One of the most intensely pursued objectives in dental materials research over the past three decades is the achievement of clinically acceptable retention, by micromechanical and/or chemical bonding mechanisms, of the restorative to the prepared enamel and dentinal tooth structure. Perfect retention, in addition to providing a major contribution to the longevity of the restoration, would offer the best protection against microleakage of oral fluids along the tooth-restorative interface, with its detrimental consequences of bacterial ingress and secondary caries development. Optimally effective interfacial bonding requires complete wetting of the adherend surfaces by the adhesive and the attainment of durable bond strengths matching the inherent strength levels of the dental and restorative components of the joint. Although materials science is still a long way from reaching such perfection, much has been accomplished in recent years in pursuit of this goal. In view of the importance of dentinal and enamel bonding in restorative practice, the subject is being treated in this section under its own separate heading. Also covered here briefly are bonding methods used for prosthodontic and orthodontic attachments and repair.
The retention of restoratives and restorations to the tooth structure is customarily measured in terms of shear bond strength and, less commonly, tensile bond strength. Peel strength measurements, as routinely performed in other segments of adhesion technology, are not particularly predictive here and hence are seldom utilized in restorative dentistry. The bond strength data reported in the dental materials literature tend to show considerable variability because of marked sensitivity to the materials and techniques employed. Type, age, and preconditioning of the tooth material, type and geometry of the prepared cavity (or other adhesion surface), and the application variables of primer and filling material all are of critical importance, and so are the details of postconditioning (e. g., storage in saline and thermocycling) of the prepared joints, and the techniques and devices used for bond strength testing.[38] The strength data given in the text should thus be accepted at best as representative, useful indicators of general bonding performance. It is equally important to realize that the data reported in the literature have been derived almost entirely from in vitro tests and thus cannot simply be correlated with in vivo results, although their value as predictors of clinical performance remains undisputed.
The composite materials presently on the market do not per se possess adhesive properties conducive to bonding to the hard tissue of tooth structure. Auxiliary techniques are available, however, which enable the clinician to overcome this inherent deficiency, and composite-type restorations are routinely placed nowadays under conditions leading to an acceptable, if not perfect degree of bond formation with the cavosurface. Thanks to these advances in dental material technology, cavity preparation with large undercuts, as with amalgam fillings, is no longer a necessity for successful restoration, and the beneficial consequences in terms of preservation of healthy tooth structure and minimization of secondary caries through reduction of microleakage are obvious. Because of differences in some of the bonding mechanisms between the resin-enamel and resin-dentin adherend pairs, the techniques required for resin bonding to enamel on the one hand, and to the dentinal tooth component on the other, differ in certain aspects. Enamel is a biomaterial of low free surface energy and thus will resist wetting by a potential adhesive. Moreover, as pointed out before, it consists of 97 wt % mineral constituent, essentially hydroxyapatite. Any adhesion process would therefore have to rely almost exclusively on reactions with the exposed apatitic hydroxyl groups, as has been established for the polycarboxylate and ionomer cements (Sections II. B.5 and II. B.6). Reactive partners of this type, however, are absent in the resin-based materials. For a mechanical joint, on the other hand, the cut enamel surface, having grooves substantially shallower than 100 nm, lacks the roughness required for retention of the intruding resin tags. The advent of the acid etch technique, developed by Buonocore in 1955, changed the situation dramatically. Acid etching, in essence an enamel-conditioning process, and by now a standard clinical procedure, involves a brief treatment of the clinically prepared enamel surface with acids, most commonly phosphoric acid, applied as an aqueous (30 to 50%) solution or, more conveniently, as an aqueous gel. The resultant increase in free surface energy enhances the wetting characteristics and so enlarges the interfacial contact area. In addition, the etching creates microporosity, which allows the subsequently placed resin to flow into the pores, forming resin tags with a typical length of 25 pm, thus efficaciously anchoring the composite to the enamel in a micromechanical fashion. The depth of hard-tissue penetration is not necessarily, however, the prime contributor to the bonding effect; tag density and inherent strength both are of at least equal importance. The placement of heavily filled and viscous composites, including the hybrid types, which may find it difficult to penetrate into the pores, is frequently preceded by application of a layer of unfilled resin of low viscosity compatible with the composite, although the success of this method is questioned by others. Typical tensile bond strengths attained between composite resin and acid-etched enamel range from 16 to 23 MPa, highest bond strength values generally being associated with surfaces cut transversely to the enamel crystallites [35]. The topic of acid etching has been reviewed by Gwinnett [36] and by Retief [37]. In addition to the acid etching technique, methods of enamel etching by laser treatment have more recently been introduced and in general appear to be similarly effective, or even superior, although more cumbersome in clinical practice.
