The ofcial journal of the Australian Dental Association

Australian Dental Journal

Australian Dental Journal 2011; 56:(1 Suppl): 310 doi: 10.1111/j.1834-7819.2010.01291.x

Smart materials in dentistry

JF McCabe,* Z Yan,* OT Al Naimi,* G Mahmoud,* SL Rolland*
*School of Dental Sciences, Newcastle University, United Kingdom.

ABSTRACT
Most dental materials are designed to have a relatively neutral existence in the mouth. It is considered that if they are passive and do not react with the oral environment they will be more stable and have a greater durability. At the same time, it is hoped that our materials will be well accepted and will cause neither harm nor injury. This is an entirely negative approach to material tolerance and biocompatibility and hides the possibility that some positive gains can be achieved by using materials which behave in a more dynamic fashion in the environment in which they are placed. An example of materials which have potential for dynamic behaviour exists with structures which are partly water-based or have phases or zones with signicant water content and for which the water within the material can react to changes in the ambient conditions. Such materials may even be said to have the potential for smart behaviour, i.e. they can react to changes in the environment to bring about advantageous changes in properties, either within the material itself or in the material-tooth complex. The controlled movement of water or aqueous media through the material may cause changes in dimensions, may be the carrier for various dissolved species, and may inuence the potential for the formation of biolms at the surface. Some of these issues may be closely interrelated. Clearly, materials which do not have the capacity for water transport or storage do not have the potential for this sort of behaviour. Some materials which are normally resistant to the healthy oral environment can undergo controlled degradation at low pH in order to release ions which may prove benecial or protective. It is doubtful whether such behaviour should be classied as smart because the material cannot readily return to its original condition when the stimulus is removed. Other materials, such as certain alloys, having no means of transporting water through their structure, can display smart behaviour by undergoing predictable changes in structure in response to applied mechanical or thermal stimuli. It has been difcult to harness such behaviour to the benet of patients but progress in this area is slowly being made.
Keywords: Smart materials, fluoride release, fluoride recharging, glass-ionomers, biofilms. Abbreviations and acronyms: GIC = glass-ionomer cement; HEMA = hydroxyethylmethacrylate; PAA = polyacrylicacid; PAN = polyacrylonitrile; PVA = polyvinylalcohol; QLF = quantitative light-induced fluorescence; RMGIC = resin modified glass-ionomer cement; SMA = shape memory alloys.

INTRODUCTION Traditionally, materials designed for long-term use in the body or more specically in the mouth are thought to survive longer if they are passive and have no interaction with their environment. Materials such as amalgams, composites and cements are often judged on their ability to survive without interacting with the oral environment. Perhaps the rst inclination that an active rather than passive material could be attractive was the realization of the benet of uoride release from materials. The nature of smart materials By denition and general agreement, smart materials are materials that have properties which may be altered in a controlled fashion by stimuli, such as stress, tempera 2011 Australian Dental Association

ture, moisture, pH, electric or magnetic elds. A key feature of smart behaviour includes an ability to return to the original state after the stimulus has been removed.1 Existing smart materials include piezoelectric materials which produce a voltage when stress is applied or vice versa.2 Structures made from these products can be made to change shape or dimensions when a voltage is applied. Likewise, a change in shape can be used to generate a voltage which can be used for the purpose of monitoring. Thermo-responsive materials, such as shape memory alloys3 or shape memory polymers4 adopt different shapes at different temperatures due to remarkable and controlled changes in structure. Magnetic shape memory alloys can change their shape in response to a change in magnetic eld. pH-sensitive polymers are materials which swell collapse when the pH of the surrounding media changes.5 Other materials change colour in response to changes in
3

