B o n d i n g of p l a s t i c t e e t h to d e n t u r e b a s e r e s i n s M. K a w a r a , D D S , P h D , a J. M. C a r t e r , M S c , P h D , b R. E. O g l e , D D S , c a n d R. R. J o h n s o n , B S d
Nihon University, School of Dentistry, Matsudo, Japan, and State University of New York at Buffalo, School of Dental Medicine, Buffalo, N.Y. This study e v a l u a t e d the bond strengths of some n e w and traditional resin denture teeth and denture base resins. It included regular monolithic acrylic resin teeth (Bioform), monolithic acrylic r e s i n - I P N teeth (Bioform IPN), and multilithic acrylic r e s i n - c o m p o s i t e resin teeth (Vivosit) , with r e l a t i v e l y n e w l i g h t - a c t i v a t e d resin (Triad), c o n v e n t i o n a l heat-cured resin (Lucitone 199), and a u t o p o l y m e r i z i n g resin (Hygenic) denture base materials. The results of four-point flexure t e s t i n g s h o w e d that the traditional m a t e r i a l s g a v e the h i g h e s t bond strength v a l u e s . The autopolym e r i z i n g resin s y s t e m s demonstrated interfacial failure with all resin denture teeth, s h o w i n g that the common practice of t r e a t i n g teeth w i t h the r e s p e c t i v e a u t o p o l y m e r i z i n g m o n o m e r failed to produce adequate bond strength. Combinations of acrylic resin, IPN, and multilithic denture teeth with l i g h t - a c t i v a t e d r e s i n s g a v e results calling for i m p r o v e m e n t s in basic bonding s y s t e m design, since interface debonding w a s prevalent. No failures occurred b e t w e e n the lap-ridge region of the multilithic tooth s y s t e m and conventional heat-cured denture base resin. (J PROSTHET DENT 1991;66:566-71.)
6 .~---~,
A d bonding e qof acrylic u resin a teeth t toethe den-
ture base resin is necessary because it increases stiffness and strength since the teeth become an integral part of the prosthesis. In most instances, the bonding seems satisfactory, although failures can be found in practice. These may be related to the basic properties of the materials, or to manipulative factors like wax contamination of the bonding surfaces. Spratley I reported that the single most common denture repair made in his clinics involved the replacement of detached acrylic resin teeth. He concluded that residual wax appeared to be the chief cause of this type of failure. Besides contamination, the factors that affect the bond strength involve the degree of cross-linking in the materials and the available monomer during processing. Sorensen and Fjeldstad 2 reported an improvement in bonding using solvents where the resultant strength was dependent on the type of solvent and the time of exposure. For acrylic resin monomer to be effective, the dough must either swell or dissolve the ridge-lap portion of the tooth. The more cross-linking present, the less the swelling action by the available monomer. Also, bonding may be compromised if the dough polymerizes too quickly after contact with the teeth. 3
Presented at the International Association for Dental Research meeting, Dublin, Ireland. aVisiting Research Associate, SUNY at Buffalo, School of Dental Medicine. bAssociate Professor, Department of Biomaterials, SUNY at Buffalo, Schoolof Dental Medicine. CAssociate Professor, Department of Removable Prosthodontics, SUNY at Buffalo, School of Dental Medicine. dlnstructional Support Technician, Department of Biomaterials, SUNY at Buffalo, School of Dental Medicine. 10/1/29234 566
~AII Supports 1.8 Dia.
,I 2
18
4 ~
Bending Moment
Shear Force All
Dimensionsin mm.
F i g . 1. Four-point flexure test for tooth-denture base
bonding evaluation. Bending moment and shear force diagrams show state of pure bending between top supports.
