Development of an Adhesive System for Bonding to Hard Tooth Tissues R.L. Bowen, D.D.S.*
W.A. Marjenhoff, Ph. D.* m s paperwas presented at the First International Conference on Adhesive Dentistry. Los Angeles, Callfomia, January 11-12,1991 [Palm Springs Dental Seminars. Inc.1.)
The development of an adhesion system for bonding dental composites to dentin and enamel is reviewed. Building on Andings concerning adhesion to enamel, a field pioneered by Dr. Michael Buonocoreand others, Dr. R.L. Bowen and his colleagues at the PaffenbargerResearch Center, National Institute of Standards and Technology,began addressingand solvingproblems associated with ( 1) silicate cements and unfilled resins, (2) bonding in an aqueous environment, and (3)the development of an adhesion system for both dentin and enamel that could withstand various stresses. Although commercial products based on this adhesion system are currently available in the dental materials marketplace, experimentation continues, focusing on the synthesis of potentially better component d o g s , the optimization of the individual components, and on improved storage stability and ease of synthesis.
D
r. MichaelBuonocoreis one of the best known pioneers in adhesive bonding of resins to teeth.’ He found that etching of enamel created a microporous surfaceinto which direct-fBlingliquid resins could flow, polymerize,and make a micromechanicalattachment. He thereby achieved his primary objective of providing a conservative means of sealing developmentalpits and fissures. with the result that innumerable carious lesions have been prevented on the occlusal surfaces of p t e r i o r teeth. He tried many kinds of resins (ageneric term for monomers or their polymers), most of which were not satisfactory,and finally settled on the kind now used in composites. Dr.George Newman,a confemporary of Dr. Buonocore. developed similar methods to bond orthodontic brackets diredlyto the enamel of teeth.2The acid-etch tedmique later was used for a variety of other dental p m c e d u r e ~but . ~ was found to be in&kcUve for the adhesivebonding of restorative materialsto dentin. Yet therearemanyinstancesinwhichitwouldbebeneficial
to bond to dentin, e.g. in restoration of teeth with root caries, fractures involving dentin, and deep coronal caries;in placement of porcelainveneers,inlays,onlays. crowns,and resin-retainedbridges; and in treatment of amelogenesis imperfecta and some cases of enamel hypophsia. Adequate adhesive bonding to both enamel and dentin could prevent marginal leakage, with its resulting problems, provided that polymerization shrinkage of dental composites can be sufRcientlyreduced. Also, the removal of healthy dentin for the mechanical retention of composite restorations, which is painful without an anesthetic, is unnecessary when effective adhesive bonding materials and techniques are successfully employed. Dentin, composed mostly of hydrophylic calcium phosphate crystals and collagen, contains a significant amount of water. Volumetrically, about half of the dentin is composed of hydrated organic material.eThe water content is thought to vary within the dentin, increasingwithproximity to the pulp.’ It is a formidable challenge to find something appropriate that can adequately compete with water for tooth surfaces, particularly dentin.*-” In addition to being adhesive to dentin and enamel in an aqueous environment, a successful bonding system must be able to withstand the stresses of compositeshrinkageduringhardening,1zthe stresses
Bonding to Hard Tooth nssues
accompanyingthermal changes of the filling material,I3 the stresses of mastication, and other stresses." In the early 1950s. there were two kinds of esthetic direct restorative materials available: silicate cements and unfilled methyl methacrylate resins. Both had important shortcomings. Silicate ~ e m e n t swere ' ~ subject to acidic degradation16 and were useful for an average of approximately 4 to 5 years before requiring replacement." Problems inherent to the unfilled resins, based on methyl methacrylate, were profound shrinkage during hardening, insufficient stiffness. and an excessive coefficientof thermal expansion compared to tooth structure.'8-20 Epoxy resins, which were relatively new in the early '50s.had some intriguing characteristics that made them candidates for use in restorative dentistry. They were known to harden at room temperature with little shrinkage and to produce an insoluble polymer with rather good physical properties. They were also adhesive to most solid surfaces, including the kind of inorganic materials that might be used forreinforcingfillers. Preliminaryexperimentswith an aggregate of porcelain or fused quartz particles bound by an epoxy resin were encouraging.I0 But epoxies were not very adhesive to tooth surfaces after immersion in water, and the aminecured epoxies did not harden rapidly enough for direct clinicaltechniques. The quick-hardeningepoxies,which used acidic catalysts for their polymerization,were not effective because tooth surfaces buffered the acid and yielded incomplete polymerization. Research to find curing agents with epoxy resins that would give quick hardening necessary