JF McCabe et al. pH, light or applied voltage. One common example of this sort of technology is the light-sensitive sunglasses which darken when exposed to bright sunlight. Polymer gels offer a potential for smart behaviour.1,6 They consist of cross-linked polymer networks which may be inated with a solvent such as water. The labile nature of the solvent enables a rapid and reversible swelling or shrinkage in response to a small change in their environment (e.g. temperature). The most common gel forming polymers are polyvinylalcohol (PVA), polyacrylicacid (PAA) and polyacrylonitrile (PAN). Microsized gel bres may contract in milliseconds, while thick polymer layers may require much longer to react. It has been suggested that these gels can potentially deliver a stress equivalent to that of a human muscle of about equivalent size. Smart materials by chance or design The future use of smart materials will involve a degree of smart behaviour by design. However, smart behaviour was rst noted in some materials by chance and the signicance of the special nature may not be recognized as being of any practical use until some time later. This was certainly the case for thermo-responsive materials, either shape memory alloys or shape memory polymers. Shape memory alloys3 based upon NiTi alloys have been used in orthodontics for many years and their remarkable properties have been commented upon without any insight into how the properties could be harnessed for a practical purpose. Likewise the potential thermo-responsive smart behaviour of some glass-ionomer cements was rst suggested by Davidson7 and was then demonstrated as a result of attempting to measure the coefcient of thermal expansion.811 Heating or cooling of these materials may result in minimal dimensional change as the expected expansions (heating) or contractions (cooling) appear to be offset by a compensating reaction related to the movement of water in or out of the structure. Smart alloys the rst smart dental materials The term smart material or smart behaviour in the discipline which is now loosely dened as dental materials science, was probably rst used in connection with nickel-titanium alloys, or shape memory alloys (SMAs), which are used as orthodontic wires. Frustratingly, although these materials were found to have fascinating characteristics in relation to the way their structure properties changed in response to strain and temperature, it has never been clearly expounded how their characteristics could be used benecially during patient treatment. The smart behaviour is essentially related to the ability of the alloy to initially undergo strain in response to stress in the normal way, but at the point identied as the yield stress there is a further
4

increase in strain which in normal alloys would be identied as irreversible yielding.12,13 In the SMA alloys, however, this yielding is related to a reversible change in the crystal structure. The reverse process is temperature dependent and this lies at the heart of the difculty in harnessing the characteristic for clinical benet as the temperatures required can be very high. However, this knowledge has been used in the manufacture of wires of similar cross-section but markedly different stiffness. The phase changes involved in the crystal transitions involve a small exothermic endothermic response which can be used to monitor or measure the extent of the change. However, the amount of energy involved is very small and very sensitive recording equipment is required (e.g. temperature modulated differential scanning calorimetry).14 In order to fully utilize the super-elastic or shape memory characteristics, it would be essential to enable the phase transitions to occur in a controlled fashion at or near mouth temperature, and this is now claimed to be possible by some manufacturers. Wires exhibiting shape-memory behaviour at mouth temperature normally contain copper and or chromium in addition to nickel and titanium. The role of water Many types of smart behaviour are related to the ability of a gel structure to absorb or release solvent rapidly in response to a stimulus such as temperature. In the oral environment, the key solvent is water and the structures may be gels or salts which contain water which may be bound either strongly or loosely and therefore may be absorbed or released at different rates. Some types of smart behaviour may also be dened by any species, such as uoride ions dissolved in the water and which are capable of undergoing reversible interactions with the gel, salt or oral structures. Depending upon the nature of the water and how strongly it is bound, the observed changes may be dependent upon the dimensions of the structures.11 Smart thermal behaviour The vast majority of materials responds to a temperature change in a predictable manner. This involves a dimensional change characterized by the coefcient of thermal expansion or expansivity. One problem with dental lling materials is their tendency to expand and contract to a greater extent than the natural tooth tissue when subjected to hot or cold stimuli. When samples of restorative materials were heated in order to determine their values of coefcient of thermal expansion, an interesting observation was made. For composite materials, expansion and contraction occurred in the expected way and a coefcient could readily be
2011 Australian Dental Association

Smart materials in dentistry determined, and whether testing was done dry or wet made little or no difference.8 For glass-ionomers, little or no change in dimension was observed when heating and cooling between 20 C and 50 C in wet conditions (Fig 1).9,10 In dry conditions, the materials showed a marked contraction when heated above 50 C. The explanation for this behaviour is that the expected expansion on heating is compensated by uid ow to the surface of the material to cause a balancing of the dimensional changes. On cooling, the process was reversed. In dry conditions, the rapid loss of water on heating results in the observed contraction. This behaviour is akin to that of human dentine (Fig 2) where very little dimensional change is observed on heating in wet conditions and a marked contraction is noted in dry conditions.11 Both results can be explained by ow of uids in the dentinal tubules. Hence, the glass-ionomer materials can be said to be mimicking the behaviour of human dentine through a type of smart behaviour. The role of porosity The smart behaviour of glass-ionomers and related materials is closely linked to their water content and the way in which this can react to changes in the environment. Clearly, there are different ways in which water can be retained in and transported through the cement structure. One important feature which may provide a location for the formation of reservoirs within the material is porosity. The number and size of pores within a cement can be controlled by the method of mixing and is conveniently measured using micro-CT scanning.15 Figure 3 shows typical scans of a glassionomer cement mixed mechanically in capsules or handmixed. The volume of bubbles can readily be calculated and is shown in Table 1. The results given in Table 1 indicate that both the method of mixing and the viscosity of the cement have an effect on porosity. In the low viscosity material, handmixing reduces the porosity signicantly compared to mechanical mixing, either by shaking or rotation. For the viscous material the levels of porosity are low and not signicantly affected by mixing. These differences in porosity are reected in differences in water absorption (Table 2).11,16 Hence, this aspect of the

