New polymer-chemical developments in clinical dental polymer materials: Enamel–dentin adhesives and restorative composites



In the past 10 years, many new components were synthesized and evaluated for an application in enamel–dentin adhesives and direct composite restoratives. New bisacrylamide cross-linkers with improved hydrolytic stability and new strongly acidic polymerizable phosphonic acids and dihydrogen phosphates, as well as novel photoinitator systems, in combination with the implementation of novel application devices, have significantly improved the performance of the current enamel–dentin adhesives. The currently used resins for direct composite restoratives are mainly based on methacrylate chemistry to this day. A continuous improvement of the properties of current composites was achieved with the use of new tailor-made methacrylate cross-linkers, new additives, and photoinitiators as well as tailor-made fillers. Nowadays, dental adhesives and methacrylate-based direct restorative materials have found wide-spread acceptance. Nevertheless, future developments in the field of dental adhesives and direct composite restoratives will focus on improving durability and biocompatibility as well as the development of materials with a broader application spectrum and of smart adhesives or composites. © 2012 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem, 2012


Basically, dental materials can be divided into clinical materials and technical materials. Clinical materials, which are described in a number of useful monographs,1–8 are mainly used by the dentist in dental surgery, whereas technical materials are mostly used by the dental technician to fabricate, for example, dentures. Polymers have been widely used in dental materials for many years. Typical polymer-based technical materials are artificial teeth, denture bases, or denture liners. These materials should be relatively cheap, have to meet the esthetic and toxicological requirements, and must be easy to apply. Therefore, weakly cross-linked poly(methacrylate)s are a suitable choice for these applications. However, expensive high-performance polymers cannot be used because of the existing strong price restriction concerning the technical dental materials. In contrast, the use of more expensive polymers in clinical dental materials such as restorative composites, cements, or adhesives, is generally conceivable. Indeed, these materials are only used in small quantities per application. For example, the application of a restorative composite or cement requires less than 1 g of the material. A few drops in the case of an enamel–dentin adhesive are sufficient to achieve efficient adhesion. It is well known that even gold is frequently used as material for the filling of larger cavities in posterior teeth. Therefore, enamel–dentin adhesives and composite restoratives are interesting fields of application for new components, which enable important properties of these materials to be improved. In 2010, the total value of the enamel–dentin adhesives and restorative composites applied worldwide was about 1.2 billion euros. However, enamel–dentin adhesives and restorative composites have to meet substantial requirements, which include various physico-chemical, processing, clinical, and toxicological properties or demands. Thus, suitable dental monomers should be liquid, colorless compounds, which show, for example, a high rate in free-radical homopolymerization or copolymerization with other monomers, long-term stability against premature polymerization during storage at room temperature or in the refrigerator, excellent resistance to oral environment, high light and discoloration stability of the polymers thereof formed, a low oral toxicity and cytotoxicity, and no mutagenic or carcinogenic potential (Scheme 1). Furthermore, monomers for restorative composites should exhibit a low volume contraction during polymerization, excellent mechanical properties after polymerization, and low water solubility. The water sorption of the formed polymer should also be low. In the case of adhesive monomers, a good solubility in solutions of ethanol, acetone or other nontoxic polar solvents or their mixtures with water, sufficient hydrolytic stability in water-based adhesives, optimal wetting and film-forming properties, and a fast and strong interaction with enamel and dentin are important. All other components included in dental adhesives and composites, such as initiators, stabilizers, fillers, or other additives, have to meet their corresponding requirement.

Scheme 1.

Requirements for dental monomers.

In our previous reviews, overviews of new components for dental composites and adhesives until 2000,9 2004,10 and 200611 were given. This article is focused on significant polymer-chemical developments and trends in enamel–dentin adhesives and dental composite restoratives in the past 5 years. The different chapters start with a brief overview of the corresponding dental material classes, present the state of the art and future trends using new components and procedures, and include our contribution in the development of polymer materials for dental application.


State of the Art of Enamel–Dentin Adhesives

Nowadays, durable esthetic tooth-colored restorative materials, particularly direct resin-based filling composites in combination with efficient enamel–dentin adhesives, play an important role in modern dentistry. In this context, the improvement of the dental adhesive technology has extensively influenced the clinical performance of modern dental restoratives.12 Nearly 60 years ago, Buonocore13 was the first who demonstrated that acid etching of enamel with phosphoric acid led to improved resin enamel bonds. Today, bonding to enamel has been proven to be durable and bonding to dentin has been significantly improved. The currently used enamel–dentin adhesives for restorative composites can be classified according to the clinical approach into etch-and-rinse adhesive (E&RA) and self-etch adhesives (SEA).14,15 The adhesion strategy of E&RAs involves three or at least two steps (Scheme 2). The use of a three-step E&RA first of all requires the application of an acid etchant, commonly a 32–37% phosphoric acid gel (pH 0.1–0.4). A primer is then applied, followed by a separate adhesive resin. In the simplified two-step systems, the second and third steps are combined. SEAs are based on the use of nonrinse acid monomers that simultaneously condition and prime dentin and enamel and involve either a two- or one-step application procedure. Presently, one-step SEAs, also called all-in-one adhesives, combine the conditioning, priming, and the use of an adhesive resin and can be applied in a single step.

Scheme 2.

Classification of current adhesives for restorative composites.

The E&R technique is the most effective approach to achieve an efficient, strong, and stable bond to enamel. The acid-etching results in a selective dissolution of hydroxyapatite (HAP) crystals. The adhesive resin fills the formed retentive etch pattern and polymerizes in situ resulting in the formation of polymerized resin tags.16 The simultaneous etching of enamel and dentin, the so-called total etching of cavities, was a significant simplification of the E&RAs. The acid etching of dentin exposes a microporous network of collagen. During the acid-etching process by E&RA, 50 vol % of the surface and subsurface mineral is solubilized and replaced by rinse water.17 The application of the primer and adhesive resin results in the infiltration of monomers into the collagen network, which replace the water between the collagen fibrils, and a subsequent in situ polymerization leads to the formation of the so-called hybrid layer and resin tags, which seal the unplugged dentin tubules.18 In the priming step, the nature of the solvents and monomers is very important. Primers are usually based on water and 2-hydroxyethyl methacrylate (HEMA) containing solutions that ensure complete expansion of the collagen network and wet the collagen fibrils with hydrophilic monomer.

In contrast to E&RAs, SEAs do not require a separate etching step.19 They contain acidic monomers, which are able to simultaneously condition and prime the dental hard tissues and therefore demineralize and infiltrate enamel and dentin to the same depth simultaneously. SEAs are used as two-step or one-step adhesives, depending on whether a self-etching primer and adhesive resin are separately provided or are combined into one single mixture. Moreover, one-step adhesives can be further subdivided into single-component and two-component SEAs. The nature of the dentin-adhesive interface formed by SEAs depends to a large extent on the structure of the functional monomers, which interact with the dentin. Thus, when decreasing the pH of the self-etch solutions, the interaction depth increases from a few hundreds of nanometers in the case of ultramild SEAs (pH > 2.5), to around 1 µm for mild SEAs (pH ≈ 2,) and several micrometers in the case of strong SEAs (pH ≤ 1). For the description of the interaction of functionalized monomers with HAP-based tissues, the adhesion-decalcification concept (AD-concept, Scheme 3) is very useful.20 According to the AD-concept, the first bonding step (P1) of an acidic monomer (e.g., a phosphonic acid R-PO(OH)2) with HAP results in the release of phosphate and hydroxyl ions, meaning that the surface remains electro-neutral. Depending on the hydrolytic stability of the ionic bond to calcium, two options are possible for the second step (P2): (a) if the ionic bond is stable, then the acid monomer will remain bonded to HAP resulting in a chemical adhesion between the acidic monomer and the HAP of the dental tissues (P2a) or (b) the ionic bond is not stable, resulting in debonding, decalcification, and release of Ca2+ and phosphonate ions from the tooth surface (P2b).

Scheme 3.

AD concept according to Yoshida et al.20

Altogether, E&RAs produce a more efficient and durable resin-dentin bond that is more durable than that of one-step adhesives, which are a mixture of hydrophilic and hydrophobic components. Major shortcomings of one-step adhesives are increased interfacial nanoleakage, limited bond durability, phase separation, or reduced shelf-life. The micromechanical interlocking is the important prerequisite to achieve a good bonding. Nevertheless, additional chemical interaction between the used functional monomers and the tooth substrate (HAP and collagen) may improve the bond durability. In this context, it was shown that insufficient resin impregnation of dentin, high permeability of the bonded interface, a too high concentration of acidic monomer, suboptimal polymerization, phase separation, and activation of endogenous collagenolytic enzymes are factors that reduce the longevity of the bonded interface.21,22

Components of Currently Used Enamel–Dentin Adhesives

The contemporary enamel–dentin-adhesive systems (primer and adhesive resin) are typically a mixture of different monomers, polymerization initiators and inhibitors, solvents, fillers, and further additives (Scheme 4). The monomers are the main component of the enamel–dentin-adhesive systems. They enable the formation of durable covalent bonds between the dental adhesive, the restorative composite, and tooth tissue. Most of the currently used monomers are methacrylates,23 which show a sufficient reactivity in free-radical polymerization. Although, acrylates are more reactive, they may increase the toxicological risk of the monomers. Additionally, acrylates are susceptible to side reactions, such as Michael addition of nucleophilic compounds to the double bond, which leads to a loss of their polymerizability. According to their functionality, the used methacrylates can be divided into nonacidic and acidic functionalized monomethacrylates, cross-linking dimethacrylates, or multifunctional methacrylates. The most frequently used nonacidic functionalized monomethacrylate is HEMA. HEMA is a water-soluble, low-viscosity monomer that improves the miscibility and solubility of the polar and nonpolar adhesive components and the wetting behavior of the liquid adhesive on the dental hard tissues. Moreover, HEMA may stabilize the collagen fibril network and improve the dentinal permeability and monomer diffusion. In addition to HEMA, 2-hydroxypropyl methacrylate (HPMA) is also used in a few enamel–dentin adhesives. Currently used acidic functionalized monomethacrylates are mainly methacryloyloxydecyl dihydrogen phosphate (MDP), 2-methacryloyloxyethyl dihydrogen phosphate (MEP), 2-methacryloyloxypropyl dihydrogen phosphate (MAP), 4-methacryloyloxyethyl trimellitic acid (4-MET) and 11-methacryloyloxy-1,1′-undecanedicarboxylic acid (MAC-10). Furthermore, acidic monomers, such as glycerol dimethacrylate dihydrogen phosphate (GDMP), dipentaerythritolpentaacryloyl dihydrogen phosphate (PENTA-P), ethyl 2-[4-dihydroxyphosphoryl)-2-oxabutyl]acrylate (EAEPA), N-tolylglycine glycidyl methacrylate (NTG-GMA), or 2-acrylamido-2-methyl-1-propansulfonic acid are used (Scheme 5). These acidic monomers have the ability to both etch the dental hard tissues and interact with the tooth structure. They may copolymerize with the other monomers contained in the primer, the adhesive, or the restorative material.

Scheme 4.

Main components in contemporary dental primers and adhesive resins.

Scheme 5.

Examples of acidic monomers used in current dental adhesives.

Cross-linking dimethacrylates are used in enamel–dentin adhesives to form a polymer network, which leads to a number of favorable effects. First, the polymerization rate is significantly increased because of the gel-effect. Second, the mechanical properties of the polymer network are improved compared to linear polymers. Third, the formed cross-linked layer is not water soluble and the degree of swelling decreases with polymer network density. The most popular cross-linking dimethacrylates used in enamel–dentin adhesives are 2,2-bis[4-(2-hydroxy-3-methacryloyloxypropoxy)phenyl]propane (BisGMA), 1,6-bis-[2-methacryloyloxyethoxycarbonylamino]-2,4,4-trimethylhexane (UDMA), triethyleneglycol dimethacrylate (TEGDMA), and glycerol dimethacrylate (GDMA, Scheme 6). The dimethacrylates exhibit different properties, such as viscosity, polarity and water solubility, polymerization shrinkage, film formation behaviour, and reactivity in radical polymerization. Thus, BisGMA shows a high polymerization reactivity, its viscosity, however, is very high and its water solubility low. Conversely, TEGDMA and GDMA are low-viscosity dimethacrylates and show improved solubility in water. Unfortunately, all these dimethacrylates are not hydrolytically stable in aqueous acid solutions and degrade under formation of the corresponding alcohols and methacrylic acid.10 In this context, we synthesized new bisacrylamides (Scheme 7), which show an adequate reactivity in free-radical polymerization as well as improved hydrolytic stability under aqueous acidic conditions compared to the dimethacrylate cross-linkers.24–26 Thus, bisacrylamides such as N, N′-diethyl-1,3-bis(acrylamido)propan (DEBAAP) were introduced as hydrolytically stable cross-linkers in our current enamel–dentin adhesives.27,28 It is noteworthy that DEBAAP exhibits a lower cytotoxicity than the currently used dimethacrylates.

Scheme 6.

Examples of cross-linking dimethacrylates used in current restorative materials.

Scheme 7.

Cross-linking bis(acrylamide)s with improved hydrolytic stability.

