Composite materials term paper

Dent Mater 10:116-121, March, 1994

Transverse strength and fatigue of denture acrylic-glass fiber composite

Pekka K. Vallittu, Veijo P. Lassila, Rolf Lappalainen

Department of Prosthetics and Stomatognathic Physiology, University of Kuopio, FINLAND

ABSTRACT Objectives. The aims of this experiment were: 1) to test the effect of a high concentration of continuous glass fibers on the transverse strength of test specimens made of heat-cured acrylic resin; and 2) to determine the fatigue weakening of both unreinforced and continuous glass fiber-reinforced specimens of heat-cured acrylic resin shaped into upper complete den- tures. Methods. A three-point loading test was used to determine the transverse strength of test specimens (n=30 per group). The fatigue test was the constant deflection test (n=l 0 per group). Results. The results revealed that, compared to unreinforced specimens, continuous glass fibers at a concentration of 58 wt% enhanced the transverse strength of the test specimens up to 146% (p<0.001) and increased the fatigue resistance (p<0.001) during 5 x 10 ~ loading cycles. Significance. This study suggests that by incorporating glass fibers into PMMA denture bases, the strength of the denture can be increased.

INTRODUCTION In an attempt to use fibers to improve the mechanical properties of dental polymers, various types of fibers, such as glass fiber, have been tested. In addition, glass fibers have been used in different forms including woven, loose (Schwickerath, 1965; Setz and Lin, 1989; Solnit, 1991) and continuous (i.e., roving form, fiber bundles) (Friskopp etal., 1979; Friskopp and Bloml6f, 1984; Inamuraet al., 1988; Vallittu and Lassila, 1992a; Goldberg and Burstone, 1992).

To strengthen polymer resins by including glass fibers, the fibers and the polymer matrix must adhere to one another. Silane compounds have been used to improve this adhesion (Clark and Plueddemann, 1963; Rosen, 1978). For optimum strengthening effect, the fibers included in the polymer material should be placed normal to the loading force (Galan and Lynch, 1989), and the concentration of fibers in the polymer matrix should be high (Schipp et al., 1992; Vallittu et al., 1994).

Various methods can be used to incorporate fibers into polymeric materials. Hand-placement of fibers in the

unpolymerized resin is the method often used by dental materials researchers (Skirvin et al., 1982; Vallittu and Lassila, 1992a). For dental purposes, Goldberg and Burstone (1992) presented the pultrusion process for producing plastic reinforced with glass fiber. They used this process to achieve a high concentration of fibers and good wetting of all fibers by the polymer resin.

Fatigue, i.e., the loss of strength as a result of stress over a period of time, generally occurs in most materials. Flexural fatigue of polymethyl methacrylate has been proposed as a cause of mid-line fractures of complete upper dentures (Smith, 1961; Beyli and von Fraunhofer, 1981). When the denture base is deformed under the loading that occurs during mastication, fatigue failure can occur. The maximal biting force of patients with complete dentures can be up to 200 N in the molar region and up to 80 N in the incisor region (Lassila et al., 1985). During normal mastication, however, the biting force remains lower than the maximal biting force. Fatigue of fiber-reinforced composites that consist of fibers depends on whether the loading is parallel to or inclined to the direction of the continuous fibers (Talreja, 1989).

The aim of the present study was to determine the transverse strength of an acrylic glass-fiber composite with a high concentra- tion of continuous glass fibers. In addition, the fatigue resistance of unreinforced and continuous glass fiber-reinforced specimens shaped like an upper complete denture was compared using the constant deflection fatigue test.