Fig 1. Dimensional change plotted against temperature for three glass-ionomers (KC, KM, F) and one resin modied glass-ionomer (FLC).

Fig 3. Micro-CT scans of a glass-ionomer cement (Ketac-Cem) mixed mechanically by shaking (CapMix) or rotating (RotoMix). L, M and H refer to low, medium and high examples of porosity within each group.

Table 1. The volume of bubbles in a low viscosity glass-ionomer (Ketac-Cem) and a high viscosity glassionomer (Ketac-Molar) using different methods of mixing
Mixing method Material Ketac-Cem Aplicap Maxicap Hand Ketac-Molar Aplicap Hand CapMix 2.8 (1.6) 2.7 (2.6) 0.2 (0.2) RotoMix 1.8 (1.8) 2.1 (2.1) 0.1 (0.1) Hand 0.2 (0.2) 0.1 (0.1) 0.1 (0.1)
5

Fig 2. Dimensional change of human dentine when heated under wet or dry conditions.
2011 Australian Dental Association

JF McCabe et al. Table 2. Porosity and water absorption of Ketac-Cem glass-ionomer cement
Mixing method CapMix Hand Porosity (%) 2.7 (2.6) 0.2 (0.2) 24 h water absorption (%) 1.32 (0.25) 0.84 (0.11)

smart behaviour of dental cements can be controlled by the operator. Expansion and radial pressure Smart materials which combine a special interactive characteristic with an acceptable durability or longevity are likely to combine some salt or gel characteristic with a resin component imparting some stability. The manufacturing of such materials presents a problem of compatibility. Traditionally, such problems are solved by also incorporating species with both hydrophilic and hydrophobic groups which have the function of bridging or coupling the two distinctly different ingredients. The most commonly used of these compounds is hydroxyethylmethacrylate (HEMA). However, the large and rapid water absorption of polymers containing HEMA can cause problems; not only does the absorption result in swelling but a considerable radial pressure can also be linked to the process.17,18 Proles of cavities containing a blend of GIC and resin phases with HEMA constrained within a cavity show the growth of the material out of the cavity and this is combined with a considerable radial pressure measured as around 26 MPa compared with <3 MPa caused by the water absorption of a typical resin matrix composite.19 Ion release and recharging The benecial effect of uoride release of dental materials has been the subject of much research over many years. There is some doubt about the efcacy of uoride release in caries prevention since even products with high initial uoride release tend to rapidly lose the ability to release uoride in signicant amounts. Even in the case of GICs, the uoride release rate can become negligible within a week.2022 However, the smart behaviour of materials containing GIC salt phases offers some long-term solutions to this problem. There is evidence that the uoride released from salt phases can be replaced when the material is bathed in a high concentration of uoride as may occur in a toothpaste or mouthrinse.16,17,23,24 In the long term, the uoride re-released after recharging may be much more important than the initial burst which is sustained only for a short time. Figure 4 shows data for one material in which uoride release over 24-hour periods into water is seen
6

Fig 4. Twenty-four hour uoride release for a resin matrix system containing some GIC salt matrix zones as a function of time of storage in water. At days 8, 17, 23 and 30 the specimens were subjected to a soak in 250 ppm NaF solution for 1 hour. The two lines are for different methods of measuring uoride concentration.