Dental adhesives harden by free-radical polymerization. In light-curing systems, mixtures of camphorquinone (CQ) and tertiary amines (e.g., ethyl 4-dimethylaminobenzoate [EMBO]) are almost exclusively used as the photoinitiator.29 The main benefits of CQ are the low toxicity of the initiator, its photodegradation products, and its excellent bleaching behavior. CQ is able to absorb light in the blue region of the visible light spectrum (400–500 nm), with the absorption maximum λmax at 465–475 nm. The first step of the radical formation is the fast electron transfer from the amine to the excited CQ (Scheme 8) followed by a proton transfer under formation of the initiating aminoalkyl radicals, whereas the ketyl radicals tend to dimerize. One crucial problem of acidic enamel–dentin adhesives is the acid–base reaction between the acidic monomers and the amine coinitiator of the photoinitiator system. Consequently, the concentration of the amine and also of the initiating amine radicals decreases. Therefore, alternative photoinitiators, such as acylphosphine oxides (e.g., 2,4,6-trimethylbenzoyldiphenylphosphineoxide [APO]), bisacylphosphine oxides (e.g., bis-(2,4,6-trimethylbenzoyl)phenylphosphine oxide [BAPO]), and aromatic iododonium or sulfonium salts (e.g., (4-isopropylphenyl-4-iodonium tetrakis(pentafluorophenyl)borate) are used (Scheme 9). However, these photoinitiators only strongly absorb in the UV region and have to be combined with CQ for a photocuring with visible light.

Scheme 8.

Radical formation and polymerization of CQ-amine photoinitiator systems.

Scheme 9.

Alternative photoinitiators for dental composites.

Moreover, redox systems are also applied in enamel–dentin adhesives.11 These are intended for applications in which light initiation does not work because light cannot pass in sufficient intensity. The redox systems consist of at least one oxidizing agent, such as a peroxide (e.g., the frequently used dibenzoylperoxide (DBPO)), and a reducing agent. In most instances, tertiary amines are used as reducing agents (e.g., N, N-diethanol-p-toluidine (DEPT) or N, N-diethyl-3,5-di-tert-butylaniline). These reducing agents accelerate the cleavage of DBPO associated with radical formation, so that polymerization can be initiated even at room temperature (cold-curing mechanism). Dual-curing initiator systems, which consist of the combination of photoinitiators (e.g., a CQ/EMBO system) and a DPBO/amine-based redox system are particularly used in luting composites.

Dental primers and adhesives contain other components beside the monomers and initiators (Scheme 4): first, inhibitors, such as 2,6-di-tert-butyl-4-methylphenol (BHT) or hydroquinone monomethyl ether (MEHQ), which are used in an amount of about 100–1000 ppm to stabilize the adhesive formulations against premature polymerization. These phenols are aerobic inhibitors because they are only fully effective in the presence of oxygen. Therefore, it is sometimes useful to combine them with anaerobic inhibitors. Anaerobic inhibitors such as phenothiazine or the stable radical (2,2,6,6-tetramethylpiperidin-1-yl)oxyl are able to react very efficiently with primary radicals or growing polymer radicals and are, therefore, used in significantly smaller quantities (20–100 ppm) compared to the phenolic inhibitors. Second, solvents are important for the viscosity, wetting, and flowing behavior of primers or adhesives. Furthermore, solvents promote the monomer penetration into the collagen network of demineralized dentin and enable the dissolution or the dispersion of different components. Water, ethanol, isopropanol, and acetone are the most commonly used solvents. These solvents are inexpensive, available in a high purity, have a good biocompatibility, an acceptable odor, and an optimal volatility. Besides many monomers, initiators and stabilizers are well soluble in these solvents or in mixtures thereof. In this context, it should be mentioned that water is an indispensable component of SEAs, to ionize the strongly acidic monomers. Third, fillers, particularly nanofillers, such as colloidal or pyrogenic silica nanoparticles, are used to modify the viscosity and flow properties of the adhesive. They may also improve the strength and elastic modulus of the formed adhesive layer. Fourth, specific additives, such as functionalized polymers, which may improve the film-forming properties of the adhesive composition, compounds that modify collagen (e.g., glutaraldehyde), antimicrobial agents used as disinfectant and to potentially prevent recurrent caries, fluoride-releasing components, and dyes can also be incorporated in the adhesive formulations. Dyes are used to facilitate the control of the mixing of adhesive components or the spreading of the adhesive across the desired tooth surface.

The improvement of the currently used enamel–dentin adhesives is mainly focused on the reduction of the technique sensitivity of the adhesives. Furthermore, the adhesives should be more simple, more user friendly, and have to show enhanced durability and biocompatibility. These improvements can be attained by the use of new components, which are described in the following chapter.

New Components for Enamel–Dentin Adhesives


In recent years, the synthesis of new acidic monomers for enamel–dentin adhesives concerned mainly strongly acidic monomers, such as polymerizable dihydrogen phosphates and phosphonic acids (PA). One drawback is the hydrolytic instability of the corresponding methacrylate derivatives. This problem can be solved, for example, using monomers, which contain a hydrolytically stable group between the polymerizable group and the strongly acidic phosphorus group. (Meth)acrylamides can be used for this purpose. In this context, we synthesized a number of new (meth)acrylamido dihydrogen phosphates.30,31 Thus, 5-methacrylamido-pentyl, 10-(N-methylacrylamido)-decyl, 11-(N-methylacrylamido)-undecyl, or 1,3-bis(methacrylamido)-propane-2-yl dihydrogen phosphate (BMAPDP) (Scheme 10) were synthesized by acylation of the corresponding aminoalkanols with methacrylic anhydrid or acryloyl chloride followed by phosphorylation with POCl3. In addition, we synthesized O-alkylated acrylic acid hydroxamides, such as 10-(N-acryloyl-N-methoxyamino)-decyl dihydrogen phosphate.32 Because of their good solubility, improved hydrolytic stability, high acidity, and sufficient potential to etch enamel, these monomers are excellent candidates to enter SEA formulations. Indeed, they are able to mediate a strong bond between the enamel or dentin surface and a restorative composite under self-etching conditions. Thus, a new hydrolytically stable one-component SEA, which can be stored at room temperature, was successfully developed using BMAPDP, the cross-linker DEBAAP (Scheme 7), and the hydrolytically stable N-(5-hydroxypentyl)methacrylamide as alternative to HEMA.28,33,34 Furthermore, novel 2-(ω-phosphonooxy-2-oxaalkyl)acrylate monomers (POAA; Scheme 10) with improved hydrolytic stability were synthesized by a four-step synthesis involving a Baylis–Hillmann reaction between ethyl acrylate and formaldehyde, followed by halogenation, a subsequent etherification with various diols and a phosphorylation.35 Compared to the (meth)acrylamido dihydrogen phosphates, however, these monomers are still less hydrolytically stable because of the presence of terminal ester groups.

Scheme 10.

Structure of different dihydrogen phosphates with improved hydrolytic stability.

A number of new polymerizable PA (meth)acrylates for dental application were synthesized by the group of Avci.36–39 Examples are aryl diphosphonic acid group-containing dimethacrylates such as 2,5-bis(methacryloyloxy)-1,4-phenylenediphosphonic acid (PA-1) or 2,2-bis[3-(dihydroxyphosphoryl)-4-(2-carboxyprop-2-en-oxy)phenyl]propane (PA-3, Scheme 11). Both monomers are high melting solids and their decomposition starts near 200°C. This is a drawback for dental application, because such solids generally show a relatively low solubility, and solutions of these monomers tend to crystallize during storage. Unfortunately, 4-dihydroxyphosphoryl-2,6-dicarboxy-4-ethoxycarbonyl-1,6-heptadiene (PA-4, Scheme 11) is also a high melting solid.40 The aqueous solution of PA-1 (1 wt %) had a relatively low pH value of 1.65. However, it was found that the addition of PA-1 to HEMA decreased the maximum rate of polymerization. Unfortunately, this was also found for two recently synthesized phosphonated bis(methacrylamide)s PA-2a, b (Scheme 11), which showed increased hydrolytic stability but also a gradual decrease in the conversion of the copolymerization with HEMA.41 The aqueous solution of PA-3 (7 wt %) was significantly less acidic (pH: 1.87) compared to the pH of 7 wt % phosphoric acid (pH: 0.90). The liquid aromatic monophonic and diphosphonic acid monomers PA-5 und PA-6 (Scheme 11) were synthesized in four-step synthesis. However, during the hydrolysis of the corresponding phosphonate ester groups, it was observed that phenyl ester linkages were also prone to hydrolysis. Thus, 2-hydroxyphenyl phosphonic acid and the corresponding carboxylic acid were formed in the case of PA-5. Furthermore, simple methacrylates bearing PA groups (PA-7 and PA-9, Scheme 12), difluoromethylphosphonic acids (PA-8), or sulfur-containing monomers (e.g., PA-10) were prepared.42,43 The ability of PA-7, PA-8, and PA-9 to bond to HAP was demonstrated using 31P cross polarization (CP) magic angle spinning (MAS) NMR spectroscopy. The bisphosphonic acid PA-9 was significantly more reactive in photopolymerization experiments than the PAs PA-7 and PA-8. This result was explained by the formation of hydrogen bonds affecting the system mobility and organization during polymerization. Unfortunately, the incorporation of the sulfur methacrylate PA-10 (Scheme 12) in a BisGMA/TEGDMA (1:1) mixture decreased its polymerization reactivity. Moreover, a number of promising cross-linking phosphonates, such as the bisphenol-A dimethacrylate derivative MAP-1, were synthesized (Scheme 13).44,45 Novel N-alkylacrylmidophosphonic or bisphosphonic acids, for example, 2-N-methylacrylamidoethyl, -hexyl- or -decylphosphonic acid (PA-11a-c, Scheme 14), and 3-(N-alkylacrylamido)propylidenebisphosphonic acids BPA-1a-d (Scheme 14) were prepared.46,47 Dentin shear bond strength measurements showed that self-etching primers based on N-alkylacrylamidophosphonic acids PA-11a-c or bisphosphonic acids BPA-1a-d in combination with the bisacrylamide cross-linker DEBAAP (Scheme 7) assured a strong bond between the tooth structure and a dental composite. The monomer PA-11c with the longest spacer group provided the highest shear bond strength. Furthermore, N-alkyl-N-(phosphonoethyl)-substituted mono- (PA-12a-f, Scheme 14), bis-, and tris(meth)acrylamides were synthesized and showed improved hydrolytic stability compared to carboxylic esters.48 The highest stability was found for the phosphonoethyl-substituted acrylamide monomers, which had a larger polymerization enthalpy compared to methacrylamides. Among the monomers presented in Scheme 14, the monomer PA-11c showed the highest adhesion values in a two-step SEA. Recently, six aminophosphonate-containing methacrylates (e.g., MAP-2a, b, Scheme 13) were described by Avci and coworkers.49 The high rate of polymerization of the monomers was attributed to both hydrogen bond interactions due to NH groups as well as chain-transfer reactions. Because of the basic properties of the amino phosphonates, the corresponding PAs were not synthesized.

Scheme 11.

Structure of polymerizable phosphonic or bisphosphonic acid monomers.

Scheme 12.

Structure of methacrylates bearing phosphonic or bisphosphonic acid groups.

Scheme 13.

Structure of methacrylate phosphonates.

Scheme 14.

Structure of novel N-alkylacrylamidophosphonic or bisphosphonic acids.

Altogether, more acidic polymerizable dihydrogen phosphates are more suitable for SEAs than less acidic PA monomers. Adhesives based on acidic monomers with long spacers between the strongly acidic and the polymerizable groups resulted in a better performance. In case of MDP, good bond strength can be explained by a self-assembled nanolayering due to the length of the decamethylene spacer and by the low solubility of the corresponding calcium salt.50–53 Moreover, (meth)acrylamido groups are more suitable polymerizable groups compared to methacrylates, because they enable the preparation of all-in-one SEAs with high storage stability.

In addition to strongly acidic monomers, such as carboxylic and PAs or dihydrogenphosphates, functionalized monomers, which are able to chelate calcium ions or react with collagen chains, have the potential to improve the performance of enamel–dentin adhesives. For instance, we have shown some years ago that 2-acetoacetoxyethyl methacrylate (AAEMA, Scheme 15) was able to increase the performance of our multistep adhesive Syntac®54 by forming an insoluble calcium chelate55 and react very fast with aldehydes56 and amines57 under mild conditions.

Scheme 15.

Structure of different functionalized monomers evaluated for dental adhesives.