In recent years, several new materials have been introduced, with claims of increased wear resistance, better esthetics, and more convenient curing methods. Trubyte Bioform IPN (interpenetrating polymer network) (Dentsply International Inc., York, Pa.) was introduced as an abrasion-resistant tooth material. IPN materials are produced by blending together two or more existing polymers having compatible networks, which form permanent entanglements and owe their improved properties to synergism of the networks. 4 When these materials are placed in solvents, they swell rather than dissolve. Other improvements in teeth have involved the combination of composite resins with conventional tooth acrylic resins to create the multilithic tooth (Ivoclar North America, Amherst, OCTOBER1991 VOLUME66 NUMBER4
BONDING OF DENTURE TEETH
Table I. Summary of specimen combinations and codings Denture teeth and No. o f s a m p l e s
Denture base materials
Heat-cured resin (HC) Lucitone 199
Autopolymerizing resin (AP) Perm Cross-Linked
Visible light-cured resin (VLC) Triad
None RE IPN ML None RE IPN ML None Regular RE IPN ML
10 10 10 10 10 10 10 10 10 10 10 10 120
Total
Denture teeth: RE, Regular acrylic resin, monolithic, Trubyte Bioform Mold 74H; IPN, acrylic resin-IPN, monolithic, Trubyte Bioferm-IPN Mold; ML, acrylic resin-Composite resin, multilithic, SR-Vivosit PE Mold A17.
N.Y.). This tooth combines the abrasion resistance of the composite resin with the denture base bonding capabilities of the acrylic resin. The introduction of a visible light-cured (VLC) denture base resin represents a new direction for denture bases (Triad, Dentsply International Inc.,). Ogle et al. 5found this material to be accurate, of superior strength, and of acceptable bond strength compared with heat-cured (HC) resin. Because of these new materials, it is necessary to evaluate the various combinations for bond strengths and to compare these with accepted standards. This was the purpose of the study. Bond strength, like any other strength property, is statistical in nature, since the presence of intrinsic or extrinsic flaws strongly influences fracture. The mechanical testing of strength is complicated by specimen geometry, size, test grip alignment, force direction, and other variables that usually produce complex stress distributions. Because of this, some tests produce more useful data than others. At least three bond tests have been used for denture bases and teeth. The American Dental Association (ADA) specification No. 156 uses a tensile test; the German proposal 7 for an International Organization for Standardization (ISO) draft involves 3-point bending; and the British Standard Specification (BSS) No. 39908 uses shear force loading. It has been suggested that the bend test is superior to the other two tests, since it showed better discriminatory power than the ADA specification No. 15, and provided more quantitative data than BSS No. 3990:1980.* We decided to use a variation of this test in the present study. Since three-point bending involves vertical shear as well as horizontal tension and compression along the beam, there is a shear stress maximum beneath the single loading roller, and the specimen tends to break here rather than elsewhere. In four-point bending, the region between the two *Combe EC, Tarigan S. Personal communication, 1988. THE JOURNAL OF PROSTHETIC DENTISTRY
•
22
-~
All Dimensions in mm.
Se0men
Compos,teRes,n
,~ DentureBaseResin •
22
All D i m e n s i o n s in mm.
Fig. 2. Cross sections of typical teeth superimposed on bend test specimen outline showing regions in teeth involved in bond testing. 567
K A W A R A ET AL
I
s,,io0 .I I I I I
t
Thrust Platen
Guide
~
/ Sub-Press Frame / Sliding Plunger Four Point Loading Rig Vee Block Anvil
Fig. 3. Flexure test sub-press. This frame was positioned between the Instron instrument crosshead and the compression load cell.
loading rollers is stressed in pure bending only, shear being absent (Fig. 1). Moreover, the horizontal tensile and compressive stresses are uniform, for a constant section cross section. This means that a series of bonded interfaces within a specified region will undergo equal loading, and thus the weakest location will be sought out. Therefore it is possible to correlate the fracture location with the failing load.