for clinical use as a direct W g material proved fruitless. However, a monomer was conceived and synthesized that resembled an epoxy but had methacrylategroups instead of epoxy groups.21Methacrylate ester groups polymerizerapidly under oral conditions by a free radical chain reaction. This monomer, the reaction product of the diglycidylether of bispenolA and methacrylic acid, was called "BIS-GMA"and became the hardeningbinder for a variety of inorganic fillers22 of Merent sizes and distributions29and having varying degrees of radiopacity24-2Bthat collectivelycomprise the classof restorative materials known as dental composites. They replaced silicate cements and unfilled methyl methacrylate resins as dentists' direct tooth-colored restorative material of choice. Although an improvement over the materials they replaced, dental composites have some troublesome properties. Compositeswith a higher proportion of filler and less resin tend to have better physical properties; but, even with a maximum proportion of inorganic filler particles, composites exhibit some hardening shrinkage, a stiilhess lower than that of the tooth, and a CoefXicient of thermal expamion higher thanthat of the tooth crown. In spite of these problems, there has been some success in solving the problem of bonding dental composite resins to dentin and e ~ m e l . ~BISGMA -~' resins have no specificchemicalgroups or mechanisms
for competing with water that interacts strongly with tooth structures. Therefore, chemical coupling agents (surface-activecomonomers) are needed. These have groups that can compete with water for the tooth surfaces, and also groups that can interact by copolymerization with dental resins such as BIS-GMA. Surface activity tests38 showed that there were groups that could competewith water for dentin or enamel surfaces. Many attempts have been made to synthesize different coupling agents for tooth surfaces. One of the earliest successful compounds tested was NPG-GMA, the reaction product of N-phenylglycine and glycidyl methac~ylate.~~ The use of this surface-active comonomer alone improved the water-resistant bonding between resins and enamel and dentin to a degree that was statistically but not clinically ~igniflcant.~O-~~ Removal of the structurally weak smear layer, pellicle, or other superficial layers of the tooth surface by the use of acidic or chelating agents might reduce the availability of calcium ions for interaction with a chelating surface-active comonomer like NPG-GMA or other multiple-bondingcoupling agent. To supplement calcium ion sites for improved bonding, certain appropriate metal cations were evaluated for use on tooth surfaces. Experiments indicated that the most effective agent might be ferric oxalate. primarily because of the iron ion's high tendency to be bound strongly by dentin and enamel and its high chelate stabilityconstants with molecules having ligand groups similarto those of "GGm4548 Furthermore, the oxalatewould form an insoluble precipitate with calcium ions,which, together with insoluble ferric phosphate. would seal the dentinal tubules to provide pulp protection and desensitization. It was then discovered that the additional use of a relatimly hydrophylic monomer containing two free carboxylgroups in addition to two polymerizablegroups on each molecule dramaticallyimprovedbond strengths to levels of clinical sign.iflcance.m This monomer was called PMDM (the reaction products of pyromellitic dianhydrideand hydroxyethylmethacrylate).Therewas a synergisticinteractionbetween the NPG-GMA and the PMDM.S2 The original adhesive system developed was a sequential application of aqueous acidic ferric oxalate, followed by an acetone solution of NFG-GMA or NTGGMA(thereaction product ofN-p-tolylglycineand glycidyl methacrylate), and then an acetone solution of PMDM. This systemwas effective only if placed in the described sequential order utilizing all three components. The acidic ferric oxalate solution was removing the orlginal smear layer, the disturbed surface layer caused by mechanical abrasion in preparing a restoration site,% and laying down a layer of precipitation product that was plugging up the lumen of dentin tubules.The latter function signLficantly reduced tooth sensitivity to the subsequent procedure. The NTG-GMAwas necessary to induce polymerization of the PMDM, but the exact mechanism(s) of this free radical initiation is still not clear. During subsequent experhentation. it was dis87
JOUWAL OF ESTliETIC DESI'ISTRY VOLUME 3. NUMBER 3 M a y / J u n e 1991