In this context, we recently synthesized a few monomers with the potential to form strong complexes with calcium ions. We started with the synthesis of a methacrylamide derivative of ethylenediamine-N, N,N′, N′-tetraacetic acid (EDTA) MAM-EDTA.58 Further examples are the polymerizable crown ethers 4-(methacryloyloxymethyl)-benzo-15-crown-5 (MA-B-15-C-5) and 4-(methacryloyloxymethyl)-benzo-18-crown-6 (MA-B-18-C-6), the polymerizable cryptant AA-B-2.2.2 (Scheme 15)59, and the dihydroxyphenylalanine (DOPA) derivative DOPA-MA.60 Among these very differently functionalized monomers, the polymerizable crown ether MA-B-18-C-6 showed the most promising properties in strongly acidic adhesive formulations for dentin, which additionally contained the PA monomer EAEPA (Scheme 5). Indeed, the addition of MA-B-18-C-6 resulted in a significant increase of the dentin shear bond strength of the corresponding adhesive. Crown ethers are used as complexing agents and phase-transfer catalysts in organic syntheses. Until now, the reason for the improvement in adhesion achieved with MA-B-18-C-6 could not be fully explained. In addition to its complexation properties, the excellent ability of the monomer to penetrate into demineralized dentin may play an important role, similarly to the surfactant dimethacrylate monomers.61 Several reasons cause the lower adhesive properties of the other monomers. With the cavity radius of 15-crown-5 (86–92 pm) being smaller than the ionic radius of Ca2+ ions (99 pm), MA-B-15-C-5 did not form a strong complex with Ca2+ ions. In contrast to this, 18-crown-6 has the optimal cavity size (134–143 pm) for the strong complexation of Ca2+ ions. In case of the polymerizable cryptant AA-B-2.2.2, the protonation of the basic nitrogen atoms probably impaired the chelating properties of this compound. Finally, DOPA-MA has a retarding effect on the photopolymerization of dimethacrylate resins (e.g., mixtures of BisGMA and TEGDMA), which can be explained by the well-known retarding effect of phenolic OH groups in the free-radical polymerization of vinyl monomers.62 Polymerizable β-cyclodextrin derivatives improved the dentin shear bond strength of formulations containing sorbitol dimethacrylate, methacrylic acid, and BAPO as photoinitiator. This result was explained by the formation of strong adhesive bonds to the hydrated dentin surface.63 The polymerizable cyclodextrin derivatives were synthesized by etherification of β-cyclodextrin with a mixture of 6-bromohexanoic acid and vinyl benzyl chloride.64 Finally, it should be mentioned that polymerizable ionic liquids, such as, 1-butyl-3-methylimidazolium 2-acrylamido-2-methyl-1-propanesulfonate (MIAMS) could improve the bond strength of conventional dental adhesives on dentin or enamel.65 In this context, it was found that polymerizable ionic liquids can significantly increase the air to nitrogen curing exotherm, meaning that mixtures containing polymerizable ionic liquids can be cured in air to a higher conversion.

Photoinitiators and Stabilizers

Beside the monomers, initiators represent the second key component in dental adhesives. By careful selection, a remarkable influence on the curing performance, such as polymerization rate and double bond conversion, and on the final adhesive properties is possible. As mentioned before, different kinds of initiators are used in adhesives. Photoinitiator systems, such as combinations of CQ with tertiary amines, are preferably used. The different redox systems used in adhesives or cements are based on boranes, monosubstituted barbituric acids, and the combinations of DBPO with aromatic tertiary amines or aromatic sulfinic acid salts. They were described in a recent review by Ikemura and Endo.66 Therefore, only the developments of a few photoinitiators are discussed below. Unlike the CQ-tertiary amine systems, APO and BAPO (Scheme 9) undergo a monomolecular α-cleavage (Norrish I). The initiating radicals are formed without the use of accelerators, such as basic tertiary amines, and these photoinitiators may show a high photoinitiation reactivity in dental resins and acidic adhesive formulations.67 Compared to APO, BAPO is a better alternative for dental application, because its absorption tails out into the visible range of the spectrum. The main drawbacks of APO and BAPO concern their low solubility in ethanol or aqueous solutions of polar solvents, their weak absorption in the visible range of the spectrum, and their limited hydrolytic stability. Salt formation is one possibility to improve the water solubility of the photoinitiators. Thus, sodium 2,4,6-trimethylbenzoyl-phenylphosphinate (APO-Na, Scheme 16) was synthesized.68 Unfortunately, the water-soluble APO-Na was insoluble in hydrophobic resins. However, the sodium salt could be dissolved in a hydrophobic resin in the presence of crown ethers (15-crown-5 and 18-crown-6). The addition of crown ethers (0.5 wt %) to one-step SEAs containing APO-Na (0.5 wt %) resulted in a significant increase of dentin and enamel shear bond strengths. No significant difference between the two crown ethers was observed and the improvement of the adhesive properties was explained by the ionophore effect. Another possibility to change the solubility of the photoinitiator is to introduce solubility mediating substituents. We followed this approach together with Liska and coworkers.69,70,71 In this context, three different (methoxyethoxy)-ethoxy-substituted bisacylphosphine oxides WBAPO-1, -2, and -3 (Scheme 16) which showed improved solubility in pure water or polar solvents were synthesized. For example, the solubility of WBAPO-2 in water was 2.11 mg L−1 compared to <0.2 mg L−1 for BAPO.69 Furthermore, it was found in photo-differential scanning colorimetric (DSC) experiments that the use of hydrophilic photoinitiators led to a significantly higher polymerization activity than BAPO up to a factor of 2 when the initiators were introduced in the same concentrations. Based on these investigations, we successfully synthesized and up-scaled the synthesis of bis(3-{[2-(allyloxy)ethoxy]methyl}-2,4,6-trimethylbenzoyl(phenyl)phosphine oxide (WBAPO-4). Self-etching enamel–dentin adhesives, which contained WBAPO-4 as a photoinitiator, an aqueous mixture of the hydrolytically stable cross-linker DEBAAP (Scheme 7), and the strongly acidic adhesive monomer BMAPDP (Scheme 10), did not show any decrease in the adhesion performance after storage at 42°C for 28 days. In contrast, the shear bond strength values of a corresponding adhesive based on CQ/EMBO decreased after only a few days of storage under these conditions.

Scheme 16.

Acyl or bisacylphosphinoxides with improved solubility in polar solvents.

As mentioned before, the weak absorption in the visible range of the spectrum is another significant drawback for the dental application of the commercially available APO or BAPO. To solve this problem, we first focused on the preparation of new chromophores exhibiting a red-shifted absorption maximum of the important n-π* transition of BAPO. However, the introduction of various substituents on the phosphorus aromatic or benzoyl groups did not result in a strong shift of the absorption maximum toward the visible light range of the spectrum. Nevertheless, it was found for the first time that in the case of the photopolymerization of butylacrylate using BAPO as photoinitiator, the first addition of an initiator radical to the acrylate double bond may occur in a reversible way.72 Second, we obtained very promising results with organometallic ketones containing germanium. Indeed, we observed about 30 nm bathochromic shift of the UV–Vis absorption maximum of the n-π* transition of benzoyltrimethylgermane Ge-1 (Scheme 17) compared to monoacylphosphine oxides.73 In photo-DSC investigations, Ge-1 showed nearly the same reactivity compared to CQ and in broad-band irradiation experiments, Ge-1 was significantly more reactive. The intensive yellow colored Ge-1 exhibited an excellent photobleaching behavior, which is very important for dental applications. Similarly to the acylphosphinoxides, Ge-1 undergoes an α-cleavage upon irradiation resulting in the formation of a benzoyl and a germyl radical (Scheme 17). It was found later that acetyltriphenylgermane CH3COGePh3 too is a high-performance photoinitiator.74 In this context, it should be mentioned that the visible light-initiated free-radical-promoted cationic polymerization of cyclohexene oxide was successfully initiated by a mixture of Ge-1 and an onium salt, such as diphenyliodonium hexafluorophosphate.75 We also had an interest in the preparation of a BAPO analogous compound based on germanium. Dibenzoyldiethylgermane Ge-2 was successfully synthesized in a two-step procedure: 2-phenyl-1,3-dithiane was first deprotonated with n-butyl lithium (n-BuLi) and reacted with diethyldichlorogermane. The subsequent removal of the dithioacetal protecting group was performed with large excess of iodine and CaCO3 to give Ge-2 (Scheme 17).76,77 As expected, the red-shift of the long wavelength absorption maximum of Ge-2 (λmax = 418 nm) was more than 20 and 50 nm compared to both n-π* transitions of BAPO at 369 and 397 nm. Furthermore, Ge-2 showed a significantly increased quantum yield of photodecomposition compared to BAPO, an excellent photobleaching effect and an outstanding photoinitiation activity in aqueous acidic formulations or dental composites.78 The photolysis of Ge-2 in the absence of monomer causes scission and multiple coupling reactions forming polygermanes.79 To optimize the properties of Ge-2, a number of new dibenzoylgermanium derivatives were synthesized (Scheme 17).80 Among these derivatives, the methoxy-substituted compound Ge-3 showed the best performance. Indeed, it exhibits a significantly stronger absorption in the visible region of the spectrum compared to CQ (Fig. 1), a high photoinitiation activity, excellent bleaching properties, and it could be synthesized in large scale with good yield. A significantly red-shifted visible absorption was found for the bis(germyl)ketone Ph3GeCO-GePh3, which showed an absorption band around 514 nm.81 Unfortunately, this compound is very unstable and we were not able to prepare resins or composites based on it. Finally, it should be mentioned that as a further approach toward photoinitiators for water-borne, amine-free dental adhesive formulations, we investigated different hydrophilic hydrogen donors, such as poly(ethylene oxide) (PEO) or phenylglycine derivatives as Norrish Type II photoinitiators and a thioxanthone derivative (TXD) as a visible light photoinitiator alternative for CQ (Scheme 18).82–84 Unfortunately, these variations of different Norrish Type II photoinitiators were significantly less efficient compared to the Norrish Type I photoinitiator Ge-3.

Figure 1.

UV–Vis spectra of Ge-3 and CQ (1.0 mmol L−1, in acetonitrile).

Scheme 17.

Benzoylgermanium derivatives: radical formation, synthesis of dibenzoyldiethylgermane, and structure variations.

Scheme 18.

Hydrophilic hydrogen donors and a TXD for water-borne, amine-free dental adhesive formulations.

Selected additives may also strongly influence the performance of dental adhesives. In this article, only stabilizers and oxygen scavengers will be briefly described. As mentioned before, phenols have been applied for many years as effective inhibitors of the autopolymerization that occurs during the synthesis and storage of acrylic monomers. In particular, BHT and MEHQ have been frequently used to inhibit the spontaneous polymerization of adhesive monomers used in dental compositions and to effectively prevent premature hardening. Indeed, the stabilizers BHT and MEHQ usually provide the required shelf-life stability for the adhesives. In context with the development of single-component SEAs based on our hydrolytically stable cross-linking bisacrylamide DEBAAP (Scheme 7) and the strongly acidic PA monomer EAEPA (Scheme 5), we found that the phenolic stabilizers are depleted from the system via a nonradical reaction with the acrylic double bond in addition to the stabilizing effect.85 We proposed a nucleophilic C-addition of the electron-rich hydroxyarenes to the acrylic double bonds (Scheme 19) based on a paper of Krawczyk et al.,86 which described the conjugate C-addition of acrylates with a series of aryl oxides. This assumption was also based on the facts that the acidic reaction environment assists the addition and that protic solvents promote proton transfer by stabilizing charged intermediates. The used model systems contained either the cross-linker DEBAAP or GDMA for comparison. Figure 2 represents the depletion of MEHQ and tert-butylcatechol (tBC) in its mixtures with a cross-linker and EAEPA at 42°C. It was demonstrated that the methacrylic vinylidene double bond is less sensitive to Michael addition than the acrylic vinyl group, it reacted 36 times slower with MEHQ and 49 times slower with tBC. We also evidenced similar depletion results with other stabilizers such as BHT, tert-butylhydrochinon, 5-methoxy, or 5-dimethylamino pyrimidol. The most notable conclusion from these experiments is that in adhesives based on acrylamides, the depletion of phenolic stabilizers is catalyzed by acidic mixtures and cannot be neglected. Increased concentration of the acid monomer and higher acidity accelerate this undesirable process. Nevertheless, we found other efficient stabilizers for acrylamides in acidic formulations.

Figure 2.

Depletion of MEHQ and tBC in the presence of EAEPA and the cross-linker during storage at 42°C: •: DEBAAP+MEHQ; ○: DAAP+tBC; ▴: GDMA+MEHQ; [trio]: GDMA+tBC.

Scheme 19.

Depletion of the stabilizer MEHQ by the reaction with the bisacrylamide DEBAAP.

The inhibition of photoinduced free-radical polymerization by molecular oxygen is one of the most challenging problems in thin film application of adhesives or coatings. It is well known that molecular oxygen deactivates excited states of photoinitiators and scavenger radicals leading ultimately to a prolonged inhibition period and tacky, insufficiently cured surfaces. This inhibitory effect is caused by the biradical ground state of oxygen, making the molecule accessible to convert propagating radical centers to peroxy radicals, which show a lower reactivity toward polymerizable double bonds.87 Because of the triplet ground state of molecular oxygen, an energy transfer with excited triplet states of the photoinitiator is also possible. As a consequence of those quenching processes, the lifetime of the excited triplet states is reduced, impeding the polymerization in the initial state, especially for Norrish Type II photoinitiators such as CQ-amine systems. The consequence of these inhibitory effects is a decreased rate of polymerization and lower double bond conversion. The oxygen inhibition takes place in particular when the radical photopolymerization of less reactive methacrylates is initiated by photoinitiators, which absorb the visible light of the spectrum. However, in case of enamel–dentin adhesives, the oxygen inhibition can be overcome by covering the adhesive layer with the restorative composite before curing. Nevertheless, oxygen scavengers are also of interest for dental applications. To prevent oxygen inhibition, various approaches have been successfully tested, such as the use of an inert gas atmosphere, the use of high-intensity irradiation sources, the addition of highly volatile monomers, such as methyl methacrylate, or of suitable additives like amines, thiols, phosphines, and so forth.87 In recent years, the addition of silanes, boranes, phosphines, germanes, or zirconium complexes found remarkable attention (Scheme 20).88–92 We tested a number of these additives (e.g., diphenylsilane and triphenylgermane) in case of the photopolymerization of dimethacrylate resins initiated with the CQ/EMBO system and found a significant reduction of the oxygen inhibition. However, compositions containing these additives showed no storage stability and were probably deactivated by oxidation with aerial oxygen. The oxygen scavengers 2,5-diphenyl-furan (DPF) and 9,10-dibutylanthracene (DBA) (Scheme 20), which react with singlet oxygen in a [4+2] cycloaddition reaction, induced improved storage stability.93 We tested these oxygen scavengers in photocuring of methacrylate and acrylate systems with a CQ/EMBO mixture as the photoinitiator. Both, photo-DSC and ATR-IR-spectroscopy measurements indicated an enhancement of the polymerization by the addition of the anthracene DBA or the furan DPF as oxygen scavenger.