MATERIALS AND METHODS The acrylic resin used was heat-cured polymethyl methacrylate (PMMA) (Paladon 65, Kulzer GmbH, Wehrheim, Germany). This acrylic resin was pressed into an aluminum mold, as described previously (Vallittu and Lassila, 1992b), to prepare Type I test specimens (3.0 x 4.5 x 50.0 mm) and into a con-ventional denture mold to prepare Type II specimens (in the shape of an upper complete denture) (Fig. 1). The thickness of the Type I test specimens was ground to the predetermined dimensions with an accuracy of _+0.10 mm. The minimum thickness of Type II test specimens was 2.17 _+ 015 mm measured from the palatal area of the specimen (Fig. 2). Thirty Type I

116 Vallittu et al./Streng~ and fatigue of PMMA composite

Fig. 1. Test specimens of unreinforced heat-cured acrylic resin (Type I to the left, Type II to the right) and amount of continuous glass fibers used as reinforcement beside the test specimens.

®l e$

coPP[i/~ ' m~iN~; ~ ~ ~ + ,m. ,~ ~ + , ~ i

Fig. 3. Schematic representation of the method for pressing acrylic resin dough for Type I test specimens in an aluminum mold. After the first test pressing, copper wires were placed (A) into the mold. Wires were separated from acrylic resin with separating sheets of polyethylene and pressed into acrylic resin dough. They formed a concavity (B) for fiber rovings. Fibers dipped in PMMA-MMA mixture were incorporated into the concavity and pressed (C) into acrylic resin. The arrow indicates the direction of the pressure load.

Fig. 2. Mid-line section of Type II test specimen, The dotted line indicates the position of glass fibers. The arrows indicate the thickness of test specimen.

c

Fig. 4. Schematic representation of position of fibers in Type II test specimens (A), area of applied force (B), and stress bearing area under the test specimens (C).

specimens and ten Type II specimens were tested both with and without fiber reinforcement.

E-glass fiber rovings (bundles of fibers) (Ahlstrom, Karhula, Finland) formulated from 55% SiO2, 22% CaO, 15% AI203 and 6% B203 were silanized by dipping them in a silane solution, silane compound A174 (gammamethacryloxypropyltrimethoxy-silane, Union Carbide Chemicals, Versoix, Switzerland) which reacts with methylmethacrylate (MMA). The fibers were air-dried for 40 min before further handling. The fiber rovings for Type I test specimens weighed 0.449 g and for Type II test specimens 0.820 g. The length of fiber rovings for Type I specimens was 48 mm and for Type II specimens 58 mm. The individual fibers were 10 ~m in diameter.

Silanized glass-fiber rovings were dipped in a mixture of PMMA power and MMA liquid (ratio 10:8 by weight), as de- scribed previously (Vallittu, 1994). After the fiber rovings were

dipped in the PMMA-MMA mixture, they were placed into the mold longitudinally to the long axis of Type I test specimens. Before this, unpolymerized acrylic resin dough had already been pressed into the mold, and the space for the fiber rovings had been made by using copper wire (2.6 mm in diameter) as a space maker (Fig. 3). The space maker for Type II test specimens was 5.0 mm in diameter, extending from the second premolar on the left side to the second premolar on the right side of the test specimen (Fig. 4). The fiber rovings were pressed into the mold at a pressure of 10 MPa. Acrylic resin was allowed to undergo short-term polymerization in a water bath according to the manufacturer's instructions (the mold was placed in boiling water and remained at 80°C for 45 min); after cooling, the specimens were removed from the mold. They were then stored in water at room temper- ature for 2 wk before testing.

The fracture loads of Type I test specimens were measured

Dental Materials~March 1994 117

300 600 900 1 2 3

l___ff_J 1 0 0 N ? N ? N

~cles

Fig. 6. Schematic representation of deflection amplitudes (A1)during loading of Type II test specimens. The initial force applied was 100 N.

