Reinforcement of poly(methyl methacrylate)denture base with glass flake

Paul Franklin*, David J. Wood, Nigel L. Bubb

Division of Restorative Dentistry, Leeds Dental Institute, Clarendon Way, Leeds LS2 9LU, UK

Received 11 December 2003; received in revised form 20 May 2004; accepted 13 July 2004

01do

34

KEYWORDSDenture base;Double torsion;Fracture toughness;PMMA;Reinforcement

09-5641/$ - see front matter Q 200i:10.1016/j.dental.2004.07.002

* Corresponding author. Tel.: C44 136165.E-mail address: p.franklin@leeds.a

Summary Objectives: Since the introduction of poly(methyl methacrylate) as adenture base material, it has suffered from having relatively poor mechanicalproperties. Many methods of improving its strength and toughness have beeninvestigated. Most of these have not been adopted due to: increased cost, the needfor specialist processing equipment or increased laboratory time due to morecomplicated procedures.

Glass flake has been used as a reinforcing agent in many industrial polymers, but isas yet untested with denture base acrylic materials. The aim of this study is toevaluate the effect of adding glass flake to denture base acrylic powder on thefracture toughness of the set material.

Methods: Glass flake was added in 5, 10 or 20% w/w to Trevalonw denture basepowder. The material was mixed, flasked, packed and processed in a manner typicalfor a denture base material. Fracture toughness was determined using a doubletorsion test technique.

Results: The addition of glass flake gave up to a 69% increase in fracture toughnesscompared to plain Trevalonw material. The addition of 5% glass flake lead to animprovement in fracture toughness that was statistically significant compared toboth plain Trevalonw and the 10 and 20% groups.

Significance: The significant improvement in fracture toughness of a denture baseacrylic material using glass flake is an extremely promising result. Other mechanicalproperties will require testing before glass flake can be recommended as areinforcing agent for denture base acrylic materials.Q 2004 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

4 Academy of Dental Materials

13 3436167; fax: C44 113

c.uk (P. Franklin).

Introduction

Fracture in poly(methyl methacrylate) (PMMA)denture base is a time consuming and costlyproblem. As well as being inconvenient for patients,

Dental Materials (2005) 21, 365–370

www.intl.elsevierhealth.com/journals/dema

. Published by Elsevier Ltd. All rights reserved.

P. Franklin et al.366

fractured denture portions can pose a significanthealth hazard if inhaled [1].

Dentures fracture essentially due to factorsleading to stress concentration and increasedflexing. The majority of denture fractures occurby the end of 3 years in service [2] and fracture mayoccur slowly during function or more rapidly if, e.g.the denture is dropped on a hard surface. Theclassical midline fracture has been extensivelystudied [3]. The factors underlying any particulardenture fracture can be difficult to ascertain due tothe very large number of variables includingfunction, processing and handling of the denture.Darbar et al. [4] reported on the results of aquestionnaire of types of fracture encountered indental laboratories, finding 34% of repairs to becarried out on upper complete dentures, and 58% ofthose repairs were to midline fractures, i.e. 20% ofthe entire sample.

Methods to improve the inherent materialproperties of PMMA have included using alternatepolymers such as polycarbonate [5], nylon [6],co-polymers [7–9] and reinforcing agents includingparticulates and fibres [10].

The most commercially successful method ofreinforcement to date is rubber toughening, how-ever, such materials have compromised flexuralproperties [11]. A rubber phase may be introducedinto the acrylic matrix as either co-polymer, rubberparticles or as core-shell particles. Rodford [8]described the dissolving of the butadiene–styreneco-polymer in methyl methacrylate monomer.There are several mechanisms in operation thataccount for the toughening behaviour of rubber-toughened PMMA including shear yielding andcavitation.

Core-shell particles are a development of therubber toughening process, utilising the arrange-ment of rubber and acrylic that imparts the great-est toughness upon the acrylic—namely a core ofrubber phase material, surrounded by an acrylicshell [12]. The structure has been developedfurther to produce particles with three or morelayers, the structure of which can be varied tomatch the material being toughened. Murakamiet al. [13] recommended a particle structure of acentral PMMA core, surrounded by a butyl-metha-crylate inner shell, with a PMMA outer shell.Particulate fillers such as ceramic or UHMWPEbeads have been tested, but have failed to provideany significant improvement in mechanical proper-ties [14,15].