The development of chemical coupling or bonding agents for resin adhesion to hard tooth structure, pioneered by Bowen several decades ago [38] and more recently reviewed by that author [39], represents a challenging chapter in contemporary dental materials research. Although applicable to resin-enamel bonding, the chemical adhesive materials currently available find their major use in resin-dentin bonding applications.
Contrasted with enamel, dentin contains only 69% hydroxyapatite matter in addition to an increased percentage of organic substance of low surface energy and aqueous fluids, which occupy the dental tubules (Table 1). On a volume basis, the overall organic — aqueous domain makes up more than one-half of the dentinal substance. The dentin surface is thus a strongly hydrophilic adherend. The bis-GMA and related resin components of the composite, on the other hand, represent hydrophobic constituents. A bonding agent intended to join dentinal and composite adherends durably must therefore be hydrophilic enough to displace the aqueous phase from the dentinal surface for subsequent bonding, by whatever mechanism, to the dentinal substrate. At the same time, however, it must comprise hydrophobic molecular entities compatible with, and capable of bonding to, the resinous restorative. Based on this rationale, early biphasic, surface-active dentin bonding agents, developed in Bowen’s laboratory [38], were of the type N-[2-hydroxy-3- (methacryloyloxy)propyl]-N-phenylglycine (NPG-GMA), N-[2-hydroxy-3-(methacryloy — loxy)propyl]-N-(4-tolyl)glycine (NTG-GMA), and related structures. These compounds are distinguished (1) by the presence of hydrophilic functional amino acid groups capable of chelating or ionic bonding to the apatitic surface calcium and other multivalent cations and to reactive amino groups in the organic (collagen) domains of dentin, and (2) by the presence of reactive vinyl groups capable of copolymerization with composite resin. Other first-generation bonding agents contained isocyanatoacrylates or diisocyanate-terminated oligourethanes designed so as to form cross-links between dentinal hydroxyl and amine functions and filler hydroxyl groups. Halogenated phosphate esters of bis-GMA, HEMA, and other methacrylate substrates, believed to function through calcium phosphate bonding to dentin and vinyl-type copolymerization with composite resin, were also developed at that time. The compounds were applied as thin layers to variously conditioned dentinal surfaces, followed by the placement of standard composites. Athough initial results were by no means impressive, shear bond strengths at the very best attaining 10 MPa, these early pioneering investigations provided a powerful impetus to dental bonding research activities worldwide, and although many a development product fell by the wayside for reasons of poor long-term clinical performance, others were developed in the following years to a fairly high level of effectiveness and produced encouraging (although not necessarily clinically acceptable) results. Among the bonding systems that have reached the third-generation stage and compete for present-day clinical acceptance are those based on combinations of (1) glutaraldehyde with HEMA; (2) arylglycine-type surface-active monomers with PMDM, the adduct of HEMA to pyromellitic dianhydride; (3) hydrophilic HEMA with hydrophobic bis-GMA; and (4) methyl methacrylate with 4-META, the adduct of HEMA to trimellitic acid anhydride. A brief discussion of these examplify bonding systems follows.