subsequent oxygen exposure. These protected solutions were stable under normal storage conditions. Two commercial products are based on this two-solution system. Current experimentation with the system is focusing on optimizingthe individual components. Nitric acid concentrations will continue to be refined to yield optimal treatment of both dentin and enamel. Different analogs of the NPG molecule are being synthesized toward improving effectiveness, storage stability, and ease of synthesis.55 PMDM is being investigated to isolate more effective linking agents between tooth surfaces and the overlying restorative resins.56 In additionto the ADA Health Foundation's system(s) being developed by research associates at the National Institute of Standards and Technology, there are, of course, a number of other methods and different materials being used now for dental adhesive bonding; most of them have certain features in common.57One of the problems with current systems for bonding with dentin is technique sensitivity, which probably accounts for the disparity between reports of the relative merits of different systems. Researchers involved in developing a given technique and those experienced in using it probably unconsciously utilize procedural details or measures that are not described explicitly and in sufficient detail in the methods for use for others to match their results exactly. Consequently, other researchers using nominally the same materials and protocols may get poorer results because of the details that are left out of the descriptions for use. That is probably one reason why the ranking of bond strengths differs between laboratories. Another more obvious reason is differences in test methods, which have not yet been standardized. Although the clinicalperformanceof products based on the adhesive bonding system described here, as well as others that have found their way to the dental materials marketplace, are worthy of use, they are still not as strong, dependable, and technique insensitive as acid-etched bonding to enamel. Therefore, continued extensive work and studies on adhesive bonding to dentin are stronglyjustified. There is good reason to be optimistic that with continued improvement of the tools of research, the developmentof more clinically satisfactory adhesive bonding to dentin probably will be forthcoming before the beginning of the 2 1st century.
covered that the smear-removing capabilities of ferric oxalate were due primarily to the presence of small amounts of nitric acid left over from the synthesis of the oxalate.% Controlled additions of nitric acid to the aqueous oxalate solution were done to determine the optimum acid concentration for this s o l ~ t i o n .A~small ' increase in the concentration of nitric acid to about 2.5 percent HNO, by weight also improved the simultaneous etching of instrumented enamel. A side effect of the application of the ferric oxalate solution was discovered: the occasional appearance of black staining at the adhesive interface in early animal trials.52This could be reproduced in the laboratory by applying a sodium sulfide solution to femc oxalatetreated dentin. The cause of this staining in uiuo is probably (although not proven to be) the reduction of ferric to ferrous ions by sulfide-forming anaerobic microorganisms, resulting in the formation of black ferrous sulfide pigments. To eliminate this, acidic aluminum oxalate was substituted, and it produced no staining on dentin.53 Aqueous solutions of aluminum oxalate and nitric acid were then applied in animal trials with no staining problems." There was some in uitro evidence that aluminum oxalate did not produce as much of the reaction products plugging the dentin tubules as had ferric oxalate. This technology was transferred from the laboratory to the dental materials marketplace in 1986, when a product incorporating aluminum oxalate and based on the patents assigned to the ADA Health Foundation became commercially available. In the continuing laboratory research, it was found that the aluminum oxalate could be eliminated entirely from the experimental system without loss in adhesion, providing that the dilute nitric acid solution was retained. None of the other acids, evaluated in a wide range of concentrations. were as good or better than the dilute nitric acid (which should be distinguished from concentrated nitric acid, a strong oxidizing agent). It was then surprisingly discovered that NPG (N-phenyl glycine) could be substituted from NPG-GMA or "EGMA. The experimental system was then reduced to the three components of (1) dilute nitric acid, (2)an NPGacetone solution, and (3)a PMDM-acetone solution. The three components still had to be applled individually in sequence to achieve adhesion, and efforts were concentrated on ways to simplify application. It was then suspected and verified that NPG would be soluble in the dilute aqueous nitric acid solution. This permitted a simplification of the procedure to the application of two solutions: (1) an acidic NFG solution and (2) a PMDM-acetone solution. However. preparation and storage of the Arst solution was ditricult because of the reactivity of the NPG molecule to atmospheric oxygen. Storage times were very short if the add-NPG solution was exposed to air. If used shortly afler mbdng. adhesion was effective. Methods were then developed for preparingthe solution under an inert atmosphere and protecting it from
REFERENCES 1. Buonocore MG. A simple method of increasing the adhesion of acrylic fllling materials to enamel surfaces. J Dent Res 1955:W. 849-853. 2. Newman GV,SnyderWH. WilsonCJr. Acrylic adhesives for bondingattachments to tooth surfaces.Angle Orthod 1968; 3812. 3. Laswell HR, Welk DA, Regenos J W . Attachment of resin restorations to acid pretreated enamel. J Am Dent Assoc 197 1;82:558-563.