MATERIAL AND METHODS Details of the denture base resins and the denture teeth (central incisors and canines only) used in this study are given in Table I. Fig. 2 shows cross sections of typical monolithic and multilithic teeth. A total of 90 combinations of bend test specimens were produced with the three denture base resins and three types of teeth, plus a control group of 30 without teeth, with 10 replicates in each group. New teeth of the various types were cleaned and were then ground with a tungsten carbide acrylic resin bur on both the labial and lingual surfaces parallel to the labial mid one third of the tooth. This method produced samples of the center portions of the teeth that were about 3 mm wide, a suitable size for the flexure test specimens. However, with the multilithic teeth this produced exposure of the transitional composite resin layer. For consistency of specimen preparation, no mechanical retention was used, although the bonding of denture base resin to this layer without the placement of diatorics is not recommended by the manufacturer.* Specimens were then washed with a detergent in hot water, followed by ultrasonic cleaning in distilled water for 5 *Tysowsky G. Personal communication, Ivoclar North America, 1990.
568
minutes. Care was taken at all stages during subsequent handling to avoid contamination. The teeth were dried in air for at least 24 hours before processing. Heat-cured specimens were made in a 10 x 10 x 25 mm dental stone mold and were cured for 9 hours at 71 ° C (160 ° F). The teeth were wiped with monomer for 1 minute and were then placed in dough made with a liquid/powder ratio of 31 to 100 by volume, following the manufacturer's recommendations. This was trial packed immediately. Some porosity and polymerization problems were encountered with the autopolymerizing resin (AP) specimens, since we used a pourable mix to simulate practical usage of this type of resin. The specimen thickness was large compared to what would be used practically, so a pilot study was made to choose parameters so that nonporous specimens could be routinely made. It was found that a processing vessel with an internal pressure of 40 psi filled with 40 ° C (104 ° F) water produced sound specimens if the mold size was 6 × 6 x 25 ram. Again, teeth were treated with monomer before the mold was filled with resin. The mix was made using Perm Material (The Hygenic Corp., Akron, Ohio) with a liquid/powder ratio of 60 to 100 by volume, following the manufacturer's recommendations. The VLC resin specimens were processed in a 10 x 10 x 25 mm clear Lucite mold. A urethane dimethacrylate-based bonding agent (Triad, Dentsply International Inc.) was painted on each side of the tooth being embedded and cured for 2 minutes in a curing unit (Triad). Small amounts of VLC resin were then pressed onto opposite sides of a tooth and were cured for 4 minutes. The remainder of the specimen was then formed with additional resin, and was cured at top and bottom for 4 minutes each. All processed specimens were sized to 4.1 x 4.1 × 1 x 22 mm using a 3/8-inch diameter tungsten carbide end cutter in a milling machine running at 2100 rpm. Light cuts were made to avoid the formation of residual stresses 9 and to prevent breaking of specimens where bonding was minimal. The bars were abraded by hand to a final size of 4 x 4 x 22 using wet 120, 320, and 400 grit silicon carbide papers. All specimens were stored at 37 ° C in distilled water for 7 days before mechanical testing. Modulus of rupture or flexural strength values were obtained by testing finished bars in the four-point bending cell (Fig. 3). This cell consisted of a rigid sub-press with a carefully aligned loading plunger. The four-point loading configuration is shown in Fig. 1. The lower bed was constructed using an engineer's VEE block. Stainless steel rods, 1.8 mm in diameter, were positioned on the block using a measuring microscope and were epoxied in position. After the resin set, the rods were then tacked in position by spot welding. Finally, the epoxy was removed, and the rods were permanently secured with soft solder. The same technique was used to mount the rods on the plunger. Using this assembly technique produced linear tolerances of _+ 0.002 mm and parallelism for the loading bars of _+ 0.01 degrees.
OCTOBER 1991
VOLUME 66
NUMBER 4
BONDING OF D E N T U R E TEETH
Modulus of Rupture in 4-Point Bending 140 T 120
T
100 n
T
80
[ ] Control I~RE 17ff/]ME
60 40 20 0
Heat-Curing
Auto-Polymerizing
Visible Light-Curing
Resin Fig. 4. Flexure test results. Bars represent upper values for 95 % confidence limits.