Fig. 5. SEM micrographs of a transverse section of Type I test specimens [top magnification 26x (bar 1.0 ram); bottom magnification 2000 x], showing PMMA beside glass fibers (bar 10 i~m). The arrow indicates small cleft between the fibers and PMMA.

with a Schleuniger 2E testing machine (Schleuniger, Solothurn, Switzerland) at a crosshead speed of 120 mndmin, which was also u s e d in previous studies (Vallittu et al., 1992a; 1992b). The span of the three-point loading test was 35.0 millimeters. The transverse strengths (S) were calculated using the formula below (Phillips, 1982):

3 W L S - (1)

2 b d 2

where W is the fracture force, L is the distance between supports (35.0 mm), b the specimen width, and d the specimen thickness.

Transverse sections of the Type I test specimens were cut with a low speed saw (Isomet, Buehler Ltd., Lake Buff, IL, USA) for further analysis. Surfaces of the cut samples were polished through No. 1200 sandpaper and KMG polishing liquid (Candulor, Zurich, Switzerland). The polished samples were cleaned for 10 rain by an ultrasonic cleaning machine in water. The surfaces of the sputtered samples were examined by scanning electron microscope (JEOL JSM-35, Tokyo, Japan) at 15 kV, and micro- graphs were taken for visual inspection (Fig. 5).

The fiber content of the Type I specimens was verified by ashing the loaded specimens for 30 rain at 750°C. Before ashing, the specimens were dried for 5 h at 100°C and weighed to an accuracy of 0.001 g for calculation of fiber content.

To examine the fatigue weakening of the Type II specimens, the constant deflection fatigue test was carried out with a pneu- matic testing machine constructed at the Technical Department of the University of Kuopio. During testing, the specimens were immersed in 37°C water. The initial force (100 N) applied to the test specimens affected their deflection amplitude, A 1 (Fig. 6). The magnitude of the deflection amplitude was measured with gauge strips, the thickness of which had been verified with a micrometer (NSK-DigitrixMKII, Japan Micrometer Ltd., Osaka, Japan). The stress-bearing area under the test specimens simu- lated the anatomical structure of the toruspalatinus. There was no contact between the test specimens and the residual ridge of the testing machine (Fig. 7). The force required to produce the initial deflection amplitude was applied to the specimens after each 300 ms (= I cycle). Changes in applied force as a function of deflection cycles were used as indicators of fatigue weakening of the test specimens.

To verify that changes of applied force are caused by loss of modulus of the material, parameters D 1 and D z (Fig. 7) were measured. The difference between these parameters indicates the abrasion of the occlusal surfaces and adaptation of the test specimen at the stress-bearing area during cyclic loading (Fig. 4). To establish the weakening of the material, the magnitude of the

118 Vallittu et al./Strength and fatigue of PMMA composite

TABLE 1: FRACTURE LOADS OF TYPE I TEST SPECIMENS AND TRANSVERSE STRENGTH

Fracture Load Transverse n M + SD Strength p

(N) (MPa)

Unreinforced 30 83.5 + 18.7 108.0

Glass fiber-reinforced 30 204.3 + 15.7 2 6 5 . 4 <0.001

*Comparison of means by Student's t-test indicated the significant differ- ence (p<0.001)

TABLE 2: COMPARISON OF THE EFFECT OF THE VARIABLES OF TYPE II TEST SPECIMENS

Deflection Amplitude Variable n M + SD p

(A1) First loading cycle with 100 N force

Unreinforced 10 0.36 (mm) _+ 0.07

Glass fiber-reinforced 10 0.28 (mm) + 0.09

(A2) After 5 x 105 cycles with 100 N force

Unreinforced 10 0.45 (mm) + 0,11

Glass fiber-reinforced 10 0.29 (mm) + 0.08

(D,)*

Unreinforced 10 0.00 (mm) + 0.00

Glass fiber-reinforced 19 0.00 (mm) + 0.00

(D2)*

Unreinforced 10 0.07 (mm) + 0.02

Glass fiber-reinforced 10 0.07 (ram) _+ 0.03

0.034

0,003

1.000

*D~ and D 2 are parameters for abrasion and adaptation of the test specimen to bearing area (see Fig. 7)

deflection amplitude with a force of 100 N was determined after 5 x 105 cycles (A2). Thus, A~ was determined using the position of parameter D 2 as the reference point for measurement.