Scheme 20.

Oxygen scavengers for reduced oxygen inhibition in radical photopolymerization.

The adhesive technology has undergone great progress in the last 10 years by the introduction of new components, optimization of the adhesive compositions, and the implementation of novel application systems. Nevertheless, further improvements are possible. Basically, the dental adhesives have to show an efficient clinical performance and should be safe and easy to apply. Future developments of the dental adhesives are the improvement of the adhesive performance, the bond durability, biocompatibility, storage stability, and the handling properties of the adhesives. Thus, new components may contribute to improve the biocompatibility and stability of adhesives to prevent the degradation of the bonding interface. For example, the use of inhibitors for matrix metalloproteinases should be able to enhance the durability of the adhesives. Compounds able to accelerate the curing (light curing combined with self-curing) are also of interest. Further trends involve the development of adhesives, which can be used in both etch-and-rinse and self-etch technology and the formulation of so-called universal adhesives applicable for all dental substrates (enamel, dentin, ceramics, metal alloys, and composites). The addition of components, which show antimicrobial activity or may contribute to the remineralization of the tooth substrate, is also under investigation. Finally, the improvement of the radiopacity of the adhesive layer, which could be realized by the introduction of radiopaque nanoparticles, as well as the creation of self-healing/self-repairing or debonding-on-demand properties are also of interest.


State of the Art and Organic Components of Current Composite Restoratives

Composites have been used as direct restorative materials to replace missing biological tissue for more than 40 years. The term “composite” subsumes materials, which are composed of at least two different phases. In the case of resin composites, these include the polymerizable organic matrix and fillers (Scheme 21). The organic matrix of a photopolymerizable composite restorative, to which the filler particles can be covalently bonded, is based on a mixture of polymerizable monomers or oligomers, the photoinitiator system, and further additives such as stabilizers, colorants, or optical brighteners. Nearly all current composite restoratives are hybrids, that is, they contain at least two different fillers, for example, of different shape, chemical composition, or size. The fillers are surface-modified with suitable coupling agents to improve the linkage between the fillers and the organic matrix. Resin composites are used for a variety of applications in dentistry, including direct restorations and indirect restorations, such as inlays or onlays, crowns, bridges, veneers as well as cements, fissure sealants, or cavity liners. The different types of dental composite materials are distinguished mainly by their viscosity, the type of fillers used, and the applied type of resin (Scheme 22).94,95 Packable composites exhibit a higher viscosity, are condensable like amalgam, and enable the formation of tight interproximal contacts. Packable composites were developed to limit the polymerization shrinkage, wear, and fracture of the restoration within the body and at the margins of the restorations. Compared to the universal composite restoratives, which can be placed with a common dental syringe, flowable composites show a lower viscosity. Thus, flowable composites can be used to penetrate small pits and grooves in teeth avoiding the need to remove additional tooth structure. Fluoride containing, flowable composites are ideally suited as preventive sealants, especially for children. The viscosity or consistency of a composite can be reduced by the reduction of the filler content, the increase of the average filler particle size, and the reduction of the viscosity of the organic resin. Because of the major influence of the fillers on the physical properties, the frequently used classification of dental composites is based on the type and the particle size of the reinforcing fillers. Subsequently, resin composites can be classified into fiber-reinforced composites (FRC) and particulate filler composites. FRCs, which can be divided based on the fiber length in long-fiber and short-fiber composites or according to the fiber orientation in composites with continuous unidirectional and bidirectional fibers. In clinical dentistry, FRC technology may solve some problems associated with a metal alloy substructure, such as corrosion or toxicity.96 However, FRCs are not used as direct restoratives, which are generally based on particulate fillers. According to the particle size, the particulate fillers can be divided into macrofillers (10–100 µm, used in the first composite restoratives), midifillers (1–10 µm), minifillers (0.2–1 µm), and nanofillers. Microfill composites contain fillers with a primary particle size between 5 and 100 nm, which have to be called nanofillers instead of microfillers.95 To enhance the handling properties and achieve a higher filler load, heterogeneous microfill composites may contain prepolymerized resin fillers that are based on a homogeneous microfill material. Microfill composites are generally weak due to their relatively low filler content. They were developed to satisfy the need for esthetic, more polishable composites. Therefore, the presently used universal composites, which can be used for most anterior and posterior applications, are hybrids based on a mixture of minifillers and nanofillers and produce an optimal combination of adequate mechanical strength with sufficient polishability. According to the particle size of the fillers, the hybrids can be further distinguished in microhybrids and nanohybrids. Microhybrids are based on a mixture of minifillers, whereas nanohybrids contain only nanoscale particles or a dominant proportion of nanoparticles.

Scheme 21.

Composition and use of dental resin composites.

Scheme 22.

Classification systems of dental composite restoratives.

Restorative composites have to meet extensive requirements, which include physical, chemical, processing, toxicological, and clinical requirements. Excellent properties of composite restoratives can be obtained by the optimum choice of the fillers, their suitable surface modification, and various components in the organic matrix. Thereby, each component influences the composite properties in a different manner. The curing rate, light sensitivity, polymerization shrinkage, color stability, water uptake, or storage stability are mainly influenced by the organic matrix (Scheme 23). Typically filler-determined properties are wear resistance, hardness, radiopacity, gloss, polishability, or the release of fluoride ions. Moreover, the flexural strength, flexural modulus of elasticity, consistency, translucency, thermal expansion coefficient, water solubility, or biocompatibility is dependent on the chemical composition of both the organic matrix and the fillers. Finally, for the long-term stability or durability of the composites, the interphase between the matrix and the fillers is very important. The clinical success of a light-cured direct composite restoration also depends strongly on additional factors, such as an optimal preparation or design of the cavity, a good bonding to the adhesive layer, the optimum placing in layers/increments, and sufficient curing of the composite.

Scheme 23.

Properties of light-curing dental composites influenced by different components.

Most of the currently used commercial direct composite restoratives or composite cements are based on a mixture of conventional dimethacrylates.97,98 The most frequently used dimethacrylates are BisGMA and UDMA (Scheme 6), which show a relatively low polymerization shrinkage of 6.1 and 6.7 vol %, a high reactivity in the free-radical polymerization, and form polymer networks with excellent mechanical properties. Because of their high viscosity, they have to be diluted with dimethacrylates of lower viscosity, such as TEGDMA (Scheme 6). Because of the relative high water solubility and high cytotoxicity of TEGDMA,7 more hydrophobic diluents, such as decanediol dimethacrylate (D3MA) or bis-(3-methacryloyloxymethyl)tricyclo-[,6]decane (TCDMA, Scheme 24) are also used. The selection of appropriate dimethacrylate cross-linkers for the formulation of a composite strongly influences the reactivity, viscosity, and polymerization shrinkage of the composite paste as well as the mechanical properties, water uptake, and swelling of the cured composite. Alternatives to BisGMA or UDMA have been used. The ethoxylated (BisEMA), propoxylated bisphenol A dimethacrylate (BisPMA), and 1,3-bis[1-(2-methacryloyloxypropoxycarbonylamino)-1-methyl-ethyl]benzene (TMXDMA) are three examples shown in Scheme 24. In this context, it can also be mentioned that the partially aromatic urethane dimethacrylate TMXDMA showed similar reactivity, flexural strength, flexural modulus of elasticity, polymerization shrinkage, and water sorption to materials based on BisGMA.99 Trimethacrylates such as trimethylolpropane trimethacrylate (TMPTMA) are rarely used in composite restorative materials, as they entail low double bond conversion. In contrast to this, monomethacrylates improve the double bond conversion; however, they may result in an increase of residual monomer, a decrease of cross-linking density, and a corresponding decrease of the modulus of elasticity of the cured materials.

Scheme 24.

Several dimethacrylates used in current resin-based composite restorative materials.

Beside the conventional dimethacrylates, acidic carboxylic groups containing dimethacrylates are use in so-called compomers. The term is a combination of the terms composite and glassionomer. These water-free, single-component, light-cured composites are based on carboxylic groups containing dimethacrylates, for example, TBCDMA or CTCDMA (Scheme 25), which can be easily synthesized by the reaction of HEMA with the anhydride of the corresponding polyfunctional carboxylic acid, such as butane-1,2,3,4-tetracarboxylic acid dianhydride (TBCDMA) or 5-(2,5-dioxotetrahydrofuryl)-3-methylcyclohex-3-ene-1,2-dicarboxylic acid anhydride (CTCDMA). The compomers are reinforced with strontium or barium alumofluorosilicate glass particles. In the presence of water, the acid groups ionize and undergo an acid–base reaction with the basic glass components and potentially show anticariogenic properties due to fluoride release. Nevertheless, conventional hybrid composites show better physical properties compared to compomers. Therefore, compomers are mainly used in cervical class V restorations or for the restoration of decayed deciduous teeth. Another group of acidic monomers containing composites are self-adhesive composite resin cements. Presently, there are more than 15 self-adhesive composite resin cements on the market, which are based on a mixture of usual dimethacrylate cross-linkers and acid-functionalized monomers.98 For this purpose, not only the strongly acidic adhesive monomers, such as the phosphate methacrylates MDP, GDMP, or PENTA-P (Scheme 5) but also the less acidic 4-methacryloyloxyethyl trimellitic anhydride are used. These self-adhesive composite cements adhere to the tooth structure without the application of a separate adhesive or etchant and have the major benefit to simplify the adhesive restorative therapy. Until now, clinical evaluations are few and short term, so long-term conclusions about the performance of these cements in the dental practice are not yet possible.

Scheme 25.

Dimethacrylate cross-linkers for dental compomers and ormocer-based composites.

A few restorative composites are on the market, which are based on organic–inorganic hybrid materials. These materials are called “ormocers” (organically modified ceramics) and the resins are polycondensates prepared by hydrolytic condensation of suitable polymerizable trialkoxysilanes (Scheme 25).100 These hybrid materials may show interesting new properties in relation to each individual component and can be fine-tuned by adjusting the composition. The main reason for incorporating inorganic units in the organic matrix is to increase the biocompatibility and wear resistance of the dental composites. The first commercialized ormocer restorative composite was based on the silane UDMASi, which was prepared by the reaction of GDMA and 2-isocyanatopropyltriethoxysilane. For the composite production, the silane was hydrolytically condensed, resulting in a liquid UDMASi-polycondensate, the ormocer. The viscosity of the UDMASi-polycondensate was very high and therefore during the composite manufacturing a certain amount of a common dimethacrylate diluent had to be added. Also in case of the polycondensates of 1,3-bis(methacryloyl)-2-(3-trimethoxysilyl)propoxy)propane, TEGDMA, which shows a relatively low biocompatibility, was used as diluent.101 In contrast to this, we prepared dimethacrylate-diluent-free composites with an ormocer, which was a polycondensate of the amide dimethacrylate trimethoxysilane ADMASi (Scheme 25).102,103 It was found that the viscosity of the obtained polycondensates was predominantly influenced by the kind of the organic spacer between the polymerizable groups and the silicon atom. The formed structure depended both on the type of catalyst used for silane hydrolysis/condensation and the bulkiness of the organic substituent at the silicon atom. Based on the low cytotoxicity of the synthesized ormocer, the prepared restorative composite showed good biocompatibility.

The fourth type of resins in presently used dental resin composites contains cationically ring-opening polymerizable cycloaliphatic epoxides. The resin is a mixture of the diepoxides bis[2-(7-oxabicyclo[4.1.0]hept-3-yl)ethyl]methylphenylsilane 1 (BECEPSi, Scheme 26) and the tetrafunctional cyclosiloxane epoxide 2,4,6,8-tetrakis[2-(7-oxabicyclo[4.1.0]hept-3-yl)ethyl]-2,4,6,8-tetramethyl-1,3,5,7-tetraoxa-2,4,6,8-tetrasilacyclooctane (TECTMSi). In addition, it was found that the spiro monomer 3,9-diethyl-3,9-bis(trimethylsilylpropoxymethyl)-1,5,7,11-tetraoxaspiro[5.5]undecane (TOSU) may significantly reduce the polymerization stress of these epoxide mixture.104–107 For the main monomer TECTMSi, the term “silorane” was introduced to represent hybrid monomer systems that contain both siloxane and oxirane structural units. The main benefits of these resins are the very low polymerization shrinkage, for example, 2.0 vol % in case of TECTMSi, the good biocompatibility, the very low water solubility of the monomers, and low water sorption of the formed polymer networks. Thus, not only the cytotoxicity rating of TECTMSi is as good or better compared, for example, to BisGMA, but it also appears to be nonmutagenic.108,109 Furthermore, it should be mentioned that polymerizable silsesquioxanes, which were synthesized by hydrosilylation first of 4-vinyl-cyclohexene epoxide and then of di(propylene glycol)allyl ether methacrylate with octakis(dimethylsiloxy)silsequioxane, were proposed as useful components for low-shrinkage composites.110 Although the silorane-based restorative composite was introduced in the dental market in 2007, it could not attain broad market acceptance until now. A number of reasons are responsible for this situation. The establishment of the mechanical properties during the light-curing of the silorane composite takes more time and the exothermic effect is twice as high compared to dimethacrylate-based composites with similar filler load. Furthermore, siloxane-based epoxides have a significantly lower refractive index compared to the commonly used radiopaque fillers, which affects the transparency and thereby the attainable curing depth of the composite. Because of the increased exothermic effect and the limited transparency, the very attractive application of a restorative composite in thick layers or in bulk is difficult to achieve. Moreover, the storage stability of the cycloaliphatic epoxide-based composites is significantly shorter compared to that of dimethacrylate-based composites. Finally, results of the long-term performance of silorane composites are not available and the composites have to be used with a special enamel–dentin adhesive. The adhesive system was a methacrylate-based, two-step self-etch system and resulted in a high degree of double bond conversion in the formed hybrid layer.111

Scheme 26.