Testing Failing loads were obtained from flexure tests using an Instron machine (Instron Corp., Canton, Mass.) to load the cell shown in Fig. 3. Specimens were wiped dry from storage, and were tested immediately, with a plunger displacement speed of i mm/min. This testing speed is given in the German proposal 7 for the ISO draft on synthetic resin teeth, which uses three-point flexure. Modulus of rupture (MR) values were calculated from the failing loads, W, in kilograms: WL MR = ~ × 9.805MPa where L = 18 mm, d = depth in millimeters, b = width in millimeters A two-way analysis of variance (ANOVA) was performed using the modulus of rupture values. The 3 × 3 experimental design excluded the "plain" specimens. Duncan's multiple range test was used for ranking, at the 0.05 % level of significance. RESULTS The plain heat-cured (HC) denture material was so tough in testing that no fractures occurred. Failing loads were taken from the maximum of the load deflection recording. Most of the plain autopolymerized (AP) specimens were similar, unlike the visible light-cured material (VLC), whose specimens broke in a brittle manner (Table II). HC material with regular acrylic resin (RE), IPN, and multilithic (ML) teeth showed the same order in strengths as the AP denture base resin specimens. The corresponding HC-tooth combinations were significantly higher in strength order and showed cohesive failures within the teeth, except for the ML teeth. In HC-ML specimens, the failures were almost equally distributed between the acrylic resin segment-composite resin interface, within the transitional composite resin, and at the transitional composite resin-denture resin interface. The AP-ML specimens gave
THE J O U R N A L OF P R O S T H E T I C D E N T I S T R Y
the lowest flexure strengths of any combination, with adhesive failures at the interface between the transitional composite resin layer and the denture resin. No failures occurred at the interface between the denture base-acrylic resin segment. The VLC denture resin combinations were not significantly different from each other and, as a group, were second lowest. The VLC-ML specimens showed mixed interface failures, equally divided between the denture resin-acrylic resin segment and the transitional composite resin-denture resin interfaces. DISCUSSION Failure modes and values for the various materials and combinations are detailed in Table II. Fig. 4 illustrates bond strength figures and includes the 95% confidence limits, which estimate the precision of the mean values. For any bond value, we are 95 % confident that the true mean falls within the indicated ranges. All the HC and 80% of the AP resin specimens showed excessive yielding without fracture. The load deflection curves went through maxima, and these loads were used to calculate an apparent rupture modulus. To achieve brittle failure, changes would be needed. These changes could involve increasing the depthto-width ratio of the cross section and the crosshead speed. All the plain VLC specimens failed in a brittle manner without appreciable plastic yielding. Because all the bonded specimens failed in a brittle manner, the flexure test was considered to be valid for evaluating bonding. The two-way ANOVA clearly shows interaction effects between denture base type and tooth type (Table III), which indicate that some combinations produce better bonding than others after accounting for the main effects. Since the HC resinto-RE plastic teeth gave the highest flexure strength and showed cohesive fracture, the bonding and therefore preparation steps are considered to be ideal. This was also true with HC-IPN combinations, although the strength values show the IPN teeth to be weaker than the RE plastic teeth.
569
KAWARAET AL
T a b l e II. Modulus of rupture (MPa) in four-point bending and failure modes Denture base resins
HC
AP
VLC
138 _+ 1.61 ductile yielding
128 _+ 7.56 ductile yielding
99.7 + 7.03 brittle fracture
112 _+_14.1 within tooth
60.8 + 8.22 interface
19.1 _+ 4.72 interface
IPN
92.2 _+ 14.4 within tooth
49.0 _+ 4.36 interface
16.8 _+ 4.75 interface
ML
31.5 + 13.3 within tooth, mixed
9.39 -+ 2 . 4 1 - interface
18.9 -+ 7.08 interfaces
Denture teeth RE
Abbreviationsas in Table I. Dispersions are standard deviations. Connecting lines join means that are not significantlydifferent at 95% confidencelevel as given by Duncan's multiple range test.