aa
Bonding to Hard Tooth Tissues
4. Doyle WA. Operative dentistry. In: Goldman HM. Forrest
25. Bowen RL, Cleek CW. A new seriesofX-ray-opaquereinforcing fillers for composite materials. J Dent Res 1972:51: 177-182. 26. Bowen RL. Reed LE. Semiporous reinforcing fillers for com-
SP, Byrd DL, McDonald RE. eds. Current therapy in dentistry, Vol. 111. St. Louis: CV Mosby, 1968: 823-849. 5. Doyle WA. Acid etching in pedodontics. Dent Clin North1973: 17: 93-104. 6. Erickson RL. Adhesive dental materials. In: OkabeT, Taka-
27.
shashi s. eds. Transactions. International Congress on Dental Materials. Joint Meeting of The Academy of Dental Materials and The Japanese Society for Dental Materials and Devices, Honolulu, HI: November 1-4, 1989: 55-69. 7. Pashley DH. Interactions of Dental Materials with Dentin. In: Setcos JC, ed. Transactions, Academy of Dental Materials, Proceedings of Conference on Enamel-Dentin-PulpBone-Periodontal Tissue Interactions with Dental Materials. Chicago, IL. February 9. 1990: 55-73. 8. Swartz ML. Phillips RW. A method of measuring the adhesive characteristics of dental cement. J Am Dent h s o c
28. 29.
30.
31.
1955;50: 172. 9. Schouboe PJ, Paffenbarger GC. Sweeney WT. Resin ce-
10. 11.
12.
13. 14.
15.
16. 17.
32.
ments and posterior-type direct filling resins. J Am Dent Assoc 1956;52:584. Bowen RL. Use of epoxy resins in restorative materials. J Dent Res 1956:35:360-369. Buonocore M, Wileman W, Brudevold F. A report of a resin composition capable of bonding to human dentin surfaces. J Dent Res 1956;35:846. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. VI. Forces developing in direct filling materials during hardening. JAm Dent Assoc 1967:74:439-445. Turner PS. Thermal-expansion stresses in reinforced plastics. J Res NBS 1946:37:239. Smith DL, SchoonoverIC. Direct filling resins: dimensional changes resulttng h m polymerization shrinkage and water sorption.J Am Dent Assoc 195346:540. Paffenbarger GC, SchoonoverIC, Souder W. Dental silicate cements: physical and chemical properties and a specification. J Am Dent Assoc 1938;25:32-87. Henschel CJ. Observations concerning in vivo disintegrationofsillcatecementrestorations.J Dent Res 1949:28:528. Bowen RL, Paffenbarger GC, Mullineaux AL. A laboratory and clinical comparison of silicate cements and a directflllingresin:aprogressreport. JProsthetDent 1968:20426-
33.
34.
35.
36.
37.
38.
437. 18. v l m a n SC.Peyton FA. Acrylics and other synthetic resins used in dentistry. Philadelphia: J B Uppincott, 1946. 19. Paffenbarger GC, Nelsen RJ. Sweeney WT. Direct and in-
39.
direct filIing resins: a review of some physical and chemical properties. J Am Dent h s o c 1953:4751&524. 20. Coy HD. Directresinfillings.JAmDentAssoc 1953:47:532-
40.
