Dynamic mechanical properties of multiphase acrylic systems Harry 0ysaed NIOM, Scandinavian Institute of Dental Materials, Haslum, Norway

The influence of type and quantity of five different dimethacrylate crosslinking agents on the dynamic mechanical properties of multiphase acrylic systems h a s been studied. These materials, commonly used in bioengineering, were processed by polymerization of a mixture of liquid methacrylate monomers, and poly(methy1 methacrylate) powder. The specimens were made with various ratios of methyl methacrylate and dimethacrylate crosslinking agents in the monomer liquid. Two different processing conditions were used, heat-polymerization at 100°C and autopolymerization at 45°C. By using a forced torsional vibration apparatus the

storage modulus (G’), loss modulus (G”), and dissipation factor (tan 6) were determined over the temperature range -60°C to 140°C at frequencies of 0.1, 1.0, 10, and 100 rad/s. In the autopolymerized materials, the glass transition temperature (TJ, as determined via tan S data, increased with increasing quantities of crosslinking agents. The storage modulus likewise increased. In the heat-polymerized materials only minor variations in modulus and tan 6 with type and quantity of crosslinking agents were observed. T g values of the heat-polymerized materials were, in all cases, greater than those of the autopolymerized materials.

INTRODUCTION

By dynamic mechanical thermal analysis (DMTA) the response of a material to a sinusoidal or other periodic stress is determined. Over a wide temperature and frequency range DMTA is sensitive to the chemical and physical structures of polymers.’ The observed shifts in dynamic mechanical properties often correlate with change in polymer chain mobility. Braden and coworkersz4 investigated different denture base materials and concluded that the dynamic mechanical properties were highly informative on the chemical composition and structure of the materials, as well as on the influence of free monomer content, internal plastization, and degree of crosslinlung. The viscoelastic properties of different dental polymeric materials have been studied by dynamic mechanical analysis. An increased glass transition temperature, TR, in heat-polymerized denture base polymers Dynamic compared to autopolymerized materials has been dem~nstrated.~ mechanical properties of resin-based filling materials have been investigated6and a Tgas low as 10°C of a photopolymerized proprietary composite has been r e p ~ r t e d . ~ A multiphase acrylic system is formed by polymerization of a mixture of liquid methacrylate monomers and poly(methy1 methacrylate) (PMMA) Journal of Biomedical Materials Research, Vol. 24, 1037-1048 (1990) 0 1990 John Wiley & Sons, Inc. CCC 0021-9304/90/081037-12$04.00

0YSiED

1038

p ~ w d e r .An ~ , ~interpenetrating polymer network (IF") is formed if the monomer contains a crosslinking agent, and the product is called a semiIF" structure." The mechanical behavior of multiphase polymers depends not only on the behavior of each phase, but also on the molecular mixing at the phase boundaries as well as within the phase structures. Incorporation of different types and quantities of crosslinking agents have different effects on mechanical properties such as tensile strength," elastic m ~ d u l u s ,ten~~,~~ sile creep,9and impact resistance.'4 Creep studies of multiphase acrylic systems showed that the type and quantity of crosslinking agents had no major effect on tensile creep at low stress levels.' However, at high stress levels the creep values varied with both type and quantity of crosslinking agents. In heat-polymerized materials the failure mode changed from brittle to ductile when the test temperature was increased from 37°C to 50°C while the autopolymerized materials failed in a ductile manner both at 37°C and 50"C.9The purpose of this study was to investigate the influence of type and quantity of crosslinking agents on dynamic mechanical properties of multiphase acrylic systems and to evaluate these results within the context of recent creep ob~ervations.~ MATERIALS AND METHODS

Specimen preparation The materials were processed by polymerization of a mixture of 30 wt% methacrylate monomers and 70 wt% uncrosslinked PMMA powder (Ivoclar). The specimens were made with various ratios of MMA and crosslinking agent in the monomer liquid (Table I). Dimethacrylates of ethyleneglycol (EGDMA), 1,3-propanediol (PDMA), 1,4-butanediol (BDMA), diethyleneglycol (DEGDMA), and triethyleneglycol (TEGDMA)were used as crosslinking agents. The different monomers were stabilized with varying quantities of the inhibitors hydroquinone and methylhydroquinone by the manufacturers. These were removed by washing a 50% solution of monomer in pentane (p.a. Merck)/dichloromethane (HPLC grade, Rathburn Chem. Ltd.) (9/1) with an equal volume of 1N NaOH followed by washing with water TABLE I Composition of the Monomer Liquid MMA (mol%)

Crosslinking Agent" (mol%)

100 98.75 97.5 95.0 90.0

"EGDMA, PDMA, BDMA, DEGDMA, or TEGDMA.