T a b l e III. Two-way analysis of variance S o u r c e of v a r i a t i o n
S u m of s q u a r e s
df
Mean square
F
p Value
Denture material Tooth material Two-way interactions Residual Total
55823.303 31221.232 18209.406 6936.425 112190.367
2 2 4 81 89
27911.651 15610.616 4552.352 85.635 1260.566
325.938 182.293 53.160
Nihon University, School of Dentistry, Matsudo, Japan, and State University of New York at Buffalo, School of Dental Medicine, Buffalo, N.Y. This study e v a l u a t e d the bond strengths of some n e w and traditional resin denture teeth and denture base resins. It included regular monolithic acrylic resin teeth (Bioform), monolithic acrylic r e s i n - I P N teeth (Bioform IPN), and multilithic acrylic r e s i n - c o m p o s i t e resin teeth (Vivosit) , with r e l a t i v e l y n e w l i g h t - a c t i v a t e d resin (Triad), c o n v e n t i o n a l heat-cured resin (Lucitone 199), and a u t o p o l y m e r i z i n g resin (Hygenic) denture base materials. The results of four-point flexure t e s t i n g s h o w e d that the traditional m a t e r i a l s g a v e the h i g h e s t bond strength v a l u e s . The autopolym e r i z i n g resin s y s t e m s demonstrated interfacial failure with all resin denture teeth, s h o w i n g that the common practice of t r e a t i n g teeth w i t h the r e s p e c t i v e a u t o p o l y m e r i z i n g m o n o m e r failed to produce adequate bond strength. Combinations of acrylic resin, IPN, and multilithic denture teeth with l i g h t - a c t i v a t e d r e s i n s g a v e results calling for i m p r o v e m e n t s in basic bonding s y s t e m design, since interface debonding w a s prevalent. No failures occurred b e t w e e n the lap-ridge region of the multilithic tooth s y s t e m and conventional heat-cured denture base resin. (J PROSTHET DENT 1991;66:566-71.)
6 .~---~,
A d bonding e qof acrylic u resin a teeth t toethe den-
ture base resin is necessary because it increases stiffness and strength since the teeth become an integral part of the prosthesis. In most instances, the bonding seems satisfactory, although failures can be found in practice. These may be related to the basic properties of the materials, or to manipulative factors like wax contamination of the bonding surfaces. Spratley I reported that the single most common denture repair made in his clinics involved the replacement of detached acrylic resin teeth. He concluded that residual wax appeared to be the chief cause of this type of failure. Besides contamination, the factors that affect the bond strength involve the degree of cross-linking in the materials and the available monomer during processing. Sorensen and Fjeldstad 2 reported an improvement in bonding using solvents where the resultant strength was dependent on the type of solvent and the time of exposure. For acrylic resin monomer to be effective, the dough must either swell or dissolve the ridge-lap portion of the tooth. The more cross-linking present, the less the swelling action by the available monomer. Also, bonding may be compromised if the dough polymerizes too quickly after contact with the teeth. 3
Presented at the International Association for Dental Research meeting, Dublin, Ireland. aVisiting Research Associate, SUNY at Buffalo, School of Dental Medicine. bAssociate Professor, Department of Biomaterials, SUNY at Buffalo, Schoolof Dental Medicine. CAssociate Professor, Department of Removable Prosthodontics, SUNY at Buffalo, School of Dental Medicine. dlnstructional Support Technician, Department of Biomaterials, SUNY at Buffalo, School of Dental Medicine. 10/1/29234 566
~AII Supports 1.8 Dia.
,I 2
18
4 ~
Bending Moment
Shear Force All
Dimensionsin mm.
F i g . 1. Four-point flexure test for tooth-denture base
bonding evaluation. Bending moment and shear force diagrams show state of pure bending between top supports.