Having reviewed the problem of denture fractureand investigated the current situation with regardto the prevalence and type of fractures, thereclearly remains a need for a method of improving

mechanical properties of PMMA that does notimpinge too heavily on cost in terms of materials,technical time and the need for costly additionalspecialist equipment. The goal of research in thisarea is to formulate a material that can be used inexactly the same way as conventional denture basePMMA.

Glass flake (Glassflake Ltd, Leeds, UK) is a highaspect-ratio reinforcing additive with many com-mercial applications. The flake is a modified ‘C’glass composition and is supplied in a range of threethicknesses: 3.5–5.5, 1.9–2.5 and 1.4–1.9 mm. Thereare also three particle size distributions to choosefrom: unmilled, milled and micronised. As yet noliterature exists regarding its ability to reinforceacrylic, though the manufacturers claim that itsaddition to some thermoplastics has resulted insignificantly improved flexural modulus and planarreinforcement. They also claim the effect of addingto PTFE outperformed glass fibre reinforced PTFE interms of: tensile strength, compressive modulus,dimensional stability, wear resistance and creep.Clearly if such properties could be improved indenture base materials, the potential benefits aregreat.

The objectives of the present study were toinvestigate the ease of addition of glass flake toregular denture base acrylic, to determine theability of the material to take a polish and todetermine the fracture toughness of a controldenture base material and ones with glass flakeadditives in various concentrations.

Materials and methods

Trevalon (Dentsply Ltd, Weybridge, UK) was chosenas both the control material and the agent to bereinforced. The material has been described andtested by Murphy et al. [16]. They listed thecomponents of the powder as 93% polymer and0.4% benzoyl peroxide. The monomer liquid alsocontains 6% ethylene glycol dimethacrylate and0.006% hydroquinone. The fracture toughness ofTrevalon has been previously investigated [17].

Untreated, micronised glass flake (product codeGF003) was chosen for the test. This consists offlake particles 1.9–2.5 mm thick, and a range ofdiameters of which 88% were below 50 mm.

Moulds were prepared in standard denture flasksusing a template measuring 80!65!4 mm3. Allspecimens were processed in a water bath, byraising the water temperature over 6 h to boiling.The water was then held at boiling for 4 h, thenallowed to cool.

Reinforcement of denture base with glass flake 367

The glass flake was added in amounts of 5, 10 and20% by mass of powder with a constant amount ofliquid. Concentrations higher than this were tooviscous to successfully mix. The flake was eitherfirst mixed with the powder, and then added to theliquid. Following the curing cycle, samples werekept at 100% humidity and stored in an incubator at37 8C for 14 days aging prior to testing.

Double torsion testing

Specimens for double torsion testing weremachined from blocks produced in the flasks aspreviously described. Each flask sample was able toproduce two double torsion specimens. The speci-mens had the dimensions 65!25!3 mm3. Thesewere shaped on a milling machine, and then finishedby hand using 400 and then wet-ground with 1200grit wet and dry paper (3M, Bracknell, UK). Alldimensions were within a tolerance of G0.2 mm.The specimen dimensions over 20 mm weremeasured using a Mitutoyo series 500 ABSolutedigital calliper, with an accuracy of G0.02 mm.Specimen depth was measured using Mitutoyoseries 293 Digimatic micrometer with an accuracyof G1 mm (Mitutoyo (UK) Ltd, Andover, HampshireUK). Once the surface finish had been created, alongitudinal 908 ‘vee’-shaped groove was milledcentrally along one of the faces of the specimen.The groove was 1 mm deep leaving a residualthickness (tn) of 2 mm, with a right-angled profile.At one end of each specimen, another vee-shapednotch was cut to aid the crack initiating within thelongitudinal groove. A specimen is shown schema-tically in Fig. 1. Six specimens per group wereprepared.