to decline sharply over just a few days. As the material becomes depleted of its inherent uoride it is interesting that the spikes of uoride in the 24-hour period after recharging appear to increase slightly with age.17 This implies that the more inherent uoride lost, the greater capacity for uptake through recharging. The levels of uoride release maintained can be increased by beginning the recharging process as soon as possible after setting. Other work has shown that the rates of uoride release and recharging are temperature sensitive.16 Hence, a more rapid recharging could be accomplished by using warm uoride containing solutions and this can generate a more sustained release at mouth temperature. Another area where smart uoride interactions can have a signicant benet is in the prevention of demineralization around orthodontic brackets (Fig 5). Using uoride releasing cements can help to prevent demineralization. This has been studied using quantitative light-induced uorescence (QLF) (Fig 6). Results of laboratory studies25 show that the use of uoride-releasing cements can minimize the

Fig 5. White lesions corresponding to areas of demineralized enamel remaining in the areas surrounding the site where orthodontic brackets were bonded.
2011 Australian Dental Association

Smart materials in dentistry appear to be almost equally effective when compared with a material having negligible uoride release, as shown in the QLF images (Fig 7). Another aspect of smart behaviour for this type of material involves the development of materials having mechanical properties adequate for the retention of brackets but which enable easy debonding of brackets at the end of treatment.26 Some resin based materials have a potential to release uoride and neutralize acids.27 Other resin based materials have been designed to release calcium, uoride and hydroxyl ions at low pH28,29 and these materials are reported to have the potential to neutralize acids and to prevent, or even reverse, demineralization. However, clinical performance of such reactive products is often reported to be unsatisfactory. This highlights the potential problem which exists for many types of smart material: their smart behaviour depends upon a degree of reactivity which may not be compatible with longevity. Taking this a stage further, smart composite materials have been developed3035 which contain amorphous calcium phosphate particles or whiskers which at low pH provide a source of calcium and phosphate ions which may act to prevent demineralization of tooth structure and reportedly maintain acceptable mechanical properties in the process.
Fig 6. QLF device used to study demineralization in teeth and the positive effects of uoride release and recharging. Taken from InspektorTM Pro Manual.

Biolms and smart behaviour Biolms formed on the surface of materials in the mouth may enhance the smart behaviour of materials containing uoride releasing salt phases.36,37 Recent work with saliva, using live dead staining techniques,23,36 has shown that uoride release from

demineralization, and that this effect is not simply dependent upon the extent of the initial uoride release. Materials having various levels of uoride release

Fig 7. Pre and post-demineralization QLF images around orthodontic brackets showing signicant demineralization around a non-uoride releasing cement and minimal demineralization around cements with large or moderate uoride release or recharging capacity.
2011 Australian Dental Association 7

JF McCabe et al. at the same time concentrate uoride which is liberated through a change in pH or mechanical debridment. Design of smart materials Now that the ways in which materials containing a polysalt matrix can exhibit smart behaviour have been demonstrated, it is appropriate to consider whether future materials can have smartness designed into them. If so, can the smart behaviour be accommodated without compromising the other key requirements, such as clinical function and longevity? Of the currently available dental materials, the products which most positively react with their environment in a manner which could be interpreted as smart are the glassionomer cements. However, these products are known to have limited durability and longevity due to their brittleness and solubility. Materials demonstrating an optimum combination of smart interactions and longevity are likely to have some combination of stable resin matrix combined with a coexistent salt matrix or discrete gel phase. The rapid developments in nanotechnology suggest that such features can be manufactured into compounds by using building blocks at a molecular or even atomic level. However, in 1996 Friend38 stated, The development of true smart materials at the atomic scale is still some way off, although the enabling technologies are under development. These require novel aspects of nanotechnology (technologies associated with materials and processes at the nanometre scale, (10)9 m) and the newly developing science of shape chemistry. This statement still holds true to an extent today. However, our understanding of the potential benets of smart behaviour have enabled scientists to appreciate the potential benets of active as opposed to passive materials and the development of materials exhibiting smart behaviour is now recognized to be possible outside the realms of nanotechnology with its rather articial and restricting boundaries and denitions. Hence, even with existing technologies we are able to consider building materials with controlled structure and properties. Within the spectrum of materials which lie in the continuous scale between resin matrix composites and salt matrix glass-ionomers,17,39 we are already able to identify various materials described as resin modied GICs (RMGICs), polyacid-modied resin composites (compomers) or glass-ionomer composites (giomers). These have been shown to exhibit some smart characteristics, albeit more through chance than design. The next stage is to harvest the current knowledge into the design of materials with controlled and designed structure in which the requirements of longevity and smart interaction are balanced. For example, when resin matrix and salt matrix setting reactions are
2011 Australian Dental Association

Fig 8. Fluoride release (24-hour values) from a GIC into different storage media. Release into natural saliva is slower than into water or articial saliva.