In recent years, several new materials have been introduced, with claims of increased wear resistance, better esthetics, and more convenient curing methods. Trubyte Bioform IPN (interpenetrating polymer network) (Dentsply International Inc., York, Pa.) was introduced as an abrasion-resistant tooth material. IPN materials are produced by blending together two or more existing polymers having compatible networks, which form permanent entanglements and owe their improved properties to synergism of the networks. 4 When these materials are placed in solvents, they swell rather than dissolve. Other improvements in teeth have involved the combination of composite resins with conventional tooth acrylic resins to create the multilithic tooth (Ivoclar North America, Amherst, OCTOBER1991 VOLUME66 NUMBER4
BONDING OF DENTURE TEETH
Table I. Summary of specimen combinations and codings Denture teeth and No. o f s a m p l e s
Denture base materials
Heat-cured resin (HC) Lucitone 199
Autopolymerizing resin (AP) Perm Cross-Linked
Visible light-cured resin (VLC) Triad
None RE IPN ML None RE IPN ML None Regular RE IPN ML
10 10 10 10 10 10 10 10 10 10 10 10 120
Total
Denture teeth: RE, Regular acrylic resin, monolithic, Trubyte Bioform Mold 74H; IPN, acrylic resin-IPN, monolithic, Trubyte Bioferm-IPN Mold; ML, acrylic resin-Composite resin, multilithic, SR-Vivosit PE Mold A17.
N.Y.). This tooth combines the abrasion resistance of the composite resin with the denture base bonding capabilities of the acrylic resin. The introduction of a visible light-cured (VLC) denture base resin represents a new direction for denture bases (Triad, Dentsply International Inc.,). Ogle et al. 5found this material to be accurate, of superior strength, and of acceptable bond strength compared with heat-cured (HC) resin. Because of these new materials, it is necessary to evaluate the various combinations for bond strengths and to compare these with accepted standards. This was the purpose of the study. Bond strength, like any other strength property, is statistical in nature, since the presence of intrinsic or extrinsic flaws strongly influences fracture. The mechanical testing of strength is complicated by specimen geometry, size, test grip alignment, force direction, and other variables that usually produce complex stress distributions. Because of this, some tests produce more useful data than others. At least three bond tests have been used for denture bases and teeth. The American Dental Association (ADA) specification No. 156 uses a tensile test; the German proposal 7 for an International Organization for Standardization (ISO) draft involves 3-point bending; and the British Standard Specification (BSS) No. 39908 uses shear force loading. It has been suggested that the bend test is superior to the other two tests, since it showed better discriminatory power than the ADA specification No. 15, and provided more quantitative data than BSS No. 3990:1980.* We decided to use a variation of this test in the present study. Since three-point bending involves vertical shear as well as horizontal tension and compression along the beam, there is a shear stress maximum beneath the single loading roller, and the specimen tends to break here rather than elsewhere. In four-point bending, the region between the two *Combe EC, Tarigan S. Personal communication, 1988. THE JOURNAL OF PROSTHETIC DENTISTRY
•
22
-~
All Dimensions in mm.
Se0men
Compos,teRes,n
,~ DentureBaseResin •
22
All D i m e n s i o n s in mm.
Fig. 2. Cross sections of typical teeth superimposed on bend test specimen outline showing regions in teeth involved in bond testing. 567
K A W A R A ET AL
I
s,,io0 .I I I I I
t
Thrust Platen
Guide
~
/ Sub-Press Frame / Sliding Plunger Four Point Loading Rig Vee Block Anvil
Fig. 3. Flexure test sub-press. This frame was positioned between the Instron instrument crosshead and the compression load cell.
loading rollers is stressed in pure bending only, shear being absent (Fig. 1). Moreover, the horizontal tensile and compressive stresses are uniform, for a constant section cross section. This means that a series of bonded interfaces within a specified region will undergo equal loading, and thus the weakest location will be sought out. Therefore it is possible to correlate the fracture location with the failing load.
MATERIAL AND METHODS Details of the denture base resins and the denture teeth (central incisors and canines only) used in this study are given in Table I. Fig. 2 shows cross sections of typical monolithic and multilithic teeth. A total of 90 combinations of bend test specimens were produced with the three denture base resins and three types of teeth, plus a control group of 30 without teeth, with 10 replicates in each group. New teeth of the various types were cleaned and were then ground with a tungsten carbide acrylic resin bur on both the labial and lingual surfaces parallel to the labial mid one third of the tooth. This method produced samples of the center portions of the teeth that were about 3 mm wide, a suitable size for the flexure test specimens. However, with the multilithic teeth this produced exposure of the transitional composite resin layer. For consistency of specimen preparation, no mechanical retention was used, although the bonding of denture base resin to this layer without the placement of diatorics is not recommended by the manufacturer.* Specimens were then washed with a detergent in hot water, followed by ultrasonic cleaning in distilled water for 5 *Tysowsky G. Personal communication, Ivoclar North America, 1990.