Testing was carried out using a Lloyd LR 10K(Lloyd Instruments, Fareham, UK) bench mounteduniversal testing machine (UTM). The UTM wasUKAS accredited, as was all measuring equipment.A 500 N load cell with an accuracy of 0.5% was fitted

Figure 1 Geometry of double torsion specimen.

to the UTM. The tests were carried out using across-head speed of 0.1 mm minK1

The double torsion testing rig comprised two3 mm diameter rollers spaced 20 mm apart. Thesample was positioned centrally upon the rollerswith the groove on the lower surface parallel to thesupports. The lead-in notch, which was firstsharpened with a razor blade, was positioned atthe end of the rollers directly beneath the loadingnose. The loading nose consisted of two stainlesssteel spheres, 3 mm diameter, set into a stainlesssteel head, attached to the cross-head of the UTM.The loading nose was allowed to swivel in one planesuch that forces were evenly distributed betweenthe two spheres. The rig was set up so that theplane of the spheres was perpendicular to the tworollers. The force required to propagate the crack(Pc) was taken from the chart produced during thetest.

Fracture toughness was calculated after Hillet al. [17] using the equation

KIc Z PcWm3ð1 CgÞ

W3t tn

� �

where KIc, fracture toughness (M N mK3/2); Wt,width (m); Wm, moment arm (m); t, specimenthickness (m); tn, specimen thickness in plane ofcrack (m); g, Poisson’s ratio; Pc, load at fracture(N).

The arrangement of a specimen within thetesting rig is shown in Fig. 2. The lead in notch,and underside groove can be seen to be placedcentrally between the spheres of the loading nose,and the rollers.

At the end of each test, the two halves of the testsamples were recovered, and tn was measured usinga toolmakers microscope (Mitutoyo (UK) Ltd,

Figure 2 Photograph of sample positioned on the rigduring testing.

Table 1 Results table showing fracture toughness values (KIc) for each group (M N mK3/2).

5% 10% 20% Control

Mean KIc (M N mK3/2) 1.972 1.858 1.774 1.168SD 0.032 0.121 0.062 0.018

P. Franklin et al.368

Andover, UK). They were then re-wrapped withtissue paper and replaced in the incubator asbefore. Subsequently, one of the broken piecesobtained in each group was examined under SEM(JEOL USA, Inc., Peabody, MA, USA).

Results

The groups with higher amounts of added flake hadan appearance that could variously be described asgrainy or even sparkly depending upon the surfacefinish of the sample, and the way the light caught it.SEMs of polished surfaces revealed the flakes to bepolished flush with the resin matrix. Fracturetoughness values for each of the groups are shown

Figure 3 (a) (left) Fractured surfaces of control specimensurfaces of 5% group specimen at 60! and 600! magnificatiomarked contrast to the relative smoothness seen for the con

in Table 1. Values were improved by between 52and 69%.

Statistical analysis of the results was carried outusing SPSS for Windows software. An ANOVA testwas performed on the data, followed by a Tukey’spost-hoc test to see where any differences lay. Thisrevealed that for each test group, the fracturetoughness was significantly higher than the controlgroup and furthermore that the 5% group had asignificantly higher fracture toughness than the 10and 20% groups.

Fig. 3a and b show the fractured surface of anunfilled control specimen and that of a 5% testspecimen following double torsion testing, respect-ively. In each set the upper view is a close up of thearea within the highlighted square of the lowerpicture. Both images show the fractured surface

at 60! and 600! magnification. (b) (right) Fracturedn. The roughness of the surface of the filled specimen is introl.

Reinforcement of denture base with glass flake 369

extending to the base of the groove. The roughnessof the surface is in marked contrast to the relativesmoothness seen for the control sample. Glassflakes can be seen on the fractured surface as canflat areas of acrylic where the flakes have been lost.

Discussion

Following curing, the glass flake imparted a granularappearance to the acrylic. On close examination thesurface was very smooth, the flake material beingpolished level with the acrylic surface.

All samples behaved predictably within thetesting rig during fracture toughness measurement.All cracks propagated along the groove with nodeviation evident. Control samples produced nearideal charts in which the period of crack growthremained very stable. Pc was relatively straightfor-ward with these samples. Working out Pc for thetest samples proved more difficult as the area of thechart relating to crack growth did not remain stableand usually rose as the test proceeded. In thesecircumstances, Pc was taken as a mean of the forcesrecorded at the start of crack propagation—justafter pop in. During mixing, the consistency of theunset material with 10 and 20% flake added wasnoticeably thicker, and stirring was difficult. Themixture was also found to be tackier to the touchwith higher concentrations of glass flake.

