retentive bond strength (p < 0.001) as cement film thickness was increased. Retentive bond strength decreased as the cement film thickness increased to 150 pm. Thermal cycling significantly reduced the bond strength of all cement-core samples (Fig. 8). DISCUSSION The main objective of this study was to assesscast restoration cementation as it related to some of the many variables that affect retention. Five variables: (1) cements; (2) cement film thickness; (3) core materials; (4) internal surface roughness of the retainer; and (5) thermal stress were selected for this study. The results substantiated the SEPTEMBER
1992
VOLUME
68
NUMBER
3
CEMENT
SPACE
Mean
AND
seuaration
SURFACE
!orce
ROUGHNESS
lean
00s)
1400 1300
1200 1100 1000
900
900
800 700
600 700
t
mMMean
400
I
300 200 100 0 ftesin
ZOP
GIC
Resin Core
Fig. 6. Histogram shows retentive strength of resin cores. Resin luting agent with cement space of 100 pm, coarse internal surface, nonthermal-cycled; zinc phosphate (ZOP) luting agent with cement space of 100 pm, coarse internal surface, nonthermal-cycled; and glass ionomer {GIG) luting agent, cement space of 100 pm, coarse internal surface, nonthermal-cycled. differences in retention attributable to the choice of core materials (amalgam versus composite resin), the cement film thickness (50, 100, and 150 pm), the internal surface roughness of retainers (fine versus coarse), and thermal stress (thermal cycled versus nonthermal cycled). Nonthermal cycled amalgam cores exhibited a higher bond strength with all the cements than all nonthermal cycled resin cores. These findings compare favorably with those of Dilts et al.,5 who reported that amalgam was the most retentive core material with all cements. Their study also found that the strongest bond occurred between a composite core and resin cement. Millstein and Nathanson reported that a composite resin core, when used with resin cement, provided the most retentive core-cement combination These conclusions are strongly supported in this study with resin cores cemented with resin cement. The consistently high bond strength of cemented amalgam cores found in this study and others is attributed for the most part to the dimensional stability of amalgam. Because amalgam neither shrinks nor swells, the thin, fragile film of set cement is less likely to suffer any damage. In addition, surface roughness probably also contributes to retention. Unpolished set amalgam, unlike the smooth resin cores, provided a surface roughness as a by-product of crystallization. Nonthermal cycled resin cores yielded retentive bond strengths significantly smaller than those found with amalgam cores, with one major exception: resin cores cemented with resin cement. The unusually high results obtained with this combination may be explained by the similarity of the chemical composition between the resin core material and the resin luting agent. The compatibility may consist of an extended chemical reactivity between the two materials or, perhaps the cohesion of physical surfaces THE
SD
500
400
100 0
0
600
500
200
(Ibs)
1300
1100 1000
300
force
1400
t
1200
600
separation
15001
1600
JOURNAL
OF PROSTHETIC
DENTISTRY
internal
Surface
7. Histogram shows differences inretentive strengths of amalgam cores with 50 wrn luting agent space, nonthermal-cycled, comparing fine and coarse internal surfaces of retainers,
Fig.
Mean
separation
force
(Ibs)
~~~-7;:
so0 700 600 500 400 300 200 100 0 50 micron
100 micron
Cement
150 micron
Space
8. Histogram depicts statistically significant differences in retentive strength as related to luting agent space and thermal cycling (hatched bars) when resin cores are luted to retainers with a fine internal surface. Fig.