0 1.25 2.5 5.0 10.0

DYNAMIC MECHANICAL PROPERTIES

1039

and drying with CaSO,. The solvents were removed from the dimethacrylates by distillation at low pressure and at room temperature. Two different processing conditions were used, heat-polymerization and autopolymerization. When heat-polymerized, the PMMA powder containing 0.5 wt% benzoyl peroxide (BPO) was mixed with the monomer liquid. After 30 min at 22°C the resulting dough was pressed into a gypsum mold with internal dimensions of 65 mm X 40 mm (top), 64 mm x 39 mm (bottom), length and width respectively, and 5 mm in height, submerged in water at 73 k 1°Cfor 90 min and then immersed in boiling water for 30 min. When autopolymerized, the PMMA powder containing 0.5 wt% BPO and the monomer liquid with 0.75 wt% N,N-dimethyl-p-toluidine (DMPT) were mixed for 20 s. After 2 min the mixture was placed in a gypsum mold and kept at 45 k 1°C for 30 min at a pressure of 220 KPa. Test specimens with a length of 60 mm, a width of 10 mm and a thickness of 2 mm were made by cutting the polymerized plates and polishing with A1,0, paste (Buehler Ltd., Greenwood, IL, USA). Two pastes with grain sizes 0.3 and 0.05 pm were used. The test specimens were stored at 22 k 1°Cand 50 2 5% relative humidity for 1 week prior to testing. Dynamic mechanical measurements All dynamic mechanical properties were determined using a Rheometrics Mechanical Spectrometer, RMS-605 (Rheometrics, Inc., Piscataway, NJ, USA). This torsion instrument works in the temperature range -160 to 350°C and the frequency range of lo-* to 16 rad/s. The temperature of the sample remains constant to within 0.2"C, minimizing errors arising from the temperature variations. The technique of dynamic mechanical testing measures the response of a viscoelastic material to a small sinusoidal strain. The resultant data are the real (G') (storage modulus) and the imaginary (G") (loss modulus) part of the complex shear modulus and the dissipation factor tan 6 = G"/G', as a function of frequency and temperature. The tan 6 is a damping term and is a measure of the ratio of energy dissipated as heat to the maximum energy stored in the material during one cycle of oscillation. The investigated materials were tested over a temperature range of -60 to 140°C. G', G", and tan 6 were measured at frequencies of 0.1, 1.0, 10, and 100 rad/s. Measurements were taken at 10" intervals below 70°C and at 5" intervals above 70°C. The oscillation strain was changed from 0.1% to 0.25% at 90°C and higher temperatures. RESULTS

The dependence of the modulus G ' and tan 6 on temperature and frequency for the heat polymerized material without crosslinking agents are

0YSED

1040

shown in Figures 1 and 2. At the main, or glass transition the modulus dropped by a factor of 1000 (Fig. 1) and in the region of this transition the material changed from glassy to rubbery. This transition was accompanied by a peak in tan 6 (Fig. 2). This peak is called the a-peak, and the temperature corresponding to the maximum tan 6 value, T,. Apart from this main transition, a secondary (p) transition in the glassy region was observed with only a minor decrease in the modulus but with a distinct maximum in the tan 6 curve. Figure 2 also shows that with increasing frequency the a and @ transitions changed to higher temperatures, The @ transition was shifted more rapidly than the CY transition. The plot of loss modulus G" as a function of temperature allowed a better separation of the a- and @-maxima(Fig. 3 ) . Figure 3 shows a minor shift between the @ transitions of heat-polymerized and autopolymerized materials

-40

0

40

a0

120

TEMPER ATURE,'C

Figure 1. Storage modulus G ' as a function of temperature at frequencies 0.1 (e), 1.0 (m), 10 (x), and 100 (A) rad/s for heat-polymerized uncrosslinked PMMA.

1041

DYNAMIC MECHANICAL PROPERTIES

-40

0

40

80

120

TEMPERATURE,%

Figure 2. Dissipation factor tan 8 as a function of temperature at frequencies 0.1 (e), 1.0 (w), 10 (x), and 100 (A) rad/s for heat-polymerized uncrosslinked PMMA.