537. 21. U.S.Patent3.066,112:DentalFillingMaterials Comprising Vinyl Silane Treated Fused Silica and A Binder Consisting
41.
of the Reaction Product of Bis Phenol and GlyddylAcrylate. 1962. 42.
22. Bowen RL,Rodriguez MS. Tensile strength and modulus of elasticity of tooth structure and several restorative materials. J Am Dent Assoc 1962;64:378. 23. Bowen RL. Effect of particle shape and size distribution in a reinforced polymer. J Am Dent Assoc 1964;69:481-495. 24. Chandler HH, Bowen RL. Paffenbarger GC,MullineauxAL. Clinical investigation of radiopaque composite restorative materials. J Am Dent Assoc 1971;50:935-940.
43.
posite resins. 1. Preparation of provisional glass formulations. J Dent Res 1976;55:738-747. Bowen RL, Reed LE. Semiporous reinforcing fillers for composite resins. 11. Heat treatment and etching characteristics. J Dent Res 1976;55:748-756. U.S. Patent 4,215,033: Composite Dental Materials. 1980. Bowen RL, Cobb EN, Rapson JE. Adhesive bonding ofvarious materials to hard tooth tissues. X X V . Improvement in bond strength to dentin. J Dent Res 1982;61:1070-1076. Bowen RL. Cobb EN, Setz LE. Adhesive bonding of various materials to hard tooth tissues. XXVII. Adhesive bonding to dentin and enamel. Dentistry 82 1982;2: 11-13. Bowen RL. Cobb EN. A method for bonding to dentin and enamel. J Am Dent Assoc 1983;107:734-736. Bowen RL, Cobb EN, Misra DN. Adhesive bonding of various materials to hard tooth tissues. XXVIlLAdhesivebonding by surface initiation of polymerization. Ind Eng Chem Prod Res Dev 1984:23:78-8 1. Bowen RL. Cobb EN. Setz LE. Recently developed concepts in adhesive bonding of composites to dentin and enamel. In: Proceedings of the James A. English Symposium, State University of New York at Buffalo. Buffalo Dent Rev 1984;1:10-12. Bowen RL, Eick J D , Henderson DA. Anderson DW. Adhesivebondingofvariousmaterials to hard tooth tissues.XXIX. Smear layer: removal and bonding considerations. Oper Dent Suppl 1984:3:30-34. US. Patent 4,521,550: Method for ObtainingStrongAdhesive Bonding of Composites to Dentin. Enamel and Other Substrates, 1985. U.S. Patent 4,588.576: Multi-step Method for Obtaining Strong Adhesive Bonding of Composites to Dentin. Enamel and Other Substrates, 1986. U.S. Patent 4,659,751: Simplified Method for Obtaining Strong Adhesive Bonding of Composites to Dentin, Enamel and Other Substrates, 1987. Bowen RL. Investigation of the surfaces of hard tooth tissues by a surface activity test. In Phillips RW. Ryge G . eds.: proceedings.Workshop on adhesiverestorative dental materials at Indiana University. Sept. 28-29. 1961. Spencer, Indiana: Owen Utho Senrice. 1961: 177-191. U.S. Patent 3,179,623: Method of Preparing a Monomer Having Phenoxy and Methacrylate Groups Linked by Hy droxy Glycerol Groups, 1965. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. 11. Bonding to dentin promoted by a surfaceactive comonomer. J Dent Res 1965:44:895-902. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. 111. Bonding to dentin improved by pretreatment and the use of a surface-active comonomer. J Dent Res 1965:M 903-905. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. IV. Bonding to dentin, enamel, and fluorapatite improved by the use of a surface-active comonomer. J Dent Res 1965:44:906-911. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. V. The effect of a surface-active comonomer on adhesion to diversesubstrates. J Dent Res 196544:13691373.