568
minutes. Care was taken at all stages during subsequent handling to avoid contamination. The teeth were dried in air for at least 24 hours before processing. Heat-cured specimens were made in a 10 x 10 x 25 mm dental stone mold and were cured for 9 hours at 71 ° C (160 ° F). The teeth were wiped with monomer for 1 minute and were then placed in dough made with a liquid/powder ratio of 31 to 100 by volume, following the manufacturer's recommendations. This was trial packed immediately. Some porosity and polymerization problems were encountered with the autopolymerizing resin (AP) specimens, since we used a pourable mix to simulate practical usage of this type of resin. The specimen thickness was large compared to what would be used practically, so a pilot study was made to choose parameters so that nonporous specimens could be routinely made. It was found that a processing vessel with an internal pressure of 40 psi filled with 40 ° C (104 ° F) water produced sound specimens if the mold size was 6 × 6 x 25 ram. Again, teeth were treated with monomer before the mold was filled with resin. The mix was made using Perm Material (The Hygenic Corp., Akron, Ohio) with a liquid/powder ratio of 60 to 100 by volume, following the manufacturer's recommendations. The VLC resin specimens were processed in a 10 x 10 x 25 mm clear Lucite mold. A urethane dimethacrylate-based bonding agent (Triad, Dentsply International Inc.) was painted on each side of the tooth being embedded and cured for 2 minutes in a curing unit (Triad). Small amounts of VLC resin were then pressed onto opposite sides of a tooth and were cured for 4 minutes. The remainder of the specimen was then formed with additional resin, and was cured at top and bottom for 4 minutes each. All processed specimens were sized to 4.1 x 4.1 × 1 x 22 mm using a 3/8-inch diameter tungsten carbide end cutter in a milling machine running at 2100 rpm. Light cuts were made to avoid the formation of residual stresses 9 and to prevent breaking of specimens where bonding was minimal. The bars were abraded by hand to a final size of 4 x 4 x 22 using wet 120, 320, and 400 grit silicon carbide papers. All specimens were stored at 37 ° C in distilled water for 7 days before mechanical testing. Modulus of rupture or flexural strength values were obtained by testing finished bars in the four-point bending cell (Fig. 3). This cell consisted of a rigid sub-press with a carefully aligned loading plunger. The four-point loading configuration is shown in Fig. 1. The lower bed was constructed using an engineer's VEE block. Stainless steel rods, 1.8 mm in diameter, were positioned on the block using a measuring microscope and were epoxied in position. After the resin set, the rods were then tacked in position by spot welding. Finally, the epoxy was removed, and the rods were permanently secured with soft solder. The same technique was used to mount the rods on the plunger. Using this assembly technique produced linear tolerances of _+ 0.002 mm and parallelism for the loading bars of _+ 0.01 degrees.
OCTOBER 1991
VOLUME 66
NUMBER 4
BONDING OF D E N T U R E TEETH
Modulus of Rupture in 4-Point Bending 140 T 120
T
100 n
T
80
[ ] Control I~RE 17ff/]ME
60 40 20 0
Heat-Curing
Auto-Polymerizing
Visible Light-Curing
Resin Fig. 4. Flexure test results. Bars represent upper values for 95 % confidence limits.