-40

0

40

80

120

TEMPERATURE,‘C

Figure 3. Loss modulus G ” as a function of temperature at frequencies 0.1 (0.0)and 100 (A,A) rad/s for heat-polymerized (.,A) and auto-polymerA) uncrosslinked PMMA. ized (0,

OYSWD

1042

tested at the same frequency, although a pronounced shift in the a transitions was found. In Table I1 the variation in T , and T , with frequency, for the materials without crosslinking agents, determined from tan 6 and G” plots are given. Only minor differences in modulus and dissipation factor were observed with incorporation of the different crosslinking agents when the materials were heat-polymerized. The a-peaks in tan 6 were all narrow, but the materials with 10 mol% crosslinking agents showed broader a-peaks than did the material without crosslinking agent (Fig. 4). In the autopolymerized materials storage modulus G ’ and T, increased with increasing quantity of all crosslinking agents (Figs. 5 and 6). The tan 6 curves of these materials also exhibited broad a-peaks which might be multiple. Only minor differences in modulus and dissipation factor with type of crosslinking agent were registered. The exception was the material with 10 mol% of TEGDMA as illustrated in Figure 7. For this material a decrease in modulus between -10°C and 70°C was registered. By testing both the heat- and autopolymerized materials in triplicate the reproducibility of the moduli and tan 6 was found to be within 22%below T,. DISCUSSION

The dynamic mechanical properties of multiphase acrylic polymers were dependent on frequency as shown in Figures 1 through 3 and Table 11. The change from glassy to rubbery state is very often related to the glass transition temperature, Tg. TRcan be determined via different experimental techniques (i.e., DMTA, differential scanning calorimetry, thermo mechanical analysis) and also by different calculations of the DMTA results (i.e., G ’, G”, tan 6). By DMTA, differences as high as 85°C between T g by reference to G’ and to tan 6 are r e p ~ r t e dThe . ~ present results demonstrated the necessity of specifying how the transition temperatures are determined.

TABLE I1 a-and &Transitions of Uncrosslinked PMMA Systems

Heat Heat Heat Heat Auto Auto Auto Auto

0.1 1.0 10.0 100.0 0.1 1.0 10.0 100.0

”Determined from tan 6 data. bDetermined from G ” data.

120 123 128 133 93 110 120 130

109 112 115 118 78 87 92 97

- 10

-20

5 30 60 -12 5 30 63

-5 20 40 -20 -4 20 35

1043

DYNAMIC MECHANICAL PROPERTIES

0

-40

0

40

80

120

TEMPERA T U R E , ~

Figure 4. Dissipation factor tan 6 as a function of temperature at 0.1 rad/s for a heat-polymerized material without crosslinking agent ( 0 ) and with 10 mol% of DEGDMA (A).

In methacrylate based polymers the p-transition may be related to the rotation of the -COOR side group around the bond linking it to the main chain.15The transition seems to be independent of polymerization conditions (Table 11) as well as type and quantity of crosslinking agents (Fig. 6). A single, narrow a-transition peak in tan 6 (Fig. 4) indicated molecular mixing between the phases of the IPN structure." This was in agreement with the results from earlier microscopic and solubility investigations of heat-polymerized It is known that crosslinking increases the Tg of a polymer by introducing restrictions on the molecular motions of a chain.' In ideal polymeric networks there is a linear variation of T , with crosslinking density.16 An increase in added crosslinking agent in the monomer liquid probably results in increased crosslinking density. Earlier investigations have demonstrated an increase in gel fraction with quantity of crosslinking agents in heat-polymerized materials for all the dimethacrylates used in this investigation.' By mixing amorphous systems it has been shown that the measured (dynamically) 7,are related to the Tgls of the pure components according to the equation In T , = m, In Tgl + m 2In Tg2 where m is wt% of each ~omponent.'~ This indicates an increase in T, with increasing quantity of crosslinking agents. However, no increase in T, with in-

1044

-40

0

40

80

120

TEMPERATURE.%

Figure 5. Storage modulus G ' as a function of temperature at 0.1 rad/s for autopolymerized materials without crosslinking agent ( 0 ) and with 1.25 (0), 2.5 (m), 5 (o), and 10 (A) mol% of DEGDMA.