Mean values and standard deviations were calculated for the transverse strengths of Type I test specimens as well as for the varibles for fatigue weakening of the Type II test specimens. The means were compared by Student's t-test. Mean curves were also calculated for fatigue weakening of Type II test specimens.

R E S U L T S

Measurements of the transverse strengths of reinforced Type I specimens revealed an increase of 146% compared to unreinforced specimens. The mean values for these two types of specimens differed highly significantly (p<0.001) (Table 1).

The SEM micrographs (Fig. 5) of transverse sections of the test specimens established that the PMMA fiber composite was quite homogeneous. The distribution of the fibers in the matrix was even, and the fibers were surrounded by a layer of acrylic resin. At higher magnification (Fig. 5), single fibers could be observed. There was PMMA material between the fibers, forming a homo-

D 1 ( 0 c y c l e s )

D 2 ( 5 x 1 0 5 c y c l e s )

Fig. 7. Schematic representation of parameters D 1 and D 2 indicating abrasion of occlusal surfaces and adaptation of the test specimen during loading. The arrow indicates the direction of the cyclic force,

geneous structure of composite with only a few minor clefts between the fibers and the PMMA.

By ashing the test specimens, the concentration of glass fibers in the PMMA matrix was found to be 58 wt%. The weight of the fibers in the test specimens after polymerization was 0.448 g, which is the same as the amount incorporated (0.449 g) within experimental error.

The results of fatigue testing (Fig. 8) showed minor weakening of glass fiber-reinforced test specimens compared to unreinforced specimens after 5 x 105 cycles (p<0.001). The mean deflection amplitude (A 2) was lower for glass fiber-reinforced specimens than for unreinforced specimens (p<0.01) (Table 2). The means for parameter D 2 did not, however, differ significantly between reinforced and unreinforced specimens (p>0.05).

D I S C U S S I O N

According to many previous studies (Yazdanie and Mahood, 1985; Vallittu et al., 1994), it is difficult to increase the concentra- tion of fibers in a polymer matrix to the level at which an adequate strengthening effect is achieved. There have been three prob- lems: 1) lateral spreading of fibers when the acrylic resin dough is pressed into the mold, which diminishes the concentration of fibers in the polymer matrix and causes inhomogeneous distribu- tion of fibers in the matrix; 2) poor wetting of fibers by acrylic resin material. The layer of acrylic resin surrounding the single fibers in the middle of the fiber rovingis not even; and 3) polymerization shrinkage of PMMA destroys the layer of acrylic resin on the surface fibers and decreases the bond between the fibers and the polymer material.

In the pultrusion process investigated by Goldberg and Burstone (1992), the thermoplastic polymer materials (polyeth- ylene terephtalateglycol and poly-1,4-cyclohexylene dimethylene terephtalate glycol) that were used do not react well with PMMA. This may cause a problem if an attempt is made to use the pultrusion-processed fiber composite as a strengthener for den- ture-base material such as PMMA.

Dental Materials~March 1994 119

== F= v

0= L,.

#_

120

1 1 0

100

90

80

70

60

50

40

30

20

10

' " ' " " " . . . . . . ~

" ' " " " . . . . . . . . . . .

'"......

1 2

I I

3 4 S NUMBER OF CYCLES (XlO 5)

p<O.O01

Fig. 8. Mean curves (dotted lines indicate the SD) for fatigue weakening of Type II test specimens plotted as the number of loading cycles, GF=glass fiber (n=lO per group),

weakening of PMMA during loading was not measured. Since the force applied to the specimens decreased with cy-

cling under constant deflection, the tendency toward heating of the material also decreased. Therefore, constant deflection test- ing does not lead to thermal softening (Hertzberg and Manson, 1980). According to Beardmore and Rabinowitz (1975), the weakening of material during cyclic loading occurs in crystalline, amorphous and composite polymers. PMMA is an amorphous polymer (Higgins, 1990).