Testing Failing loads were obtained from flexure tests using an Instron machine (Instron Corp., Canton, Mass.) to load the cell shown in Fig. 3. Specimens were wiped dry from storage, and were tested immediately, with a plunger displacement speed of i mm/min. This testing speed is given in the German proposal 7 for the ISO draft on synthetic resin teeth, which uses three-point flexure. Modulus of rupture (MR) values were calculated from the failing loads, W, in kilograms: WL MR = ~ × 9.805MPa where L = 18 mm, d = depth in millimeters, b = width in millimeters A two-way analysis of variance (ANOVA) was performed using the modulus of rupture values. The 3 × 3 experimental design excluded the "plain" specimens. Duncan's multiple range test was used for ranking, at the 0.05 % level of significance. RESULTS The plain heat-cured (HC) denture material was so tough in testing that no fractures occurred. Failing loads were taken from the maximum of the load deflection recording. Most of the plain autopolymerized (AP) specimens were similar, unlike the visible light-cured material (VLC), whose specimens broke in a brittle manner (Table II). HC material with regular acrylic resin (RE), IPN, and multilithic (ML) teeth showed the same order in strengths as the AP denture base resin specimens. The corresponding HC-tooth combinations were significantly higher in strength order and showed cohesive failures within the teeth, except for the ML teeth. In HC-ML specimens, the failures were almost equally distributed between the acrylic resin segment-composite resin interface, within the transitional composite resin, and at the transitional composite resin-denture resin interface. The AP-ML specimens gave
THE J O U R N A L OF P R O S T H E T I C D E N T I S T R Y
the lowest flexure strengths of any combination, with adhesive failures at the interface between the transitional composite resin layer and the denture resin. No failures occurred at the interface between the denture base-acrylic resin segment. The VLC denture resin combinations were not significantly different from each other and, as a group, were second lowest. The VLC-ML specimens showed mixed interface failures, equally divided between the denture resin-acrylic resin segment and the transitional composite resin-denture resin interfaces. DISCUSSION Failure modes and values for the various materials and combinations are detailed in Table II. Fig. 4 illustrates bond strength figures and includes the 95% confidence limits, which estimate the precision of the mean values. For any bond value, we are 95 % confident that the true mean falls within the indicated ranges. All the HC and 80% of the AP resin specimens showed excessive yielding without fracture. The load deflection curves went through maxima, and these loads were used to calculate an apparent rupture modulus. To achieve brittle failure, changes would be needed. These changes could involve increasing the depthto-width ratio of the cross section and the crosshead speed. All the plain VLC specimens failed in a brittle manner without appreciable plastic yielding. Because all the bonded specimens failed in a brittle manner, the flexure test was considered to be valid for evaluating bonding. The two-way ANOVA clearly shows interaction effects between denture base type and tooth type (Table III), which indicate that some combinations produce better bonding than others after accounting for the main effects. Since the HC resinto-RE plastic teeth gave the highest flexure strength and showed cohesive fracture, the bonding and therefore preparation steps are considered to be ideal. This was also true with HC-IPN combinations, although the strength values show the IPN teeth to be weaker than the RE plastic teeth.
569
KAWARAET AL
T a b l e II. Modulus of rupture (MPa) in four-point bending and failure modes Denture base resins
HC
AP
VLC
138 _+ 1.61 ductile yielding
128 _+ 7.56 ductile yielding
99.7 + 7.03 brittle fracture
112 _+_14.1 within tooth
60.8 + 8.22 interface
19.1 _+ 4.72 interface
IPN
92.2 _+ 14.4 within tooth
49.0 _+ 4.36 interface
16.8 _+ 4.75 interface
ML
31.5 + 13.3 within tooth, mixed
9.39 -+ 2 . 4 1 - interface
18.9 -+ 7.08 interfaces
Denture teeth RE
Abbreviationsas in Table I. Dispersions are standard deviations. Connecting lines join means that are not significantlydifferent at 95% confidencelevel as given by Duncan's multiple range test.
T a b l e III. Two-way analysis of variance S o u r c e of v a r i a t i o n
S u m of s q u a r e s
df
Mean square
F
p Value
Denture material Tooth material Two-way interactions Residual Total
55823.303 31221.232 18209.406 6936.425 112190.367
2 2 4 81 89
27911.651 15610.616 4552.352 85.635 1260.566
325.938 182.293 53.160