Cationically polymerizable cross-linkers for dental composites.

Dental composite restoratives based on difunctional or polyfunctional methacrylates (conventional, acid-modified, or inorganic polycondensates) harden by free-radical photopolymerization. As mentioned in the previous chapter, the most widely used photoinitiator is the mixture of CQ with tertiary amines, for example, EMBO.29 In contrast to adhesives, special topics for photocured composites are the depth of cure and the bleaching behavior, particularly for highly esthetic composite applications. The depth of cure of a resin-based composite strongly depends on the transparency and the shade of the composite; darker shades may require extended irradiation. Additionally, the depth of cure is influenced by the concentration, the absorption behavior, and reactivity of the applied photoinitiator system. Further parameters are the irradiation time, light intensity, emission spectrum of the curing light, and the distance of the light probe to the restoration. The transparency of a resin composite again is mainly influenced by the refractive index of the used monomers and fillers, the particle size, shape and content of each filler, and the color and the content of pigments. In case of nanosized fillers, the agglomeration of the nanoparticles may significantly impair the transparency of the composite. A method to improve the depth of cure, bleaching behaviour, and color stability of the composite restoratives is the combination of CQ-amine with other photoinitiators, for example, the acylphosphine oxide APO, the bisacylphosphine oxide BAPO (Scheme 9), or 1-phenyl-propandione-1,2 (PPD). In the silorane composite, a mixture of CQ, EMBO, and a diphenyl-iodonium salt is used as photoinitiator.103

Fillers in Current Composite Restoratives

The particulate fillers used in dental composites directly influence properties such as radiopacity, abrasion resistance, intrinsic surface roughness, flexural modulus, coefficient of thermal expansion, and translucency. Composite restoratives require filler materials that can adjust the mechanical properties, color, translucency, and so forth, of teeth. The fillers are characterized by their shape, morphology, average particle size and corresponding surface, particle size distribution, chemical composition, density, and refractive index. As mentioned before, the particulate fillers can be classified according to the particle size into macrofillers, midifillers, minifillers, and nanofillers (Scheme 23). In principle, the particle size can be realized by either a top-down manufacturing approach, for example, a conventional milling process of glass, or bottom-up approach, for example, new sol-gel routes or flame-spray pyrolysis (FSP) starting from molecular precursors. The size of the filler particles incorporated in the resin matrix of commercial composites has continuously decreased over the last 10–15 years from the traditional composites to the nanohybrids. Many modern dental composites use fillers listed in Table 1. The scanning electron microscopic (SEM) picture of some fillers, which we use in our current direct composite restoratives, is shown in Fig. 3. Typical midifillers and minifillers are purely inorganic, usually splinter-shaped, and prepared by grinding larger particles of radiopaque glass, quartz, or ceramics into smaller particles. Today, they have an average particle size between 0.2 and 5.0 µm and a specific surface between 2 and 25 m2 g−1. The glass fillers are made from melts of different mixtures of SiO2, Al2O3, B2O3, BaO, SrO, CaO, Na2O, and, if necessary, additional compounds. Thereby, elements such as barium or strontium contribute to a high radiopacity (300–700% Al) and a high refractive index (1.50–1.55) of the glass fillers. Radiopacity, the visibility of a filling in X-rays, is an important clinical requirement for direct composite restoratives. This requirement can be very efficiently fulfilled by the use of ytterbium trifluoride (YbF3), which we widely use as radiopaque filler in our composites. A benefit of this filler is that with YbF3 radiopaque composites can be prepared, which additionally release fluoride ions. Moreover, fluorosilicate glasses or sparingly soluble fluoride salts are also added to composites for fluoride release. Because of the milling process, the ground glasses have a relatively broad particle size distribution. The abrasion resistance of the cured composite is strongly influenced by the particle size and form of the fillers. In the case of ground glasses, the mean particle size is less crucial than the maximum particle size.

Figure 3.

SEM pictures of different fillers used in current dental composites: a: Ba-silicate glass (0.4 µm), b: Ba-silicate glass (0.7 µm), c: spherical Zr/Si-oxide, d: pyrogenic silica, e: ytterbium trifluoride, and f: prepolymer filler.

Table 1. Types of Fillers and Filler Size Used in Currently Dental Composites
inline image

Pyrogenic silica is an important representative of nanofillers. The nanoparticles are frequently spherical in shape. The most commonly used pyrogenic silica shows a primary particle size of 40 nm and a specific surface of about 50 m2 g−1. Because of the large specific surface area of the nanofiller particles, they have a pronounced thickening effect. Often, the primary particles of nanofillers are agglomerated and thus the thickening effect is reduced or can be reduced by special thermal treatments. However, the formed agglomerates influence the transparency of the composite. Another possibility to attain maximum inorganic filler loading with nanofillers is to incorporate the nanofillers, for example, pyrogenic silica, into an organic resin matrix, heat-cure the mixture, and mill the formed composite material to the desired particle size of 10–100 µm (Table 1). With these larger prepolymerized filler particles, a higher filler content can be achieved, which reduces the polymerization shrinkage. Furthermore the density of the prepolymerized filler particles is significantly lower than that of conventional pure inorganic fillers. Therefore, composites containing prepolymerized fillers exhibit less polymerization shrinkage than composites composed of inorganic filler with the same wt % content of monomers.

Tailor-made nanofillers can be prepared via the sol-gel route, starting from tetraalkyl orthosilicates or metal alkoxides such as titanium(IV) and zirconium(IV) ethoxide or mixtures thereof. Sol-gel nanofillers mostly consist of spherical primary particles and particles with an average particle size less than approximately 40 nm and of very narrow particle size distribution. The particle size distribution of such nanoparticle fillers is frequently monodisperse. However, metal alkoxides are relatively expensive, their hydrolytic condensation is accompanied by large volume shrinkage, and the sol-gel process is difficult to scale-up. Therefore, one interesting alternative approach to sol-gel or pyrogenic nanofillers is the use of so-called organosols.100 Organosols are stable colloidal dispersions of inorganic nanoparticles in organic media (solvents or monomers). Organosols may contain discrete almost nonagglomerated nanoparticles with a very narrow particle size distribution. For example, transparent organosols of silica particles with a size of 10–30 nm are commercially available, which are prepared starting from silicic acid in water under special conditions. The water of the resulting aqueous silica sol is subsequently replaced by water-soluble solvents, such as alcohols. Then the solvent can be replaced by monomers, for example, HEMA or dimethacrylates. Before the silica nanoparticles are incorporated into the polymerizable monomers, they are thoroughly surface silanized with a polymerizable silane, for example, 3-methacryloyloxypropyltrimethoxysilan (MTMSi) to stabilize the particles in the dispersion and to obtain a covalent bond between the filler and the matrix after curing the composite. Thus, we prepared different dental dimethacrylate monomer-based organosols and investigated their viscosity and transparency as a function of varying amounts of silica nanofillers with a particle size of 13 nm in diameter. Compared to dispersions of the usual pyrogenic silica, the silica organosols showed a significantly lower viscosity and a higher transparency. It was found that the capacity of the filler load is strongly dependent on the initial viscosity of the dimethacrylate, but the hydrophilicity and polarity of the monomer as well as an optimal surface modification of the SiO2 nanoparticles were also important. For example, in the case of TEGDMA, SiO2 organosols can be filled with about 50 wt % of nonagglomerated silica nanoparticles without showing an enormous increase in viscosity or any pseudoplastic behavior (Fig. 4). In contrast to this, a dispersion of pyrogenic silica in TEGDMA, which contains agglomerated SiO2 nanoparticles of the same primary particle size (12 nm), showed a significant increase of viscosity at a SiO2-content of only few wt %. Compared to TEGDMA, Bis-EMA is less polar and shows a higher viscosity. Therefore, Bis-EMA-based organosols exhibited a significant increase in viscosity already at a lower SiO2-content of about 30 wt %. Silica sols also enable the synthesis of tailor-made agglomerates. One example is the so-called nanoclusters, which were synthesized from a silica sol and a zirconyl salt.112 The formed zirconia-silica nanoparticles are not clusters in the sense of a chemical definition but are rather a specific kind of agglomerates. Nevertheless, the dental nanocomposites based on them showed high translucency, high polish, and polish retention similar to microfill composites while physical properties and wear resistance were equivalent to commercial hybrid composites. Silica nanofillers, which are prepared by usual pyrolysis of SiCl4 in the H2/O2-flame, contain mainly agglomerates or aggregates of nanoparticles and produce composites with limited transparency. Using the versatile FSP, we synthesized radiopaque nonporous mixed-oxide nanoparticles, for example, ytterbium oxide mixed silica (6–14 nm in diameter) or Ta2O5-containing SiO2 particles (6–10 nm).113,114 By controlling the FSP parameters, such as the precursor mixture, the solvent composition, or the liquid-to-gas ratio in the feed of the methane flame, it was possible to adjust both the composition, size of particles, and the refractive index of the nanofillers. Based on the synthesized nanoparticles, transparent nanocomposites with a high radiopacity were prepared. Finally, it should be mentioned that glass-ceramics with a wide range of compositions can also be prepared by sol-gel technique. For example, a ceramic filler with the composition SiO2:60-P2O5:3-Al2O3:14-CaO:6-Na2O:7-K2O:10 (wt %) was manufactured for dental application from an aqueous mixture of the precursors nitric acid, tetraethoxysilane, triethyl phosphate, aluminum nitrate nonahydrate, potassium nitrate, sodium nitrate, and calcium nitrate tetrahydrate.115

Figure 4.

Shear viscosity of dispersed SiO2 particles (12 nm primary particle size) measured at a shear rate of 1 s−1 at 20°C: silica organosol in TEGDMA (♦) or Bis-EMA (▴). Dispersion of pyrogenic silica: TEGDMA (▪) or Bis-EMA (•).

Before the composite paste preparation, the surface of the fillers has to be modified with a suitable coupling agent to ensure a strong and durable bond between the formed polymer network matrix and the filler particles. The filler treatment is very important for the long-term durability of the cured composite, in particular, for the mechanical properties, for example, tensile strength or fracture toughness, fatigue resistance, and wear resistance. In case of metal oxide nanoparticles, which typically show a large specific surface, the surface treatment also may result in a significant reduction of the thickening effect of the nanoparticles, which is caused by the attractive interaction of the large number of free Me[BOND]OH groups (Me: Si, Zr, or Al) on the surface. Most of the usual dental fillers (Table 1) contain silica, which can be efficiently surface-treated with functionalized silanes, such as the already mentioned MTMSi. Furthermore, polymerizable dihydrogen phosphates or phosphonates can be efficiently used as coupling agent for ZrO2 or other metal oxides.116

The physical properties of dimethacrylate-based composite filling materials such as flexural strength, flexural modulus, or polymerization shrinkage mainly depend on the used filler types, particle size, and filler load (Table 2).95,97,117 Most of the current commercial composites meet the required flexural strength (≥80 MPa) for light-curing direct filling resins according to the ISO Standard 4049.118 Flowable microfill and nanohybrid composites show the poorest mechanical properties. Nevertheless, a direct comparison of the composite performance in a clinical situation based on selected properties is not possible. The improvement of the performance of composite restoratives using new components involves the following issues:94,95,119,120 (1) reduction of polymerization shrinkage and the contraction stress. (2) Enhancement of wear resistance, strength, and fracture toughness. (3) Improvement of the handling (less stickiness to instruments and less technique sensitivity) and shortening of the processing time (shorter curing, larger increments, or bulk placement). (4) Improvement of biocompatibility (low cytotoxicity and less elution of components) and the incorporation of antibacterial agents and pH controlling components.

Table 2. Correlation Between Composite Type, Filler Load, and Physical Properties of Dimethacrylate-Based Composites
inline image

New Components for Direct Composite Restoratives

A number of advances and developments in composite dental restorative materials mainly until 2009 addressed to dental researchers were reviewed by Cramer et al.121 and, recently, a short review highlighted the use of natural compounds for the development of dental resins.122 In the following chapters, new polymer-chemical developments of direct restorative composites, that means new methacrylates monomer resins, photoinitiators and additives based on scientific papers, patents and own investigations are described and their practical relevance is discussed.