The determination of the critical stress intensityfactor KIc is only a reliable means of measuringfracture toughness if the material being testedconforms to linear elastic fracture mechanics(LEFM), and fractures in mode I plane-strainmanner. The addition of substances to improvethe mechanical properties of PMMA, such as cross-linking agents or materials for reinforcement, maycause the PMMA to fracture in a way that is notpurely mode I, and so LEFM should not be applied.Stafford et al. [18] discovered that high impactacrylics did not fail catastrophically using tapered-cleavage and single edge notch experiments. Thismeant that KIc could not be determined for thosematerials. Hill et al. [17] determined KIc usingcompact tension and double torsion experiments,and found that a rubber toughened denture baseacrylic (Lucitone 199) adhered to the laws of LEFMsufficiently for KIc to be satisfactorily calculated.

Compared to other tests of fracture toughness,the specimen preparation for the double torsiontest is relatively simple, not requiring complexmoulds or machinery, though when using the test ondental composites, the testing procedure has beenfound to be difficult with success rates at about 50%

[19]. The current study produced a success rate of100% confirming this test to be suitable for acrylicmaterials.

For the purposes of this study, a Poisson’s ratio of0.33 was adopted in order for direct comparisons tobe made to the results obtained by Hill et al. [17],but with the recommendation that the Poisson’sratio of denture base acrylic is investigated shouldfuture work in this area be required.

A preliminary tests of Perspexe gave a KIc of1.273 M N mK3/2. This was very close to the1.3 M N mK3/2 obtained by Hill et al. [17]. In thesame study Hill obtained a KIc for Trevalon of1.51 M N mK3/2. This is considerably higher than the1.168 M N mK3/2 measured in the control groups,and may be explained by the materials coming fromdifferent batches or different curing cycles used.

In this study, the fracture toughness measure-ments of all of the test samples were significantlyhigher than the control group. Using the doubletorsion method, a KIc of a high impact denture basematerial (Lucitone) of 3.00 M N mK3/2 has beenreported [20], though this was carried out underwater. This is higher than the results obtained inthis study, but rubber toughened acrylics have anincreased KIc and impact strength at the expense ofthe modulus, which was measured at 2280 M N mK2

as opposed to that of Perspexe of 2640 M N mK2.Using the compact tension method of determiningfracture toughness, KIc values for Trevalon of1.28 M N mK3/2 and Trevalon Hi, a high impactrubber toughened denture base polymer of2.90 M N mK3/2 have been reported [21].

Porosity was a problem when making samplesusing higher glass flake loadings. The use of thelowest percentage loadings gave a higher fracturetoughness value compared to the higher loadingscarried out, with the added value of being easier tomix and less chance of porosity. It’s acceptance inthe dental laboratory would thus be high.Reinforcement of denture base materials by glassflake is an attractive option as by and large thematrix to be reinforced is standard denture baseresin which can be processed in the usual way,requiring no new equipment outlay.

Pre-soaking polyethylene fibres allowed Ladi-zesky et al. [24] to add significantly higherpercentages of fibres than had been possibleearlier. In this a preliminary study, we found thatpre-soaking the flakes showed no effect on the finalmaterial properties and made the mixture harder tomix, as the flakes settled quickly, and the mixtureneeded constant stirring to ensure a homogenousmix. It also seems doubtful that higher loadingsthan those investigated would be required.

P. Franklin et al.370

Fibre reinforced composites have the advantagethat if the matrix should fail catastrophically, thenthe fractured portions are likely to remain in closeproximity, held together by the fibres [25].Reinforcement with particles is a different conceptfrom that of fibre reinforcement, in that thematerial is not affected by the orientation of thefiller. Fibres are most effective when used as matsor sheets. The incorporation of such devices intodenture bases is time consuming and demanding.Such fibres can only be placed where the denturebase material is thickest. Repair of fibre-reinforceddentures is complicated by fibres protruding fromthe fractured surfaces. These problems are over-come by using particulate fillers. Rubber reinforce-ment has remained the most effective method oftoughening PMMA for several decades, its’ greatestproblem being significant additional cost.