creased quantity of crosslinking agents in the heat-polymerized materials was observed in the present study (Fig. 3). The reason may be competing effects. One factor which may depress Tg is the incorporation of plasticizers." Both residual unreacted monomers and pendant methacrylate groups (PMG) may act as plasticizer^."^^'^^^^ The number of PMG increases with increasing quantity of crosslinking agents. An increase in both crosslinking density and PMG may be the reason why there was a slight broadening in a-peak and no changes in T, with increasing quantity of crosslinking agents in the heatpolymerized materials. Although the results showed little variation in p-transition between heatand autopolymerized materials, a great variation in a-transitions was found (Fig. 3, Table 11). At a frequency of 0.1 rad/s a decrease of approximately 30" in T, was observed between heat-polymerized and autopolymerized materials. Kusy and Greenberg" have investigated the influence of molecular weight on the dynamic mechanical properties of PMMA over a wide range

1045

DYNAMIC MECHANICAL PROPERTIES

-40

0

40

80

120

TEMPERATURE,C

Figure 6. Dissipation factor tan 6 as a function of temperature at 0.1 rad/s for autopolymerized materials without crosslinking agent (0) and with 1.25 (o),2.5 (w), 5 (o),and 10 (A) mol% DEGDMA. The scale on the tan 6 axis is not shown, as the curves are artificially displaced vertically to facilitate comparison of pertinent differences.

of molecular weights (1500 < M , < 600,000), and reported an increase in T, with molecular weight. An earlier investigation has demonstrated higher molecular weight in heat-polymerized compared to autopolymerized materials without crosslinking agents.' The much higher values of residual MMA, acting as plasticizer, in the autopolymerized materials (2.3 wt%) compared to the heat-polymerized materials (0.6 ~ t %may ) ~also result in depression of Ta.18Another important factor may be the variation in stereoregularity between auto- and heat-polymerized materials. The glass transition temperature depends on the tacticity of the polymer, increasing with increasing stereoregularity.' A high degree of syndiotacticity in PMMA denture base materials has been reported."-24 Optical micrographs of the autopolymerized materials show a multiphase system consisting of a dispersed phase formed from the polymer beads originally present in the PMMA powder and a matrix formed from the monomer liquid.',' The short period between mixing and polymerization of the autopolymerized materials prevented an efficient penetration of monomers into the PMMA beads, resulting in less IPN formation. This is illustrated by a broad, multiple a-peak in tan 6 (Fig. 6). In uncrosslinked materials a difference between M, = 190,000 in PMMA powder, and M , = 56,000 in autopolymerized MMA in bulk has been demonstrated.' Furthermore, the degree

MYSWD

1046

-40

0

40

80

120

TEMPERATURE’C

Figure 7. Storage modulus G ’ as a function of temperature at 0.1 rad/s for autopolymerized materials without crosslinking agent (0) and with 10 mol% of TEGDMA (0) and DEGDMA (A).

of unreacted monomers may be different in the various phases. The result of these effects may be splitting of the a-peak in tan 6 as most clearly seen in the uncrosslinked material (Fig. 6). Although the effect of adding crosslinking agents to the monomer liquid may both increase Tgdue to crosslinking and decrease T g due to PMG, the results demonstrate that the total effect in the autopolymerized materials is an increase in Tg with increasing quantity of crosslinking agents. Creep studies of multiphase acrylic systems show that among both heatpolymerized and autopolymerized materials, the systems with DEGDMA clearly deviated from the others by showing higher creep values.’ No change in dynamic mechanical properties in the materials with DEGDMA compared to the other crosslinking agents was found in this investigation. The only variation in results with type of crosslinking agent was registered in the autopolymerized materials with 10 mol% of TEGDMA (Fig. 7). In this material a drop in G ’ between -10°C and 70°C compared to the other auto-

DYNAMIC MECHANICAL PROPERTIES

1047

polymerized materials with the same quantity of crosslinking agent was registered, probably due to the long, flexible chain between the two methacrylate groups in TEGDMA. The author wishes to thank Trine-Lise Rolfsen, Center for Industrial Research, Oslo, for her skillful technical assistance.