The original glutaraldehyde-HEMA system, developed in Asmussen’s laboratory [40] and commonly known as GLUMA, contains as the critical component a primer consisting of an aqueous solution of glutaraldehyde (5%) and HEMA (35%), which was applied onto the dental surface precleansed with alkali-neutralized (pH 7.4) ethyle — nediaminetetraacetic acid (17% in water) for smear layer removal and superficial decalcification. This was overlaid with a sealer consisting of unfilled, light-cured resin of the bis-GMA type, onto which in turn the composite was placed. The primer mixture in this system interpenetrates and forms bonds with the top zone of the partly demineralized dentin matrix, to which it anchors the resinous overlays upon free-radical homo — and copolymerization. The bonding effects achieved with this early system were unsatisfactory; average shear bond strengths generally failed to exceed 10 MPa even after the implementation of further (minor) improvements. Bond failure occurred along the weakened decalcified dentin zone, as neither the primer nor the sealer diffused through that zone into the underlying calcified matrix. Adhesive failure at the sealer-composite interface was also observed [41]. Subsequent improvements and simplifications of the GLUMA system included changes in pre-treatment and conversion of the primer into a self-contained bonding resin through inclusion of bis-GMA monomer and initiator. A typical present — day GLUMA bonding procedure [42] comprises the following steps:
1. Cavosurface cleansing by treatment with an aqueous solution of aluminum oxalate (ca. 5%) and glycine (2.5%) adjusted to pH 1.5. This results in both enamel and dentin etching and in amino acid infiltration into the etched dentin.
2. Brush application of bonding resin consisting of glutaraldehyde (5%), HEMA (33%), bis-GMA (2%), camphorquinone photoinitiator (0.1%), water (55%), and acetone (5%), followed by light curing.
3. Conventional placement of composite resin.
In this and similar systems (e. g., with pyruvic acid and glycine as cleanser components) [43,44] the amino acid infiltrated into the dentinal surface zone adds to the concentration of amino groups in that layer and thus contributes to glutaraldehyde bonding; in addition, it is believed to act as the reductant in conjunction with the camphorquinone photooxidant component in the interpenetrating resin, thus upon photoirradiation, initiating resin polymerization right along the contact surface with the cleanser. Shear bond strength values as high as 16 to 18 MPa to dentin, and up to 23 MPa to enamel, can be attained with this and similar third-generation GLUMA recipes.
In the field of bonding agents based on arylglycine-PMDM combinations, numerous advanced versions have originated from Bowen’s early concept of biphasic monomers with both hydrophilic and hydrophobic functional sites as exemplified by the aforementioned NPG-GMA system. In our initial version, a second biphasic monomer, 2,5-bis[2-(metha — cryloyloxy)ethoxycarbonyl]terephthalic acid (PMDM), an addition product of HEMA to pyromellitic dianhydride, was added. The dentinal surface was first conditioned with an aqueous acidic solution of iron(III) oxalate, which removed the smear layer and deposited iron cations, contributing to the bonding effect through chelation. Next, an acetone solution of NPG-GMA or NTG-GMA was applied, followed by treatment with an acetone solution of PMDM and placement of the composite. The PMDM comonomer interacted synergistically with the precursor component, spontaneously inducing free-radical polymerization. Having passed through various stages of improvement, a current version, available commercially, comprises dentin conditioning with aluminum oxalate (6%) in dilute (2.5%) aqueous nitric acid, followed by application of a premixed acetone solution of NTG-GMA and PMDM. After solvent volatilization, this is overlaid with an unfilled, light-curing bis-GMA resin of low viscosity, to be followed by composite placement [39]. The micromechanical processes constituting the overall bonding effect have been studied by transmission and scanning electron microscopy[39] techniques [41,46]. Mean shear bond strengths of 17 to 18 MPa have been reported [47,48]; however, lower and quite variable values are also on record, once again stressing the need for standardization of bonding and testing techniques [49].