89
JOURNAL OF ESTHEnC DEl\mSTRY VOLUME 3. NUMBER 3 May/June 1991
50. Cobb EN, Blosser RL, Bowen RL. Johnston AD. Ferric oxalate with nitric acid a s a conditioner in an adhesive bonding system. J Adhesion 1989:28:41-49. 51. Blosser RL. Bowen RL. Effects of purified ferric oxalate/ nitric acid solutions as a pretreatment for the NTG-GMA and PMDM bonding system. Dent Mater 1988:4:225-231. 52. Stanley HR. Bowen R L, Cob b EN. Pulp responses to a dentin and enamel adhesive bonding procedure. Oper Dent 1988:13: 107-113. 53. Bowen RL, Tung MS, Blosser RL, Asmussen E. Dentine and enamel bonding agents. Int Dent J 1987:37:158-161. 54. Blosser RL, Rupp NW.Stanley HR, Bowen RL. Pulpal and micro-organismresponses to two experimentaldental bonding systems. Dent Mater 1989:5:140-144. 55. Johnston AD, Asmussen E. Bowen RL. Substitutes for NPhenylglycine in adhesive bonding to dentin. J Dent Res 1989:68: 1337-1344. 56. Johnston AD, Farahani M. Bowen RL. Effect of catalyst on restorative dental material synthesis. J Dent Res 1990: 69:233 (Abstr 995). 57. Bowen RL. Eichmiller FC, MarjenhoffWA, Rupp NW. Adhesive bonding of composites. J A m Coll Dent 1989:56: 10-13.
44. Bowen RL.Adhesive bonding of various materials to hard tooth tissues. VII. Metal salts as mordants for coupling agents. In: Moskowitz HD. Ward GT. Woolridge ED, eds. Dental adhesive materials. Proceedings for a symposium held November 8-9. 1973at the Hunter-Bellewe School of Nursing. New York: Prestige Graphic Services. 1974:205221. 45. Misra DN. Bowen RL. Wallace BM. Adhesive bonding of various materials to hard tooth tissues. VIII. Nickel and copper ions on hydroxyapatite: role of ion exchange and surface nucleation. J Colloid Interface Sci 1975:51:3643. 46. Misra DN. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. X. Initial rates of adsorption of nickel or copper ions on hydroxyapatite surface. J Biomed Mater Res 1978:12:505-515. 47. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. X N . Enamel mordant selection assisted by ESCA KPS). J Dent Res 1978:57:551-556. 48. Bowen RL. McClendon LT. Gills TE. Adhesive bonding of various materials to hard tooth tissues. XV. Neutron acti vation analysis of dentin sorption of mordant salts.J Dent Res 1978;57 255-260. 49. Chen RS,Bowen RL. The use of N-phenylglycinein a dental adhesive system. J Adhesion Sd Techno1 1989;3:49-54.
90
W.A. Marjenhoff, Ph. D.* m s paperwas presented at the First International Conference on Adhesive Dentistry. Los Angeles, Callfomia, January 11-12,1991 [Palm Springs Dental Seminars. Inc.1.)
The development of an adhesion system for bonding dental composites to dentin and enamel is reviewed. Building on Andings concerning adhesion to enamel, a field pioneered by Dr. Michael Buonocoreand others, Dr. R.L. Bowen and his colleagues at the PaffenbargerResearch Center, National Institute of Standards and Technology,began addressingand solvingproblems associated with ( 1) silicate cements and unfilled resins, (2) bonding in an aqueous environment, and (3)the development of an adhesion system for both dentin and enamel that could withstand various stresses. Although commercial products based on this adhesion system are currently available in the dental materials marketplace, experimentation continues, focusing on the synthesis of potentially better component d o g s , the optimization of the individual components, and on improved storage stability and ease of synthesis.