Methacrylate Monomers

The main motivation for the synthesis of new methacrylate cross-linkers is to overcome the drawbacks of the conventional dimethacrylates, such as high polymerization shrinkage and contraction stress, severe stickiness, or limited reactivity in the free-radical homopolymerization and copolymerization. Furthermore, publications concerning the estrogenic effect of bisphenol A and of thereon-based polymers have also stimulated the synthesis of bisphenol A core-free cross-linkers and new multimethacrylates with improved biocompatibility. One current main synthetic strategy for dental cross-linkers has focused on the synthesis of urethane group-containing monomers or oligomers, because the reaction of different isocyanates with diols or polyols enables the targeted synthesis of a variety of different monomer structures. Moreover, besides BisGMA, the urethane dimethacrylate UDMA is one of the most frequently used dental monomers and shows polymerization shrinkage (6.1 vol %) similar to BisGMA (6.0 vol %) and low viscosity of about 10 Pa s (BisGMA: 800 Pa s). UDMA can be easily synthesized by a simple one-pot addition reaction of the commercially available 2,4,4-trimethylhexamethylene diisocyanate (TMDI) and HEMA and thereon-based materials show an improved toughness compared to that of BisGMA. New urethane methacrylates were synthesized via a number of different routes: first, dimethacrylates and trimethacrylates prepared by the reaction of commercially available diisocyanates and triisocyanates with HEMA, HPMA, or tailor-made hydroxylalkyl (meth)acrylates in stoichiometric ratio. Second, urethane dimethacrylates or multimethacrylates synthesized by the reaction of diols or polyols with the commercially available, but expensive, 2-isocyanatoethyl methacrylate (IEMA). Third, oligomers by the reaction of oligomeric diols with an excess of diisocyanate followed by the reaction of the terminal isocyanate groups with hydroxyalkyl methacrylates. According to the first route, a number of new aliphatic, cycloaliphatic, and aromatic urethane dimethacrylates, for example, the phenyl group-containing monomer P-UDMA (Scheme 27), were synthesized and showed a significantly reduced water sorption and solubility in comparison to UDMA.123 Analogously, the hydrophobic urethane dimethacrylate TCDA bis-(3-acryloyloxyethyloxy carbonylamino methyl) tricyclo-[] decane was synthesized by the reaction of the corresponding tricyclo[]decane diisocyanate with 2-hydroxyethyl acrylate and resulted in dental composites with low polymerization shrinkage stress, good mechanical properties, and a low cyctotoxicity.124,125 The low cyctotoxicity of TCDA was surprising because acrylates usually show a stronger cytotoxicity compared to methacrylates and also mostly a mutagenic effect. Indeed, we found no mutagenic effect of pure TCDA in the Ames-Test. According to the second route, 1,1,1-tri-[4-[(methacryloylethylaminocarbonyloxy)-phenyl]ethane (UTMA, Scheme 27) was synthesized by the reaction of 1,1,1-tris(4-hydroxyphenyl)]ethane with IEMA.126 Although UTMA was proposed in combination with BisGMA and HEMA as suitable cross-linker for adhesives, we also found very promising mechanical properties for thereon-based composites. The reaction of IEMA with 2,2,4,4-tetramethyl-1,3-cyclobutanediol (TMCD) resulted in the formation of TMC-UDMA as a semicrystalline resin.127 Better miscible urethane dimethacrylates were synthesized by a two-step synthesis: first, the reaction of the diol TMCD with a diisocyanate, such as isophorone diisocyante (IPDI) or TMDI in the molar ratio 1:2, and second, the reaction of the formed diisocyanate with HEMA. The synthesized TMCD-based urethane dimethacrylates enabled the reduction of the polymerization stress in resins or composites with TEGDMA. Lower shrinkage strain and a higher degree of conversion was found for the new oligomeric urethane dimethacrylate IEO-UDMA (Scheme 28), which was prepared via the reaction of 2 equiv. IPDI and 1 equiv. polyethylene glycol 400, followed by the reaction with HEMA.128 However, analogous urethane dimethacrylates with a longer spacer, for example, based on PEO diol Mn = 1000, enhanced the hydrophilic character of the cross-linker, which is usually a general drawback for dental composite application.129 By incorporation of L-tartaric acid, acidic urethane dimethacrylates such as the monomer ITA-UDMA were synthesized, which can be used as cross-linkers in compomers.130 Furthermore, methacrylate-based monomers containing an urethane linkage and a bisphenol A central spacer group were synthesized.131 Thus, highly filled composites (filler content: 81.4 wt %) with a polymerization shrinkage of less than 1.4 vol % were prepared with the dimethacrylate EBP-UDMA (Scheme 28). It was synthesized by the reaction of ethoxylated bisphenol A with IEMA. Multifunctional uretane-methacrylate derivatives showing a high monomer conversion, low volume shrinkage, and a refractive index similar to that of the usual dental glass fillers were synthesized by the reaction of diisocyanates or triisocyanates with phenyl group-containing hydroxyalkyl methacrylates.132 One example is the urethane dimethacrylate TBPP-UDMA, a reaction product of TMDI with a 1:1 (mol/mol) mixture of 2-hydrox-3-phenoxy- and 3-(4-tert-butylphenoxy)-2-hydroxypropyl methacrylate. This dimethacrylate contributes to a relatively low polymerization shrinkage of the corresponding composites. However, the viscosity of this very sticky monomer is very high and a so filled composite is not easy to handle. Finally, another advantage of the synthesis of urethane dimethacrylates is the simple incorporation of functional units. Thus, for example, monomers with thermoresponsive or photoresponsive spacer units are available by the reaction of OH-terminated functional compounds with IEMA or 1:1 reaction products of diisocyanates with HEMA or HPMA. One example is the monomer UV-UDMA (Scheme 28), reaction product of the commercially available UV photoinitiator 1-[4-(2-hydroxyethoxy)-phenyl]-2-hydroxy-2-methyl-propanone with IEMA. This cross-linker enables the synthesis of polymer networks, which may undergo photodegradation and can be used for the preparation of low stress flowable dental composites.133

Scheme 27.

Urethane (meth)acrylates synthesized for dental composites.

Scheme 28.

Urethane (meth)acrylates synthesized for dental composites.

Several examples of hydrophobic monomers based on a dimer acid structure were synthesized by Stansbury and coworkers.134 The polymerizable dimer acid derivatives, for example, the cross-linker DA-1 (Scheme 29), produced higher degrees of conversion combined with lower polymerization shrinkage and water sorption values in comparison with conventional dental dimethacrylates. However, the cured materials based on DA-1 showed a higher flexibility and a low modulus of elasticity. Nanohybrid composites based on the dimethacrylate derivatives of dimer acid performed similar with regard to the mechanical properties and the behavior of the materials after aging.135 The polymerization shrinkage of polymerizable dimer acid derivatives can be further reduced by the introduction of free-radically ring-opening groups instead of the methacrylate groups. Thus, we synthesized polymerizable dimer acid derivatives, which contain 7-methylene-1,5-dithiaoctan-3-yl groups (DA-2) or 2-vinyl cyclopropyl groups (DA-3).136 Hybrid monomers containing both polymerizable methacrylic and 7-methylene-1,5-dithiaoctan-3-yl groups were combined in an isocyanurate methacrylate cross-linker and resulted in dental composites with very low polymerization shrinkage.137 Hydrophobic cross-linkers were also synthesized starting from biphenyl-2,2′-diol. Although the dimethacrylate EBP-DMA, which contains an ethylene spacer, crystallizes during storage, the corresponding monomer with a hexamethylene spacer was obtained as a liquid. Composites of these monomers showed improved stability. For example, the flexural strength of composites decreased by less than 10% after storage in water at 60°C for 7 days.138 A very promising class of hydrophobic compounds are calixarenes. Calixarenes are macrocyclic compounds, which are easily formed by condensation, for example, of p-tert-butyl phenol with formaldehyde under formation of the cyclic tetramer, hexamer (so-called calix[6]arene), or octamer in high yields.139 Usually, calixarenes show relatively high melting points and a low solubility in common organic solvents and can be modified by O-acylation and O-alkylation. Accordingly, an acid-catalyzed esterification of p-tert-butylcalix[6]arene with octanoic anhydride followed by the methacrylation of the residual hydroxy groups with methacrylic anhydride (Scheme 30: (a)) mainly resulted in the formation of a pentaoctanoatemonomethacrylate CAOMA, which showed a significantly improved solubility, for example, in TEGDMA.140,141 Dimethacrylate-containing composites with a filler load of 60 or 85 wt % based on a monomer matrix of cross-linking dimethacrylates and 0–30 wt % CAOMA were prepared. The addition of CAOMA to the dental composite formulations resulted in a significant decrease of the polymerization shrinkage, whereas the modulus of elasticity of the visible light-cured composites was not affected. Furthermore, we also observed the same effects for polymerizable calixarenes in composites based on calix[4]arene-dimethacrylate derivatives CA4DMA (Scheme 30: (b)), which we synthesized by dialkylation of p-tert-butyl-calix[4]arene followed by diacylation with methacryloyl chloride.142 A functionalized calix[4]arene CA4N bearing 1,3-dipolaric nitrone groups was prepared by the reaction of N-methylhydroxylamine with the carbonyl groups of 5,11,17,23-tetraformyl-25,26,27,28-tetrabutoxycalix[4]arene.143 The nitrone groups may undergo a 1,3-dipolar addition reaction with (meth)acrylic double bonds. Thus, calixarene nitrones were able to react with typical dental dimethacrylates under formation of new multifunctional cross-linkers of higher molecular weight. Accordingly, the polymerization shrinkage of composites based on a BisGMA/TEGDMA mixture could be reduced from 2.8 to 1.8 vol % by substitution of 7 wt % of the resin matrix with the calix[4]-tetranitrone CA4N.

Scheme 29.

Free-radically polymerizable dimeric acid derivatives.

Scheme 30.

Synthesis of the methacrylated calix[6]arene CAOMA (a), calix[4]arene-dimethacrylate derivatives CA4DMA (b), and calix[4]arene-tetranitrone CA4N (c).

A further strategy to improve the performance of dental resins is the use of natural compound-based monomers. The main goal of this concept is to improve the biocompatibility of the used resins. A common approach is to synthesize methacrylate cross-linkers by the use of natural compounds as the basic scaffold. Thus, bile acids were used as natural precursors for the synthesis of multimethacrylates. Bile acids are steroid acids found predominantly in the bile of mammals. Cholic acid, one of prime bile acids, possesses three hydroxyl groups as well as a carboxylic acid moiety and was methacrylated by the use of methacrylic acid, methacryloyl chloride, or methacrylic anhydride.144 Cholic acid derivatives with 2, 3, or 4 methacrylate groups (BAMA, Scheme 31) were synthesized and composites were prepared based on these multimethacrylates.145 Unfortunately, the not very biocompatible TEGDMA has been used as diluent, because the cholic acid derivatives are solids or high-viscous liquids. Nevertheless, the multimethacrylate derivatives of cholic acid were less cytotoxic compared to BisGMA; most of them showed no effect on cell viability over the range of their solubility. In addition, materials containing these new monomers generally had physical, thermal, and mechanical properties compared to those of materials containing Bis-GMA or UDMA and had a lower polymerization shrinkage. These advantageous properties make the polymerizable cholic acid derivatives attractive for further evaluations in dental composites. However, suitable biocompatible diluents have to be found for their successful application. Perhaps, the recently synthesized acetylated GDMA (AGDMA, Scheme 31),146 a mixture of 2-acetoxypropylene 1,3-dimethacylate and 1-acetoxypropylene 2,3-dimethacylate, can be used instead of TEGDMA. The water uptake properties, polymerization shrinkage, and modulus of elasticity at body temperature were shown to be much better than those of TEGDMA. However, AGDMA showed a lower degree of conversion and lower glass transition temperature of the formed polymer network, and the cytotoxicity of AGDMA was not investigated. In this context, it is interesting to mention that the biocompatibility of a large number of free-radically polymerizable monomers, such as vinyl carbonates, vinyl carbamates, or phosphorus-containing vinyl esters and vinylcarbamates, was investigated.147,148 These monomers were used in the fabrication of 3D scaffolds by lithography-based additive manufacturing technology for implants, for example, for rebuilding bone structures. Recently, four new dimethacrylates PADMA (Scheme 31) were prepared by the reaction of glycidyl methacrylate with dicarboxylic acid esters obtained from phthalic anhydride and 1,3-propylene, 1,4-butylene, 1,5-pentylene, and 1,6-hexylene glycol.149 The flexural properties and storage moduli were shown to be better than those of the corresponding BisGMA mixture. Finally, two other strategies for the development of new dental resins are the synthesis of monomers, which enable the fabrication of composites with a high content of inorganic fillers and without the use of additional diluents, or of dimethacrylates with relatively high molecular weights. According to the first strategy, a number of O-alkylated BisGMA derivatives Bis-R-GMA (Scheme 31, R = ethyl, propyl, or butyl) were synthesized by etherification of BisGMA with the corresponding alkyl iodides.150 These alkylated Bis-GMA derivatives showed a significantly reduced viscosity, for example, of 0.21 Pa s in the case of the propoxy derivative, which was also lower than the viscosity of a mixture of BisGMA and TEGDMA in the weight ratio of 70:30 (1.86 Pa s). Bis-R-GMA-based composites also showed a lower water uptake and curing shrinkage and the mechanical properties are similar to those of compared composites with the BisGMA/TEGDMA mixture. According to the second strategy, novel bicycloaliphatic dimethacrylates of different chain lengths were prepared by the reaction of glycidyl methacrylate with dicarboxylic acid esters obtained from nadic anhydride and ethylene (NAEDMA, Scheme 31), 1,4-butylene or 1,6-hexamethylene glycols.151 The mechanical properties, polymerization shrinkage, and degree of double bond conversion were shown to be better than those of BisGMA, TEGDMA, and UDMA analog materials.