Fig 9. Fluoride release into natural saliva in acidied, neutral or cycled (A&N or N&A) conditioned. Note the marked increase in uoride release over both the rst and second day of placement into acidic conditions.

materials does not prevent biolm formation or growth. Figure 8 shows that the daily uoride release in natural saliva is signicantly lower than the release into distilled water or articial saliva. However, when samples are stored in acidied (pH 4) saliva the rate of uoride release is markedly increased (Fig 9). More interesting are the results which show that when the specimens are cycled through both acidic and neutral conditions, an increase in uoride release is seen at day 1 and then also at day 2 after placing into acidic conditions.36 This offers some proof that uoride becomes concentrated within the biolm and is made available when the lm is disturbed. The presence of a biolm on the surface of a material alters the interaction of the surface with the environment and in the case of a restorative material, the biolm may act as a lubricant which prevents abrasive wear (Fig 10).36 The formation of biolms and the way in which this changes the interaction of the materials with the environment represents a clear example of smart behaviour for these materials. It seems that biolms can protect surfaces from abrasive forces and
8

Smart materials in dentistry

Fig 10. Surface proles of a GIC at baseline and after brushing in acidied saliva (natural or articial). This illustrates the protective effect of the biolm formed in natural saliva.

competing during the setting of an RMGIC material, it is possible to conceive of means of controlling the extent to which one or other of the processes dominates and hence to inuence the structure and properties of the set material. ACKNOWLEDGEMENT The authors are grateful to Professor Seiji Ban, editor of the Dental Materials Journal, for permission to reproduce large parts of this paper which was originally presented at the International Dental Materials Congress, Bangkok in 2003 and published in Dent Mater J 2009;28:3743. REFERENCES
1. Zrinji M, Szilagyi A, Filipcsei G, Feher J, Szalama J, Moczar G. Smart gel-glass based on the responsive properties of polymer gels. Polym Adv Technol 2001;12:505510. 2. Allameh SM, Akogwu O, Collinson M, Thomas J, Soboyejo WO. Piezoelectric generators for biomedical and dental applications: effect of cyclic loading. J Mater Sci Mater Med 2007;18: 3945. 3. Gil FJ, Planell JA. Shape memory alloys for medical applications. Proc Inst Mech Eng 1998;212:473488. 4. Lendlein A, Langer R. Biodegradeable, elastic, shape-memory polymers for potential biomedical applications. Science 2002;296:16731676. 5. Stayton PS, El-Sayed ME, Murthy N, et al. Smart delivery systems for biomolecular therapeutics. Orthod Craniofac Res 2005;8:219225. 6. Prabaharan M, Mano JF. Stimuli responsive hydrogels based on polysaccharides incorporated with thermo-responsive polymers as novel biomaterials. Macromol Biosci 2006;8:9911008.
2011 Australian Dental Association

7. Davidson CL. Glass ionomer cements, intelligent materials. Bull Group Int Rech Sci Stomatol Odontol 1998;40:3842. 8. Sidhu SK, Carrick TE, McCabe JF. Temperature mediated coefcient of dimensional change of dental tooth-coloured restorative materials. Dent Mater 2004;20:435440. 9. Yan Z, Sidhu SK, Carrick TE, McCabe JF. Response to thermal stimuli of glass ionomer cements. Dent Mater 2007;23:597600. 10. Yan Z, Sidhu SK, McCabe JF. The inuence of microstructure on thermal response of glass ionomers. J Mater Sci Mater Med 2007;18:11631166. 11. Yan Z. Smart materials in dentistry. Newcastle University, United Kingdom, 2006. PhD thesis. 12. Gil FJ, Planell JA. Shape memory alloys for medical applications. Proc Inst Mech Eng H 1998;212:473488. 13. Gil FJ, Planell JA. Effects of copper addition on the superelastic behaviour of Ni-Ti shape memory alloys for orthodontic applications. J Biomed Mater Res 1999;48:682688. 14. Brantley WA, Iijima M, Grentzer TH. Temperature-modulated DSC provides new insight about nickel-titanium wire transformations. Am J Orthod Dentofacial Orthop 2003;124:387394. 15. Nomoto R, Komoriyama M, McCabe JF, Hirano S. Effect of mixing method on the porosity of encapsulated glass ionomer cement. Dent Mater 2004;20:972978. 16. Yan Z, Sidhu SK, Mahmoud GA, Carrick TE, McCabe JF. Effects of temperature on the uoride release and recharging ability of glass ionomers. Oper Dent 2007;32:138143. 17. McCabe JF. A laboratory evaluation of the water and uoride equilibrium characteristics of reactmer giomer. International Meeting Proceedings, Shofu Inc., 2001. 18. McCabe JF, Rusby S. Water absorption, dimensional change and radial pressure in resin matrix dental restorative materials. Biomaterials 2004;25:40014007. 19. Momoi Y, McCabe JF. Hygroscopic expansion of resin based composites during 6 months of water storage. Br Dent J 1994;176:9196. 20. Shaw A, McCabe JF, Carrick TE. Fluoride release from glass ionomers and compomers. J Dent 1998;26:355360.
9