Fatigue weakening of denture base material in the oral environment increases the deflection amplitude when biting force is applied. Plastization of denture base material by the water of saliva (Higgins, 1990) also increases the deflection of denture bases. Deflection requires a stress-bearing area such as the toruspalatinus, resilient alveolarridges and sufficient occlusal biting force. Cyclic deflection may eventually lead to fatigue failure, e.g., mid-line fractures, in upper complete dentures.

The results of the present study suggested that by incorporat- ing continuous glass fibers into PMMA denture bases, the strength of the denture can be significantly increased and fatigue weaken- ing significantly reduced. Propagation of fatigue cracks in den- ture bases as well as the effect of environment over time on the strength should be further investigated.

Using hand placement of the fibers with the simple method described in this study, a high concentration of fiber can be achieved in the PMMA matrix. It is thus practical to use this method as part of the conventional dough-molding technique when acrylic-based dentures, such as the Type II test specimens used in the present study, are processed. By pretreating fiber rovings with a PMMA-MMA mixture, the effect of polymerization shrinkage within the fiber roving can be minimized. Minor shrinkage of PMMA allows the fibers to bond to each other via an even layer of PMMA (Vallittu, 1994). The amount of PMMA between the fibers is shown in Fig. 5. Ladizeskyetal. (1990,1992) used PMMA-MMA mixture as a dipping solution for polyethylene fibers but did not report how the polymerization shrinkage of the polymer material between fibers affected the strength of the composite.

It is important to note that only minor clefts were found between the fibers and the PMMA matrix because wide clefts would increase the absorption of water into the composite by capillary effect. Such absorption may induce corrosion of the composite by damaging the PMMA-fiber interface (Ehrenstein et al., 1990). Water attacks the surface of glass fiber and destroys the silicate structure of the glass fiber by leaching and hydrolysis. Thus, bonding of PMMA and glass fiber is affected, due to a decrease in the strength of the composite.

The fatigue test used in the present study revealed weakening of denture-shaped test specimens after cyclic loading. However, the results also indicated that, by reinforcing the denture base by the method developed here, a clinically significant increase in strength can be achieved. The initial force (100 N) applied to the specimens was somewhat lower than the maximal biting force of complete denture wearers (Lassila et al., 1985). The pneumatic testing machine for in vitro fatigue tests simulated the oral environment very well in terms of biting force, temperature, moisture, and form of the palatal area. The constant deflection test has also been used previously (Fujii, 1989) to investigate the fatigue properties of PMMA. In Fujii's study, however, the

Received April 12, 1993/Accepted February 27, 1994

Address correspondence and reprint requests to: Pekka K. Vallittu Faculty of Dentistry University of Kuopio P.O.B. 1627 SF-70211 Kuopio Finland

REFERENCES

Beardmore P, Rabinowitz S (1975). Treatise on materials science and technology. Arsenault RJ, editor. Vol 6. New York: Academic Press, 267.

Beyli MS, von Fraunhofer JA (1981). An analysis of causes of fracture of acrylic resin dentures. J Prosthet Dent 46:238- 241.

Clark HA, Plueddemann EP (1963). Bonding of silane coupling agents in glass-reinforced plastics. Modern Plastics 40:133- 138.

Ehrenstein GW, Schmiemann A, Bledzki A, Spaude R (1990). Corrosion phenomena in glass-fiber reinforced thermo- setting resins. In: Cheremisinoff NP, editor. Handbook of Ceramics and Composites Vol 1. New York: Marcel Dekker, 231-239.

Friskopp J, Bloml6f L, SSder P-O (1979). Fiber glass splints. J Periodontol 50:193-196.