Scheme 31.

Alternative methacrylate cross-linkers for dental composites.

Because of the numerous requirements for monomers in restorative composites, it takes at least some years to introduce simple new dimethacrylates. More than 10 years were needed for the development of new basic resin cross-linkers, such as ormocer methacrylates or completely new resins such as the cationically polymerizable siloranes. Ultimately, the variety of basic requirements was not met by free-radical ring-opening polymerizable cyclic monomers such as cyclic ketene O, O- or S, S-acetals, 2-vinylcyclopropanes or bicyclic acrylates, which we described in our last review in 2007.11 Therefore, many investigations of the last few years were focused on the optimization of the photocuring process and the composition of conventional dimethacrylate-based restorative composites. The state-of-the-art of dental photocuring of the polymer network formation and property development processes using conventional dimethacrylate formulations was described in two excellent reviews by Rueggeberg152 and Stansbury.153 Thus, new photochemical trends are the development of light sources with improved energy efficiency and more uniform output characteristics. There is the need of light sources, which enable the simultaneous and wide area illumination of teeth, for example, for the photocuring of sealants. Moreover, “curing from within” is a further very attractive future concept, which probably can be realized by the incorporation of quantum dots in photopolymerizable restorative composites.152 In the characterization of dental dimethacrylate polymer networks, mixtures of Bis-GMA and TEGDMA are repeatedly in the focus of investigations and the application of new techniques. Thus, it was shown by positron annihilation lifetime spectroscopy that the addition of TEGDMA to BisGMA decreases the free-volume up to 40% due to increased conversion. Above that concentration, however, free volume pore size increases despite the increase in conversion due to the high concentration of the more flexible TEGDMA.154 In this context, it is interesting to mention that photopolymerized TEGDMA absorbed a higher amount of an ethanol/water solution (50/50 or 75/25 [v/v]) compared to photocured BisGMA, UDMA, or D3MA.155 With regard to the mechanical properties, water solubility, or polymerization shrinkage, BisGMA or TEGDMA was substituted by other commercial acrylates.156 However, the used diacrylates, triacrylates, and tetraacrylates showed a relatively low purity and they are significantly more toxic. How much do resin-based dental materials release? The answer to this question was given in a recent review by Van Meerbeek and coworkers.157 A meta-analytical approach out of an initial set of 71 studies showed that compared to monomers, such as BisGMA, TEGDMA, or UDMA, similar or even higher amounts of additives, such as accelerators, inhibitors, stabilizers, may be eluted, even although composites generally only contain very small amounts of additives.

Resins with Reduced Polymerization Shrinkage Stress

In the last 20 years, the main motivation to change the resin chemistry of restorative composites was to reduce the polymerization shrinkage, which occurs during the curing of the restorative composite in the tooth cavity and may result in the formation of marginal gaps, cusp fractures, microleakage, and marginal staining up to recurrent caries. Except the less successfully introduced cationically polymerizable silorane resin, the new monomer resins used for restorative composites are based to this day on methacrylate chemistry. However, a continuous reduction of the polymerization shrinkage of the composites from about 3–4 vol % in the 90s of the last century to about 1.5–2.5 vol % of current composites was achieved by optimization of the resin and composite composition and mainly by innovative improvements in the filler technology (optimal filler surface treatment, mixture of fillers with adapted particle size distribution, tailor-made fillers, etc.). Moreover, the significant improvements in the adhesive technology during the last two decades have led to the current situation that there is an excellent acceptance of methacrylate-based direct restorative materials.94 Nevertheless, for the further improvement of the marginal quality particularly of large cavities and fillings and for the substitution of the incremental technology by a bulk placement, many current research activities are focused on the reduction of the polymerization shrinkage stress, which is mainly responsible for microleakage formation or debonding. The polymerization stress development has to be considered as an extremely complex multifactorial phenomenon.158 Polymerization stress mainly depends on the volume shrinkage, the elastic modulus, the polymerization rate, and the degree of conversion. The mechanical properties of the dental tissues, the tooth geometry, cavity shape and size, and the curing process are also very important for the build-up of shrinkage stress. Furthermore, an increasing compliance of the cavity walls, for example, by applying an intermediate low-modulus layer may also lead to significant stress relief. In this context, an increase of the viscous flow of the composite is one main material-based key concept to reduce the polymerization stress development. The viscoelastic behavior of a resin composite is influenced by the resin chemistry, the filler content, and degree of conversion. In the initial stage of the polymerization, the composites show a predominantly viscous behavior (pregel phase) and gradually after polymer network formation, they exhibit prevalently elastic properties. Thus, in the pregel phase, the viscous flow may accommodate a relevant fraction of the total polymerization shrinkage. Therefore, in step-growth polymerization resins, such as thiol-ene or epoxy-amine resins, which show a significantly prolonged pregel phase usually at high monomer conversion, a significant part of the volume shrinkage does not produce shrinkage stress. However, after polymer network formation, monomer conversion contributes to the stress build-up. This is more dominant in polymer networks with a high stiffness. The present photopolymerizable composite restoratives based on cross-linking methacrylate resins form a polymer network by a chain-growth polymerization mechanism and are usually characterized by a very short pregel phase. Nevertheless, the polymerization stress development can be influenced by changing the matrix composition. For example, in the case of mixtures of TEGDMA with BisGMA or UDMA, it was shown that it was possible to formulate composites with a relatively low TEGDMA-content and still obtain a high conversion and relatively low polymerization stress compared to BisGMA-based composites.159 In the case of BisGMA/TEGDMA mixtures, the shrinkage stress decreased with decreasing TEGDMA content160 and increasing content of a barium glass filler.161 The latter was explained by the reduction of the polymerization shrinkage by the addition of the filler. Compared to BisGMA, BisEMA resulted in a greater polymerization stress in mixtures with different diluents, despite the lower modulus, which was caused by a higher conversion and polymerization shrinkage.162 Furthermore, it was shown that the glass transition temperature and viscosity of the dimethacrylate monomers, which characterize the molecular dynamics, were in close relationship with the shrinkage stress. The shrinkage stress development increased with the molecular mobility of the resins.163 In another investigation, it was shown that for the reduction of the polymerization stress, a low postgel shrinkage must be combined with a relatively low elastic modulus.164 Thus, the silorane-based composite showing the lowest polymerization shrinkage (1.4 vol %) of the current composite restoratives did not exhibit the lowest polymerization stress because of its relatively high flexural modulus.165 In this context, Watts and coworkers166 demonstrated that correlations of material parameters of light-cured dental composites, for example, the relationship between Knoop microhardness and polymerization shrinkage, can be used to estimate the temporal variations of Young's modulus and viscosity of the composite during curing. In general, the polymerization stress development of photocurable composite restoratives is the result of the complex interplay of the volumetric polymerization shrinkage, the photopolymerization kinetics, and the viscoelastic properties. Therefore, irradiation strategies also strongly influence the build-up of stress, which is summarized in the already mentioned review by Cramer et al.121

The thiol-ene-methacrylate-based resins, delay of gelation of dimethacrylate networks, and the use of photocleavable dimethacrylates are an additional recent concept for the reduction of shrinkage stress in dental composite restoratives. As noted in our last review,11 thiol-ene resins that means mixtures of multifunctional thiols (e.g., pentaerythritol tetrakis(3-mercaptopropionate) PETMP, Scheme 32) with multifunctional allyl compounds (e.g., triallyl-1,3,5-triazine-2,4,6-trione [TATATO]) show many promising properties for dental restorative applications, such as low volume shrinkage and shrinkage stress, a high polymerization rate, high conversion, and no oxygen inhibition. However, the strong odor of commercially available thiols, the limited storage stability of the thiol-ene resins, and the partially insufficient mechanical properties of the formed polymer networks mainly prevented the use of thiol-ene resins as matrix for composite restoratives until today. In recent years, improvements were achieved by the combination of conventional thiol-ene resins and methacrylate resins.167–170 Thiol-allyl ether-methacrylate ternary systems exhibit a reaction mechanism that is a combination of both step- and chain-growth polymerization.171,172 Because of the significantly higher addition rate of thiyl radicals to methacrylates compared to allylethers, the early stage of the polymerization is dominated by the methacrylate homopolymerization and chain transfer, whereas the latter stages are controlled by thiol-ene polymerization. As a result of this hybrid polymerization mechanism, the polymerization shrinkage stress is reduced. Thus, with resins or composites, for example, based on mixtures of BisEMA (n + m ≈ 3, Scheme 24, 70 wt %) with a 1:1 stoichiometric mixture of PETMP/TATATO (30 wt %), a dramatically reduced polymerization shrinkage stress was achieved compared to materials based only on a BisGMA/TEGDMA (70/30 wt %) resin. In addition, the methacrylate-thiol-allyl ether systems exhibited an equivalent flexural modulus and slightly reduced flexural strength compared to the BisGMA/TEGDMA materials. Furthermore, it was found that resins with excess of thiol resulted in a further reduction of the shrinkage stress compared to resins with a 1:1 thiol-ene molar ratio.168 Instead of thiol-ene resins, the photoinitiated polyaddition of multifunctional thiols, such as PETMP, to multifunctional alkynes, for example, the tetrafunctional monomer PETYNE (Scheme 32), is a promising concept to improve the properties of thiol-based materials.173,174 Although the methacrylate-thiol-ene systems show a very low polymerization shrinkage stress, attractive mechanical properties, improved functional group conversion, reduced water sorption, and a better storage stability, the thiol components are used in a relatively high amount and are not odor-free. To reduce the odor, the use of oligomeric polythiol monomers was proposed,175 which can be prepared by the prepolymerization of polyvinyl monomers in the presence of an excess of polythiol compounds. Another possibility to solve the odor problem is to use a smaller content of thiols, for example, only as chain-transfer additive to delay gelation and so to reduce the stress formation in methacrylate networks. Hence, BisEMA was combined with 5, 10, or 20 mol % related to the methacrylate functionality of monothiols, trithiols, or tetrathiols and resulted in a significantly delayed gelation, for example, after the addition of 10 mol % PETMP, the gelation occurred at 12% conversion compared to approximately 5% conversion for the thiol-free control.176 Unfortunately, the thiol addition also resulted in a decrease of the maximum reaction rate and deceleration was observed at later stages in the conversion. A more efficient and promising concept for stress-reduced polymer networks is the incorporation of reversible addition-fragmentation chain-transfer (RAFT) moieties that promote network stress accommodation by molecular rearrangement.177–179 In these covalent adaptable networks (CANs), the bond structure is covalent, yet each individual bond can be broken and reformed. On one hand, CANs were synthesized by photopolymerization of resins on the basis of a stoichiometric mixture of the allyl sulfide 2-methylene-propane-1,3-di(thioethyl vinyl) ether (Scheme 32) and PETMP by adding BisEMA in various ratios.178 On the other hand, CANS were obtained by addition of 0.5, 1.5, or 2 wt % of the RAFT agent trithiocarbonate 2,2′-[thiocarbonylbis(sulfanediyl)bis(2-methylpropanoic acid) (TCSPA, Scheme 32) to a mixture of BisEMA and TEGDMA (70/30 wt %).179 In the latter case, the volumetric shrinkage stress was reduced by more than 50% compared to the pure dimethacrylate system by the addition of as little as 2 wt % of TCSPA, whereas the favorable mechanical properties of the methacrylate-cross-linked networks were not influenced. However, the polymerization rate was reduced for samples containing the RAFT agent TCSPA compared to the pure BisEMA/TEGDMA resin.

Scheme 32.

Components of thiol polyaddition resins and RAFT agents.

The use of photopolymerizable and photocleavable resins is an additional concept for the reduction of shrinkage stress in dental composite restoratives.180,181 According to this approach, stress decreasing resins (SDR™) are used, which contain a photocleavable dimethacrylate, for example, the already mentioned cross-linker UV-UDMA (Scheme 28) and enable a controlled stress release in the postgel stage by a selective photoinduced network cleavage by irradiation with near UV light. Indeed, experimental methacrylate-based flowable composites using the SDR™ technology revealed the lowest shrinkage stress and shrinkage-rate values in comparison to regular methacrylate composites but mediocre micromechanical properties.182 A recent approach183 proposes free-radical polymerizable macrocyclic compounds and compositions, which are characterized by a low polymerization shrinkage and low contraction stress upon polymerization. Thus, a BisGMA-based macrocyclic oligomer MC-BisGMA (Scheme 33) was prepared under pseudo-high-dilution conditions. Unfortunately, no properties were presented in the patent application. Finally, the controlled in situ nanocavitation in polymeric materials is also an interesting new concept to reduce the volume shrinkage and shrinkage stress of dental composites.184 According to this concept, a small amount of a soluble cavitation agent (up to 2 wt %) was introduced in a photopolymerizable material. The cavitation agent decomposes, for example, thermally induced simultaneously with the photo-induced cross-linking, and produce gaseous molecules. The volatile components are trapped in the polymer network, resulting in nanovoids in the polymerized material, and thereby counteract the polymerization shrinkage. Acetone dicarboxylic acid (ADCA), which decarboxylates under mildly elevated temperatures under formation of gaseous CO2 and acetone, was used as cavitation agent. With the addition of only 0.5 wt % ADCA, dimethacrylate-based composites showed a significant reduction in volume shrinkage without a decrease in the mechanical properties.