The recent finding in Bowen’s laboratory that the oxalate conditioning and subsequent NPG-GMA coating steps can be replaced by a treatment with acidic NPG without loss of bonding strength has led to a related bonding system, also available commercially, in which the dentin is pretreated with a dilute (2.5%) aqueous nitric acid containing NPG (4%) [39]. This removes the smear layer, partially decalcifies the upper dentin layer, and permits interpenetration of the amino acid. Subsequent application of a 5% acetone solution of PMDM, with or without added HEMA, provides an overlay of resin, which penetrates into, and through, the decalcified zone and polymerizes spontaneously in contact with the amino acid, forming a resin-reinforced demineralized zone, which then bonds to the subsequently placed composite [46]. Tensile bond strengths are 12 to 16 MPa at best, and frequently much lower. On the other hand, and in contrast to the behavior of most other contemporary bonding agents, strength tends to increase slightly upon saline storage and thermocycling [50]. Failure typically occurs along the adhesive-tooth surface, and the adhesive resin itself is probably the weakest part of the joint.
Outstanding adhesion perfomance has recently been documented for a modified system in which the key ingredient is a combination of NTG-GMA and BPDM, a biphe- nyldimethacrylate derivative related to PMDM. The two components (called primers), dissolved in acetone, are premixed just prior to multiple-brush application onto the dentinal surface preconditioned either by etching with 10% aqueous phosphoric acid or by treatment with a succinic anhydride-modified HEMA (SA-HEMA) (a hydrophilic/hydro — phobic methacrylate possessing a propanoic acid terminal). The low-viscosity primer mixture displaces surface moisture on the dentin and interpenetrates the partly demineralized collagen layer exposed by the etching process and fills the dentinal tubule orifices. Subsequent application of an unfilled, photocuring methacrylate bonding resin causes further resin reinforcement of the demineralized zone and subsequent copolymerization. This is followed by conventional composite application. Mean shear bond strengths range from about 27 to nearly 40 MPa, depending on details of the application technique, and failure is cohesive in dentin. The phosphoric acid-etching pre-treatment and tolerance of a certain degree of surface moisture (by blotting or mild air drying) both combine to result in optimal bonding, whereas aggressively air-dried surfaces give considerably weaker bonds [51]. The system described also lends itself exceedingly well to metal and porcelain bonding and has therefore found application in luting operations and prosthodontics [51,52]. For example, a Ni-Cr-Be base metal alloy is bonded to composite with a mean shear bond strength in the vicinity of 25 MPa. Key aspects of the NTG-GMA-BPDM primer application have recently been discussed in some detail [52].
The development of HEMA-bis-GMA combinations as bonding agents has culminated in a number of recipes showing encouraging performance, and one major representative now on the market, defined as a dentin-enamel bonding system, has received wide attention. In a typical protocol, the enamel portions of the prepared cavity are conventionally acid etched, and the dentinal surfaces are primed with an aqueous solution of the hydrophilic HEMA and maleic acid as comonomers. This removes the smear layer and provides dentin interpenetration by the two monomers. Priming is followed by brush application, in a fairly thick layer (75 to 100 pm), of a resin adhesive composed of HEMA, bis-GMA, and a photoinitiator, with a few percent of a low-viscosity monomer added for viscosity reduction. After brief light curing of the adhesive coat, the composite is placed conventionally. Because of polymerization inhibition by oxygen, a reactive surface layer containing incompletely polymerized resin is left on the adhesive coat, and subsequent copolymerization with the composite resin overlay affords effective adhesive-composite bonding. Although earlier strength data reported were not particularly convincing, recent publications [41] cite mean shear bond strength values as high as 23 MPa, well on a par with enamel bonding data, with fracture for the major part cohesive in dentin or composite. Excellent performance with respect to minimal microleakage and marginal gap dimensions relative to competitive bonding systems tested are also on record [53]. On the other hand, this bonding system has been found to weaken on storage and thermocycling [41,50].
A combination of modified features of the last-named two bonding systems is realized in an adhesive application known as the Kanca technique, in which dentin and enamel pre-treatment by phosphoric acid etching is followed by the consecutive layering of NTG-GMA, PMDM, and HEMA-bis-GMA adhesive resins, onto which the restorative is placed by conventional manipulation. Low microleakage, and composite shear bond strengths to enamel/dentin at the 18-MPa level, have been reported [54].