References 1. L. E. Nielsen, Mechanical Properties of Polymers and Composites, Marcel

Dekker, Inc., New York, 1974. 2. M. Braden and G. D. Stafford, ”Viscoelastic properties of some denture base materials,” J. Dent. Res., 47, 519-523 (1968). 3. M. J. Barsby and M. Braden, ”Visco-elastic properties of pour (fluid) denture base resins,” J. Dent. Xes., 60, 146-148 (1981). 4. R. L. Clarke and M. Braden, “Visco-elastic properties of some roomtemperature polymerizing resins,” J. Dent. Res., 61, 997-1001 (1982). 5. W. Finger, “Mechanisch-dynamische Eigenschaften von ProthesenKunststoffen,” Dtsch. Zuhniirztl. Z . , 30, 665-671 (1975). 6. R. Whiting and P. H. Jacobsen, “Dynamic mechanical properties of

resin-based filling materials,” J. Dent. Res., 59, 55-60 (1980). 7. T. W. Wilson and D. T. Turner, ”Characterization of polydimethacrylates and their composites by dynamic mechanical analysis,“ 1. Dent. Xes., 66, 1032-1035 (1987). 8. H. 0ysaed and I. E. Ruyter, ”Formation and growth of crazes in multiphase acrylic systems,” I. Muter. Sci., 22, 3373-3378 (1987). 9. H. 0ysaed and I. E. Ruyter, ”Creep studies of multiphase acrylic systems,” J. Biomed. Muter. Res., 23, 719-733 (1989). 10. L. H. Sperling, Interpenetrating Polymer Networks and Related Materials, Plenum Press, New York, 1981. 11. E. M. Wollff, “The effect of cross-linking agents on acrylic resins,” Aust. Dent. I . , 7, 439-444 (1962). 12. L. J. Broutman and F. J. McGarry, ”Fracture surface work measurements on glassy polymers by a cleavage technique. 11. Effects of crosslinking and preorientation,” J. Appl. Polym. Sci., 9, 609-626 (1965). 13. M. Atsuta and D. T. Turner, “Strength and structure of glassy networks formed from dimethacrylates,” Polym. Eng. Sci., 22, 438-443 (1982).

14. C. A. Price, ”The effect of cross-linking agents on the impact resistance of a linear poly(methy1 methacrylate) denture-base polymer,” J. Dent. Res., 65, 987-992 (1986). 15. J. Heijboer, ”Secondary loss peaks in glassy amorphous polymers,” Int. J. Polym. Muter., 6, 11-37 (1977). 16. F. Rietsch, D. Daveloose, and D. Froelich, “Glass transition temperature of ideal polymeric networks,” Polymer., 17, 859-863 (1976). 17. J. M. Pochan, C. L. Beatty, and D. F. Pochan, ”Different approach for the correlation of the Tg of mixed amorphous systems,” Polymer, 20, 879-886 (1979). 18. S. Kalachandra and D. Turner, ”Depression of the glass transition temperature of poly(methy1 methacrylate) by plasticizers: Conformity with free volume theory,” J. Polym. Sci. Purt B., 25, 1971-1979 (1987). 19. R. P. Kusy, J. R. Mahan, and D. T. Turner, “Influence of application technique on microstructure and strength of acrylic restorations,” J. Biomed. Muter. Res., 10, 77-89 (1976). 20. I. E. Ruyter and H. Myszd, ”Conversion in denture base polymers,” J. Biomed. Muter. Res., 16, 741-754 (1982).

0YSXD

1048

21. R. P. Kusy and A. R. Greenberg, "Influence of molecular weight on the dynamic mechanical properties of poly(methy1methacrylate)," f . Therm. Anal., 18, 117-126 (1980). 22. I. E. Ruyter and S . A. Svendsen, "Flexural properties of denture base polymers," f . Prosthet Dent., 43, 95-104 (1980). 23. C. Bessing, B. Nilsson, and M. Bergman, "SR 3/60 and SR-Ivocap. A comparison between two heat cured denture base resins with dissimilar processings," Swed. Dent. f . , 3, 221-228 (1979). 24. R. Huggett, S. C. Brooks, and I. H. Sadler, "Stereochemistry of poly(methyl methacrylate) acrylic resin denture base material," Biornateriuls, 5, 118-119 (1984).

Received September 29, 1989 Accepted December 19, 1989

Dynamic mechanical properties of multiphase acrylic systems.

The influence of type and quantity of five different dimethacrylate crosslinking agents on the dynamic mechanical properties of multiphase acrylic sys...
537KB Sizes 0 Downloads 0 Views