D
r. MichaelBuonocoreis one of the best known pioneers in adhesive bonding of resins to teeth.’ He found that etching of enamel created a microporous surfaceinto which direct-fBlingliquid resins could flow, polymerize,and make a micromechanicalattachment. He thereby achieved his primary objective of providing a conservative means of sealing developmentalpits and fissures. with the result that innumerable carious lesions have been prevented on the occlusal surfaces of p t e r i o r teeth. He tried many kinds of resins (ageneric term for monomers or their polymers), most of which were not satisfactory,and finally settled on the kind now used in composites. Dr.George Newman,a confemporary of Dr. Buonocore. developed similar methods to bond orthodontic brackets diredlyto the enamel of teeth.2The acid-etch tedmique later was used for a variety of other dental p m c e d u r e ~but . ~ was found to be in&kcUve for the adhesivebonding of restorative materialsto dentin. Yet therearemanyinstancesinwhichitwouldbebeneficial
to bond to dentin, e.g. in restoration of teeth with root caries, fractures involving dentin, and deep coronal caries;in placement of porcelainveneers,inlays,onlays. crowns,and resin-retainedbridges; and in treatment of amelogenesis imperfecta and some cases of enamel hypophsia. Adequate adhesive bonding to both enamel and dentin could prevent marginal leakage, with its resulting problems, provided that polymerization shrinkage of dental composites can be sufRcientlyreduced. Also, the removal of healthy dentin for the mechanical retention of composite restorations, which is painful without an anesthetic, is unnecessary when effective adhesive bonding materials and techniques are successfully employed. Dentin, composed mostly of hydrophylic calcium phosphate crystals and collagen, contains a significant amount of water. Volumetrically, about half of the dentin is composed of hydrated organic material.eThe water content is thought to vary within the dentin, increasingwithproximity to the pulp.’ It is a formidable challenge to find something appropriate that can adequately compete with water for tooth surfaces, particularly dentin.*-” In addition to being adhesive to dentin and enamel in an aqueous environment, a successful bonding system must be able to withstand the stresses of compositeshrinkageduringhardening,1zthe stresses
Bonding to Hard Tooth nssues
accompanyingthermal changes of the filling material,I3 the stresses of mastication, and other stresses." In the early 1950s. there were two kinds of esthetic direct restorative materials available: silicate cements and unfilled methyl methacrylate resins. Both had important shortcomings. Silicate ~ e m e n t swere ' ~ subject to acidic degradation16 and were useful for an average of approximately 4 to 5 years before requiring replacement." Problems inherent to the unfilled resins, based on methyl methacrylate, were profound shrinkage during hardening, insufficient stiffness. and an excessive coefficientof thermal expansion compared to tooth structure.'8-20 Epoxy resins, which were relatively new in the early '50s.had some intriguing characteristics that made them candidates for use in restorative dentistry. They were known to harden at room temperature with little shrinkage and to produce an insoluble polymer with rather good physical properties. They were also adhesive to most solid surfaces, including the kind of inorganic materials that might be used forreinforcingfillers. Preliminaryexperimentswith an aggregate of porcelain or fused quartz particles bound by an epoxy resin were encouraging.I0 But epoxies were not very adhesive to tooth surfaces after immersion in water, and the aminecured epoxies did not harden rapidly enough for direct clinicaltechniques. The quick-hardeningepoxies,which used acidic catalysts for their polymerization,were not effective because tooth surfaces buffered the acid and yielded incomplete polymerization. Research to find curing agents with epoxy resins that would give quick hardening necessary for clinical use as a direct W g material proved fruitless. However, a monomer was conceived and synthesized that resembled an epoxy but had methacrylategroups instead of epoxy groups.21Methacrylate ester groups polymerizerapidly under oral conditions by a free radical chain reaction. This monomer, the reaction product of the diglycidylether of bispenolA and methacrylic acid, was called "BIS-GMA"and became the hardeningbinder for a variety of inorganic fillers22 of Merent sizes and distributions29and having varying degrees of radiopacity24-2Bthat collectivelycomprise the classof restorative materials known as dental composites. They replaced silicate cements and unfilled methyl methacrylate resins as dentists' direct tooth-colored restorative material of choice. Although an improvement over the materials they replaced, dental composites have some troublesome properties. Compositeswith a higher proportion of filler and less resin tend to have better physical properties; but, even with a maximum proportion of inorganic filler particles, composites exhibit some hardening shrinkage, a stiilhess lower than that of the tooth, and a CoefXicient of thermal expamion higher thanthat of the tooth crown. In spite of these problems, there has been some success in solving the problem of bonding dental composite resins to dentin and e ~ m e l . ~BISGMA -~' resins have no specificchemicalgroups or mechanisms