The last bonding system to be dealt with in this section, presented in Section II. B.8 as a luting agent, contains as the key monomer the addition product of HEMA to trimellitic acid anhydride, 4-(2-methacryloyloxyethoxycarbonyl)phthalic anhydride (4-META). Following early reports of excellent dentin-composite bonding results with 4-META — containing adhesives (tensile bond strengths typically 17 to 18 MPa), preeminently from Nakabayashi’s group and reviewed by that researcher [55], the 4-META system has since been refined to the stage of commercialization and routine clinical use [10,56]. It typically comprises the following steps:
1. Short (10 to 30 s) pre-treatment of prepared dentinal surface with the familiar citric acid-iron(III) chloride system (10% and 3%, respectively, in water).
2. Application of bonding resin, composed of 5% 4-META in MMA and premixed with the initiator, a partially oxidized tri-n-butylborane [57].
3. Overlaying of bonding resin coat with a thin layer of powdered poly(methyl methacrylate), followed by placement of composite.
The acidic iron(III) chloride etchant, as pointed out before, removes the smear layer and acts as a decalcifying agent. In addition, just like 4-META itself, it appears to promote acrylate monomer penetration into the etched and partly demineralized dentinal surface. The interpenetrated bonding agent containing the hydrophilic-hydrophobic 4-META comonomer may be retained inside the demineralized zone by adsorption onto the hydrophilic and hydrophobic domains present in that zone so that, upon polymerization, a hybrid zone is generated, which consists of resin-reinforced dentinal matter capable of copolymerization with the adjacent overlay of composite restorative. Restricting the duration of the etching treatment to the short period indicated is a vital prerequisite for strong dentin-composite bond formation, as this will keep the depth of demineralization to less than 5pm (ca. 2 pm in noncarious dentin) and maintain the collagen phase in a reactive (nondenatured) state, thus ensuring complete penetration of the demineralized stratum by the MMA/4-META agent down to the virgin (calcified) dentin matrix before polymerization sets in under the influence of the borane initiator. This, in turn, will ensure that no interlayer of decalcified and weakened dentinal material is left between virgin dentin and resin-impregnated stratum, as the exposed collagen, unprotected by infiltrated resin, is susceptible to degradation in an aqueous environment and thus would represent a weak link of the joint [56,58]. An outstanding advantage of the borane derivative as the initiator of this 4-META bonding system rests on its activation by water and oxygen as described by Nakabayashi et al. [58]. The moisture on the dental surfaces in combination with air triggers free-radical generation and thus the initiation of polymerization by the borane at the dentin interface rather than throughout the bulk of the resin layer as in other free-radical-initiated systems. This ensures that resin shrinkage proceeds toward the dentin adherend rather than away from it and so provides forceful counteraction against microleakage. In a further (commercialized) version, etching with citric acid-iron(III) chloride [containing poly(vinyl alcohol) for viscosity control] is followed by brush application of HEMA monomer (containing hydroquinone monomethyl ether), a subsequent application of the HEMA-4-META combination premixed with the tributylborane initiator, and the final placement of the restorative resin [59]. Excellent shear bond strength data, up to nearly 23 MPa, paired with a remarkably low degree of microleakage, have variously been reported [10,41,59,60], and fracture is cohesive in dentine and/or composite. The last-named adhesive system is also quite efficacious in prostho — dontic and orthodontic bonding applications [61] and in the bonding of amalgam fillings, which in general practice, plugging into an undercut cavity, are retained solely by a micromechanical mode. Although dentin-amalgam shear bond strengths, just above 3 MPa, are weak in relation to corresponding dentin-composite strength data, the bond is effective in reducing microleakage appreciably in comparison to conventionally placed amalgam restorative.
Representative shear bond strength ranges for the bonding agents discussed in the foregoing are listed in Table 5, and the structural representations and universally used abbreviations for the principal methacrylate and dimethacrylate monomers are found in Tables 6 and 7. Detailed characterization techniques for methacrylates and derived polymers have been described by Ruyter and 0ysaed [62].