that approximate one another so closely in microscopic detail. In all but one instance zinc phosphate and resin luting agents were the most retentive, regardless of the nature of the core materials. These findings again support the work by Dilts et al.,5 in which resin and zinc phosphate cements were most retentive when luted to composite resin and amalgam cores. The one exception was that of a resin luting agent and a resin core, which yielded the highest mean retentive value recorded in this study. This finding corresponds to that of Al-Quoud et al.,lr who reported that composite resin cores luted with polyurethane resin luting agents were the most retentive. The glass ionomer luting agent was generally less retentive than other agents, but revealed a greater retentive capacity with amalgam cores than with resin cores. It is con485
JUNTAVEE
ceivable that the explanation for this unusual observation may be found in the displacement of hydrogen bonds, inherent in glass ionomer luting agents, by accessible atoms of tin (Sn++) from the surface of the amalgam core.12 The general superiority of a zinc phosphate luting agent with respect to retentive bond strength has been ascribed to its ability to wet the surface to which it has been applied, (that is, its intrinsically low surface tension). This characteristic is a function of molecular cohesiveness that must be of a relatively low order in the freshly mixed zinc phosphate luting agent, and is expressed by its reduced viscosity and excellent flow. It is this ability to flow into or about each microscopic crevice or roughness before setting that enhances the retentive property of the hardened luting agent. During luting, phosphoric acid may contribute to overall retention by its “cleansing” activity in that it might etch the core or casting surface, thereby increasing the effective surface area of retention. Resin luting agents yielded retentive bond strengths in the same range as zinc phosphate luting agents with one exception when they were used with resin cores. The effect of luting agents’ film thickness and type, as they relate to the bond strength of cast retainers, was significant. An increase of luting agent film thickness reduced the bond strength of castings with all types of luting agents and with the two types of core material. The film thicknesses of 50 and 100 pm were not statistically significantly different in retentive bond strength; however, there was a significant reduction in retentive bond strength of the retainers when the luting agent film thickness was increased to 150 pm. Variation of the internal surface roughness of the retainers was sufficiently significant to affect the retentive bond strength. Retainers with a coarse internal surface exhibited a higher bond strength with all types of luting agents used than the retainers with finer internal surfaces. This finding correlates well with that of Phillips,13 who reported that castings with roughened surfaces provided greater retention than castings with smooth surfaces. CONCLUSIONS 1. Amalgam was superior to other core materials regardless of the luting agent used. Resin cores yielded a lower bond strength than amalgam except when used with resin luting agents.
486
AND
MILLSTEIN
2. Zinc phosphate and resin luting agents were more retentive than glass ionomer luting agents for all core materials. 3. Thermal cycling reduced the retentive bond strength for all samples. 4. Luting agent film thicknesses of 50 and 100 pm were more retentive than aluting agent film thickness of 150 pm. 5. Retainers with coarse internal surfaces exhibited a higher bond strength than those with smooth internal surfaces. We thank L. Corman,DMD, PhD, for sharing his knowledgein the chemistry of cementationand Ms. Linda Rosefor her help in analyzing the statistical data. REFERENCES 1.
2. 3. 4. 5. 6.
I. 8. 9. 10. 11. 12. 13.
Ghan KC, Azarbal P, Kerber PE. Bond strength of cements to crown bases. J PROSTHETDENT 1981;46:297-9. Oldman D, Swartz M, Phillips R. Retentive properties of dental cement on crown retention. J PROSTHETDENT 1964;14:760-8. Worley JL, Hamm RC, van Fraunbofer JA. Effects of cement on crown retention. J PROSTHETDENT 1982;48:289-91. O’Brien WJ. Dental materials: Properties and selection. Chicago: Quintessence Publishing Co, 1989527-8. Dilts WE, Duncanson MG, Miranda J, Brackett SE. Relative shear bond strengths of luting media with various core materials. J PROSTHETDENT 1985;53:505-8. Millstein P, Nathanson D. Temporary cement effect on composite core retention with permanent cement in vitro [Abstract]. J Dent Res 1988;67(special issue):344. Hormati AA, Denehy GE. Retention of cast crowns cemented to amalgam and composite resin cores. J PROSTHETDENT 1981;45:525-8. Ady AB, Fairhust CW. Bond strength of two types of cement to gold casting alloy. J PROSTHETDENT 1973:29:217-20. Jorgensen KD. The relationship between the retention and convergence angle in cemented veneer crowns. Acta Odontol Stand 1955;13:35-45. Norman RD, Swartz ML, Phillips RW. Study on film thickness, solubility and marginal leakage on dental cements. J Dent Res 1963;42:950-8. Al-Quoud OA, Millstein P, Nathanson D. Effects of thermalcycling on retention of different cement-core material combinations [Abstract]. J Dent Res 1989;68(special issue):955. Hotz P, McLean JW, Seed I, Wilson AD. The bonding of glass ionomer cements to metal and tooth substrates. Br Dent J 1977;142:41-7. Phillips RW. Skinner’s Science of dental materials. 8th ed., Philadelphia: WE Saunders, 1982463.
Reprintrequeststo: DR. PHILIP L. MILLSTEIN 15 LANCDONST. CAMBRIDGE,MA 02138
SEPTEMBER
1992
VOLUME
68
NUMBER
3