Scheme 33.

Structure of a BisGMA-based macrocyclic oligomer.

New Photoinitiator Components

Nowadays, photoinitiators for direct composite restoratives have to meet the following basic requirements: the photoinitiators should show a strong visible light absorption mainly in the blue region of the visible light spectrum (400–500 nm), which widely corresponds to the emission spectrum of the current light sources, such as quartz-tungsten-halogen lights and especially the very efficient and cost effective blue light emitting diodes (LEDs). A high photoreactivity, good solubility in and compatibility with the currently used dental resins, sufficient thermal storage stability, and toxicological harmlessness of the photoinitiator and the corresponding photolytic reaction products are also important. Furthermore, the photoinitiator should exhibit an excellent bleaching behavior and not form colored by-products. These demands are widely met by mixtures of CQ with tertiary amines as coinitiators, which are therefore the most frequently used photoinitiator systems in dental materials.29 Nevertheless, in the view of the photocuring of direct composite restoratives, the search for new photoinitiator components is mainly motivated by the elimination of the drawbacks of the CQ-amine systems, by the development of faster curing photoinitiators and by a significant improvement of the curing depth of the composites, allowing a curing of the entire filling without applying increments. The disadvantages of the CQ-amine photoinitiators concern the toxicity of the used amines and the discoloration of the cured materials caused by amine impurities. CQ exhibits a rather low extinction coefficient (ε = 380 dm2 mol−1) at the absorption maximum at λmax = 468 nm and a very low quantum yield ∼0.07 was measured for the photodecomposition of CQ in a dental resin formulation containing 0.7 wt % of CQ and 0.35 wt % of 2-dimethylaminoethyl methacrylate (DMAEMA) as coinitiator.185 Furthermore, in CQ-amine photoinitiators as bimolecular photoinitiating systems, the interaction of the partners is strongly influenced by the viscosity of the medium. Such binary systems also tend to produce a characteristic oxygen-inhibited layer. Finally, as mentioned before regarding acidic self-adhesive composite cements, the acid–base reaction of strongly acidic monomers used in these materials with the amine coinitiator affects the initiating efficiency of the CQ-amine system. Numerous papers deal with CQ-amine system-initiated photopolymerization in dental formulations and we have discussed it in our review.29 Recent papers by Nie and coworkers186–189 concern the use of cyclic acetals such 1,3-benzodioxolane or its derivatives as coinitiator in CQ-induced photopolymerizations. Thus, the natural component sesamin SA (Scheme 34) was used as coinitiator for dental composites. Compared to the CQ-EMBO system (Scheme 8), the final double bond conversion of the photopolymerization of a BisGMA/TEGDMA mixture (70/30 wt %) with the CQ-SA system was only slightly lower.186 In addition, the dimethacrylate N, N-bis(methacryloylethoxycarbonylethyl)-N-(1,3-benzodioxole-5-yl-methyl) amine (DMEBM, Scheme 34), which contains both a coinitiating amine and 1,3-benzodioxole moieties, was synthesized and used to replace both TEGDMA as diluent and the usually nonpolymerizable coinitiator.189 The incorporation of the coinitiator into the polymer network of a composite filling may reduce the possibility of its diffusion out of the cured material into the surrounding tissue to a minimum. Unfortunately, results of the biocompatibility and color stability testing of the polymerizable coinitiator DMEBM have not been published yet. The color stability is an important property for dental composite restoratives as it can significantly impact the long-term esthetic properties of the restorations. Many amine coinitiators are susceptible to yellowing mainly as a result of a photo-oxidation process that can occur after curing of the composite. Although the addition of UV stabilizers can prevent this color formation, it would be beneficial to have color stable amine coinitiators. In this context, simple N-methyl-N-phenyl-3-amino-propionic acid derivatives, for example, ethyl N-methyl-N-phenyl-3-aminoproprionate (MPAP, Scheme 34) were claimed to show an improved color stability.190 The influence of different tertiary amines (1 wt %: N, N-cyanoethylmethylaniline, N, N-dimethyl-p-toluidine, DEPT, EMBO, and DMAEMA) on the color and transmittance stability, degree of conversion, shrinkage strain, and so forth, of BisGMA/TEGDMA resins (3:1 wt) containing 0.25 wt % CQ was recently evaluated by Watts and coworkers.191 Only the resin with DMAEMA did not show dark or yellow shifts after artificial daylight aging. As mentioned before, another method to improve the color stability, bleaching behavior, and depth of cure of restorative composites is the use of mixtures of different photoinitiators, which also helps to prevent discoloration. Thus, polymerizable compositions containing photoinitiator components consisting of only a bisacylphosphine oxide, a α-diketone, and a tertiary amine were recently proposed,192 which hardly show any discoloration even if stored for a long time and exhibit high photocurability both in case of halogen lamp curing and LED light-curing. The photoinitiation behavior of acylphosphine oxide (APO) and bisacylphosphine oxide (BAPO) derivatives in unfilled, light-cured dental resins and composites was intensively investigated by Ikemura and Endo.67,193,194 It was shown that the APO and BAPO photoinitiators exhibit a reactivity comparable to that of the CQ/tertiary amine system. In addition, a novel CQ derivative, 7,7-dimethyl-2,3-dioxobicyclo[2.2.1]heptane-1-carbonyldiphenyl phosphine oxide (CQ-APO, Scheme 35) bearing an acylphosphine oxide group, was synthesized. CQ-APO was obtained as a pale yellow solid and CQ-APO-containing resins showed an excellent shade and a good photopolymerization reactivity. Compositions containing mixtures of a BAPO and an α-diketone as photoinitiator systems can also be used as two-phase light-curing (TPLC) materials.195 Thus, in a first light-curing step, curing light of a wavelength of more than 420 nm is used to activate mainly the CQ-amine systems. In a second step, curing light of a wavelength of less than 420 nm may activate mainly the α-cleavage of BAPO and results in a fully cured material. For cement application, such a TPLC material would enable the easier removal of excess cement in the mouth before the final curing. Additional current examples of the use of mixtures of different photoinitiators concern the application of CQ/PPD mixtures to reduce the stress development without compromising the final properties of dental composites196 and the use of CQ-amine systems in combination with diphenyliodonium hexafluorophosphate (Ph2IPF6).197 For the three-component initiation system (CQ/amine/Ph2IPF6), an enhanced photopolymerization reactivity of dimethacrylates was found. The proposed photoinitiating mechanism involves the reduction of the excited CQ molecule by the amine to form ketyl and aminoalkyl radicals, followed by the irreversible oxidation of the amine, and, to a lesser extent, the ketyl radical by the iodonium salt, to form an initiating radical. The result is a more efficient and faster initiation of the polymerization but also a slower consumption of CQ. The three-component systems showed an approximately fivefold increase in the polymerization rate. Three-component photoinitiator systems including a light-absorbing photosensitizer, an electron donor, and an electron acceptor for visible light-activated free-radical polymerization were also intensively investigated by Stansbury and coworkers.198–202 They used xanthene dyes (e.g., rose bengal and fluorescein), a phenazine dye (methylene blue), a porphin dye (5,10,15,20-tetraphenyl-21H,23H-porphin zinc; Scheme 35), or CQ as photosensitizer, various amines as electron donors, and a diphenyliodonium salt as electron acceptor. Based on selected three-component photoinitiator systems, an extensive dark curing could be achieved, for example, for mixtures of HEMA with 1,6-hexandiol dimethacrylate.202 Such a dark curing is useful for dental applications to reduce the processing time or lower the initiator concentration as well as to achieve a photocuring in shadow regions. However, the usual radical photopolymerization is characterized by a rapid drop of the polymerization rate, when the photocuring light is turn off. In contrast, cationic photopolymerization allows significant dark curing because of the long life span of the active species. Using the three-component photoinitiator systems with a photo-oxidizable photosensitizer and monomers with at least one abstractable hydrogen, a controlled radical polymerization may take place in which the radical species are not terminated and therefore result in an extensive curing in the dark. Finally, it should be mentioned that a promising concept to increase the depth of polymerization is the use of frontal polymerization, which is an unusual method of polymerization converting monomers into polymers by means of a localized reaction zone that propagates by the coupling of thermal diffusion and the heat released by the polymerization.203,204 Frontal polymerization is induced at the surface of a formulation and a polymerization front migrates through the material. For example, under irradiation, photopolymerization is initiated in the upper layer of the material. This exothermal process raises the temperature of the adjacent region of the formulation, where the thermal initiator is cleaved, and new radicals can start the exothermic polymerization, leading to the reaction zone that propagates through the reaction medium as a thermal wave. Unfortunately, the temperatures of the propagating front rise to above 200°C, so an application in dental materials is impossible. We could introduce a new concept of frontal polymerization involving a copolymerization-induced destabilization of (meth)acrylate-based peroxides that allows lowering the front temperatures.205 This concept is based on the fact that with increasing the degree of branching of the substituent next to the carbonyl group of acylperesters, the decomposition temperature significantly drops. Thus, for the use of a peracrylate as a thermal initiator, the relatively high stability guarantees a good storage stability. The actual initiator for the frontal polymerization is the in situ formed copolymer containing a perester unit with an increased degree of branching, which has a lower stability. A propagating front at lower temperatures could therefore be enabled. Nevertheless, the lower front temperatures are still too high for oral curing and further significant progress is necessary to use frontal polymerization for the preparation of sensitive materials.

Scheme 34.

Structure of promising cyclic acetal or amine accelerators for CQ.

Scheme 35.

Structure of CQ-APO and of selected photosensitizers.


In the last 10 years, many new monomers and initiator components were synthesized and evaluated for an application in enamel–dentin adhesives and direct composite restoratives. The new components and the thereon-based adhesives or composites have to meet many different requirements, before they can really be used in dental materials. It also goes without saying that they should contribute to an improvement of the corresponding materials or products. Basically, dental adhesives and composite restoratives have to show good clinical performance and should be safe and easy to apply. New bisacrylamide cross-linkers with improved hydrolytic stability, new strongly acidic adhesive monomers that exhibit both excellent adhesive properties and storage stability, and novel photoinitator systems, which enable fast and efficient curing, in combination with the implementation with novel application devices, have significantly improved the adhesive performance, durability, and handling properties of the current enamel–dentin adhesives. Apart from the less successfully introduced cationically polymerizable silorane resin, the currently used resins for direct composite restoratives are based on methacrylate chemistry to this day. Nevertheless, a continuous improvement of the physico-mechanical and processing properties as well as durability and esthetics of current composites could be achieved using new tailor-made methacrylate cross-linkers, by a few new additives and photoinitiators, and mainly by the introduction of new fillers and the optimization of the composite composition. Compared to adhesives, the application of new components in composites is more difficult and frequently prevented by trivial problems concerning discoloration, odor, or stability. Nowadays, the current dental adhesives and methacrylate-based direct restorative materials have found wide-spread acceptance. Future developments of dental adhesives and direct composite restoratives will be based on different strategies: first, improvements of the durability, biocompatibility, storage stability, and the handling properties. Second, the development of materials with a broader application spectrum, for example, one adhesive for all dental substrates (enamel, dentin, ceramics, metal alloys, and composites) or well performing self-adhesive composite restoratives. Third, smart adhesives or composites, which, for example, allow the control of color or pH value, show bioactive or remineralization properties as well as self-repair properties or enable easy removal.

Biographical Information

original image

Norbert Moszner was born in Jena (Thuringia), Germany in 1953. He studied chemistry at the Friedrich-Schiller-University of Jena from 1972 to 1976, received his Ph.D. from the Friedrich-Schiller-University of Jena in 1980 and his habilitation in Polymer Chemistry from the Friedrich-Schiller-University of Jena in 1987. He became a full professor of Polymer Chemistry at the Technical University of Leuna-Merseburg (Saxony-Anhalt), Germany in 1989. In 1990, he changed to the International Dental Company Ivoclar Vivadent AG in Schaan, Liechtenstein, with worldwide about 2400 employees. He is responsible for the polymer-chemical basic research and the development of new monomers, polymers, and additives for dental applications. He has published more than 132 papers in refereed scientific journals and filed 72 patents. His research activities concern the synthesis and radical polymerization of functionalized, low-shrinking, adhesive or hybrid monomers, ring-opening polymerization, smart materials, nanotechnology, photopolymerization, and composites.

Biographical Information

original image

Thomas Hirt was born in Basel, Switzerland in 1967. He studied chemistry at the ETH in Zürich from 1986 to 1991 and received his Ph.D. from ETH in 1995 for his work with new biocompatible, degradable polymers for medical applications. Thereafter, he joined the central research laboratories of Ciba Geigy in Basel and moved Ciba Vision in Atlanta, USA, after 1 year to where he worked from 1996 to 1999. He then joined SEFAR AG, Printing Division as Head of R&D and New Business Development. From 2008 to 2010, he worked at BioCure (Schweiz) GmbH and BioCure as VP of R&D in the area of Medical Devices. In 2010, he joined Ivoclar Vivadent AG in Schaan, Liechtenstein as Director of R&D/Organic Chemistry. He is responsible for the development of dental products for the clinical use, which include composites, adhesives, luting cements, and prevention products.