JF McCabe et al.
21. Momoi Y, McCabe JF. Fluoride release from light-activated glass ionomer restoration cements. Dent Mater 1993;9:151 154. 22. McCabe JF, Carrick TE, Sidhu SK. Determining low levels of uoride released from resin based dental materials. Eur J Oral Sci 2002;110:380384. 23. Al Naimi OT, Itota T, Hobson RS, McCabe JF. Fluoride release for restorative materials and its effect on biolm formation in natural saliva. J Mater Sci Mater Med 2008;19:1243 1248. 24. Itota T, Carrick TE, Yoshiyama M, McCabe JF. Fluoride release and recharge in giomer, compomer and resin composite. Dent Mater 2004;20:789795. 25. Mahmoud G, Pretty IA, Gordon PH, McCabe JF. Effect of uoride release on enamel de-mineralisation adjacent to orthodontic brackets. Proceedings BSDR 2007, Abstract 217. 26. Larmour CJ, McCabe JF, Gordon PH. An ex vivo assessment of resin-modied glass ionomer bonding systems in relation to ceramic bracket debond. J Orthod 2000;27:329332. 27. Itota T, Al-Naimi OT, Carrick TE, Yoshiyama M, McCabe JF. Fluoride release and neutralizing effect by resin-based materials. Oper Dent 2005;30:522527. 28. van Dijken JWV. Three-year performance of a calcium-, uoride-, and hydroxyl-ions- releasing resin composite. Acta Odont Scand 2002;60:155159. 29. Heintze SD. A new material concept for inhibiting the formation of secondary caries. Am J Dent 1999;12:S4S7. 30. Skritic D, Antonucci JM, Eanes ED. Amorphous calcium phosphate bioactive composites for mineralized tissue regeneration. J Res Nat Inst Stand Tech 2003;108:167182. 31. Xu HHK, Weir MD, Sun L. Calcium and phosphate ion releasing composite: effect of pH on release and mechanical properties. Dent Mater 2009;25:535542. 32. Xu HHK, Weir MD, Sun L. Nanocomposites with Ca and PO4 release: effects of reinforcement, dicalcium phosphate particle size and silanization. Dent Mater 2009;23:14821491. 33. Antonucci JM, Fowler BO, Weir MD, Skrtic D, Stansbury JW. Effect of ethyl-a-hydroxymethylacrylate on selected properties of copolymers and ACP resin composites. J Mater Sci Mater Med 2008;19:32633271. 34. Namen FM, Galen J, De Deus G, Cabreira RD, Filho FC. Effect of pH on the wettability and uoride release of an ion-releasing resin composite. Oper Dent 2008;33:571578. 35. Xu HHK, Sun L, Weir MD, et al. Nano DCPA-whisker composites with high strength and Ca and PO4 release. J Dent Res 2006;85:722727. 36. Al Naimi OT. Surface interaction of biomaterials with biological uids. Newcastle University, United Kingdom, 2005. PhD thesis. 37. Rolland SL, McCabe JF, Robinson C, Walls AWG. In vitro biolm formation on the surface of resin-based dentine adhesives. Eur J Oral Sci 2006;114:243249. 38. Friend C. Smart materials: the emerging technology. Materials World 1996;4:1618. 39. McCabe JF. Resin-modied 1998;19:521527. glass ionomers. Biomaterials

Address for correspondence: Professor JF McCabe School of Dental Sciences Newcastle University NE2 4BW United Kingdom Email: j.f.mccabe@newcastle.ac.uk

10

2011 Australian Dental Association