Friskopp J, Bloml6fL (1984). Intermediate fiberglass splints. J Prosthet Dent 51:334-337.

Fujii K (1989). Fatigue properties of acrylic denture base mate- rials. Dent Mater 8:245-259.

Galan D, Lynch E (1989). The effect of reinforcing fibers in denture acrylics. J I r D e n t A s s o c 35:109-113.

Goldberg AJ, Burstone CJ (1992). The use of continuous fiber reinforcement in dentistry. Dent Mater 8:197-202.

120 Vallittu et al./Strength and fatigue of PMMA composite

Hertzberg RW, Manson JA (1980). Fatigue of Engineering Plastics. London: Academic Press, 55.

Higgins RA (1990). Properties of Engineering Materials. Lon- don: Edward Arnold Co., 259-262.

Inamura M, Nagato H, Aida E, Adachi T, Ishiguro H, Ono T, Kurosu K (1988). Self-cure dental acrylate resin reinforced with glass fiber - Part 1: Bending and tensile test. Jpn J Pedodontics 26:328-335.

Ladizesky NH, Chow TW (1990). The effect of highly drawn polyethylene fibres on the mechanical properties of denture base resins. Clin Mater 6:209-225.

Ladizesky NH, Ho CF, Chow TW (1992). Reinforcement of complete denture bases with continuous high performance polyethylene fibers. J Prosthet Dent 68:934-939.

Lassila V, Holmlund I, Koivumaa KK (1985). Bite force and its correlations in different denture types. Acta Odontol Scand 43:127-132.

Phillips RW (1982). Skinner's Science of Dental Materials. 8th ed. Philadelphia: WB Saunders Company, 39.

Rosen MR (1978). From treating solution to filler surface and beyond. The life history ofa silane coupling agent. JCoat Tech 50:70-82.

Schipp C, RSber S, Zachmann HG, Seferis JC (1992). Determina- tion of the fiber content in polymer composites by means of X- ray absorption measurements. J AppI Polym Sci 44:1631- 1634.

Schwickerath H (1965). Glasfaser und Kunststoffverst~irkung. Zahn(trztl Welt / Ref 66:364-367.

Setz J, Lin W (1989). Biegefestigkeit Kurzfaservert~irkter

Provisorienkunststoffe. Schweiz Monatsschr Zahnmed 99:544-546.

Skirvin DR, Vermilyea SG, Brady RE (1982). Polymethy- lmethacrylate reinforcement: Effect on fatigue failure. Mil Med 147:1037-1040.

Smith DC (1961). The acrylic denture: Mechanical evaluation of mid-line fracture. B r D e n t J 110:257-267.

Solnit GS (1991). The effect of methyl methacrylate reinforce- ment with silane-treated and untreated glass fibers. JProsthet Dent 66:310-314.

Talreja R (1989). Fatigue of composites. In: Kelly A, editor. Concise Encyclopedia of Composite Materials. Oxford: Pergamon Press, 77.

Vallittu PK, Lassila VP (1992a). Reinforcement of acrylic den- ture base material with metal or fiber strengtheners. J Oral Rehabil 19:225-230.

Vallittu PK, Lassila VP (1992b). Effect of metal strengtheners surface roughness on fracture resistance of acrylic denture base material. J Oral Rehabil 19:385-391.

Vallittu PK, Lassila VP, Lappalainen R (1994). Acrylic resin- fiber composite - Part I: The effect of fiber concentration on fracture resistance. JProsthet Dent 71:607-612.

Vallittu PK (1994). Acrylic resin-fiber composite - Part II: The effect of polymerization shrinkage of polymethyl methacry- late applied to fiber roving on the transverse strength. J Prosthet Dent 71:613-617.

Yazdanie N, Mahood M (1985). Carbon fiber acrylic resin com- posite: An investigation of transverse strength. J Prosthet Dent 54:543-547.

Dental Matenals/March 1994 121