The glass flake material has been able tosignificantly improve the mechanical propertiesthat were tested. The flake can simply be added tothe polymer powder and mixed in prior to adding themonomer. The dentures can then be constructed inexactly the same way that they would be normally.

Conclusions

Current methods of reinforcement are expensive orhave too many drawbacks to encourage theirroutine use. For the particular flake size examinedin this study, the addition of glass flake to PMMAdenture base resin considerably improved thefracture toughness of the material for all additions;fracture toughness for the lowest loading level wasstatistically significantly higher than for the othergroups. These improvements were achieved with-out the use of silane coupling which offers thepotential to further improve properties. Furtherwork to look at the effect of glass flake additions onother properties of reinforced denture basematerials is required.

Acknowledgements

Jane Lockwood and Ian Smith in the Division ofRestorative Dentistry for their assistance and Glassflake Ltd for providing the glass flake.

References

[1] Knowles JE. Inhalation of dental plates—a hazard ofradiolucent materials. J Laryngol Otol 1991;105(8):681–2.

[2] Hargreaves AS. The prevalence of fractured dentures: asurvey. Br Dent J 1969;126:451–5.

[3] Rees JS, Huggett R, Harrison A. Finite element analysis ofthe stress-concentrating effect of fraenal notches incomplete dentures. Int J Prosthodont 1990;3:238–40.

[4] Darbar UR, Huggett R, Harrison A. Denture fracture—asurvey. Br Dent J 1994;176:342–5.

[5] Stafford GD, Smith DC. Polycarbonates—a preliminaryreport on the use of polycarbonates as a denture basematerial. Dent Pract 1967;17(6):217–23.

[6] Hargreaves AS. Nylon as a denture-base material. DentPract 1971;22:122–8.

[7] Ping Chaing BK. Polymers in the service of prostheticdentistry. J Dent 1984;12(3):203–14.

[8] Rodford R. The development of high impact strengthdenture-base materials. J Dent 1986;14:214–7.

[9] Rodford R. Further development and evaluation of highimpact strength denture base materials. J Dent 1990;18:151–7.

[10] Vallittu PK. A review of fibre-reinforced denture basedresins. J Prosthodont 1996;5:270–6.

[11] Jagger DC, Harrison A, Jandt KD. The reinforcement ofdentures. J Oral Rehabil 1999;26:185–94.

[12] Hill RG. The role of microstructure on the fracturetoughness and fracture behaviour of rubber-reinforcedacrylics. J Mater Sci 1994;29:3062–70.

[13] Murakami A, Behiri JC, Bonfield W. Rubber-modified bonecement. J Mater Sci 1988;23:2029–36.

[14] Carlos NB, Harrison A. The effect of untreated UHMWPEbeads on some properties of acrylic denture base material.J Dent 1997;25(1):59–64.

[15] Harrison A, Costantinidis VI, Vowles R. The effect of surfacetreated UHMWPE beads on some properties of acrylic resindenture base material. Eur J Prosthodont Rest Dent 1997;5(1):39–42.

[16] Murphy WM, Huggett R, Handley RW. A laboratory andclinical study of Trevalonw denture base material. J OralRehabil 1982;9:401–11.

[17] Hill RG, Bates JF, Lewis TT, Rees N. Fracture toughness ofacrylic denture base. Biomaterials 1983;4:112–20.

[18] Stafford GD, Huggett R, Causton BE. Fracture toughnessof denture base acrylics. J Biomed Mater Res 1980;14:359–71.

[19] Fujishima A, Ferracane JL. Comparison of four modes offracture toughness testing for dental composites. DentMater 1996;12:38–43.

[20] Hill RG, Bates JF, Lewis TT, Rees N. The fracture of acrylicpolymers in water. J Mater Sci 1984;19:1904–16.

[21] Stafford GD, Huggett R, MacGregor AR, Graham J. The use ofnylon as a denture-base material. J Dent 1986;13(1):18–22.

[22] Ladizesky NH, Chow TW, Cheng YY. Denture base reinforce-ment using woven polyethylene fiber. Int J Prosthodont1994;7:307–14.

[23] Berrong JM, Weed RM, Young JM. Fracture resistance ofKevlar-reinforced poly(methyl methacrylate) resin: a pre-liminary study. Int J Prosthodont 1990;3:391–5.