for competing with water that interacts strongly with tooth structures. Therefore, chemical coupling agents (surface-activecomonomers) are needed. These have groups that can compete with water for the tooth surfaces, and also groups that can interact by copolymerization with dental resins such as BIS-GMA. Surface activity tests38 showed that there were groups that could competewith water for dentin or enamel surfaces. Many attempts have been made to synthesize different coupling agents for tooth surfaces. One of the earliest successful compounds tested was NPG-GMA, the reaction product of N-phenylglycine and glycidyl methac~ylate.~~ The use of this surface-active comonomer alone improved the water-resistant bonding between resins and enamel and dentin to a degree that was statistically but not clinically ~igniflcant.~O-~~ Removal of the structurally weak smear layer, pellicle, or other superficial layers of the tooth surface by the use of acidic or chelating agents might reduce the availability of calcium ions for interaction with a chelating surface-active comonomer like NPG-GMA or other multiple-bondingcoupling agent. To supplement calcium ion sites for improved bonding, certain appropriate metal cations were evaluated for use on tooth surfaces. Experiments indicated that the most effective agent might be ferric oxalate. primarily because of the iron ion's high tendency to be bound strongly by dentin and enamel and its high chelate stabilityconstants with molecules having ligand groups similarto those of "GGm4548 Furthermore, the oxalatewould form an insoluble precipitate with calcium ions,which, together with insoluble ferric phosphate. would seal the dentinal tubules to provide pulp protection and desensitization. It was then discovered that the additional use of a relatimly hydrophylic monomer containing two free carboxylgroups in addition to two polymerizablegroups on each molecule dramaticallyimprovedbond strengths to levels of clinical sign.iflcance.m This monomer was called PMDM (the reaction products of pyromellitic dianhydrideand hydroxyethylmethacrylate).Therewas a synergisticinteractionbetween the NPG-GMA and the PMDM.S2 The original adhesive system developed was a sequential application of aqueous acidic ferric oxalate, followed by an acetone solution of NFG-GMA or NTGGMA(thereaction product ofN-p-tolylglycineand glycidyl methacrylate), and then an acetone solution of PMDM. This systemwas effective only if placed in the described sequential order utilizing all three components. The acidic ferric oxalate solution was removing the orlginal smear layer, the disturbed surface layer caused by mechanical abrasion in preparing a restoration site,% and laying down a layer of precipitation product that was plugging up the lumen of dentin tubules.The latter function signLficantly reduced tooth sensitivity to the subsequent procedure. The NTG-GMAwas necessary to induce polymerization of the PMDM, but the exact mechanism(s) of this free radical initiation is still not clear. During subsequent experhentation. it was dis87
JOUWAL OF ESTliETIC DESI'ISTRY VOLUME 3. NUMBER 3 M a y / J u n e 1991
subsequent oxygen exposure. These protected solutions were stable under normal storage conditions. Two commercial products are based on this two-solution system. Current experimentation with the system is focusing on optimizingthe individual components. Nitric acid concentrations will continue to be refined to yield optimal treatment of both dentin and enamel. Different analogs of the NPG molecule are being synthesized toward improving effectiveness, storage stability, and ease of synthesis.55 PMDM is being investigated to isolate more effective linking agents between tooth surfaces and the overlying restorative resins.56 In additionto the ADA Health Foundation's system(s) being developed by research associates at the National Institute of Standards and Technology, there are, of course, a number of other methods and different materials being used now for dental adhesive bonding; most of them have certain features in common.57One of the problems with current systems for bonding with dentin is technique sensitivity, which probably accounts for the disparity between reports of the relative merits of different systems. Researchers involved in developing a given technique and those experienced in using it probably unconsciously utilize procedural details or measures that are not described explicitly and in sufficient detail in the methods for use for others to match their results exactly. Consequently, other researchers using nominally the same materials and protocols may get poorer results because of the details that are left out of the descriptions for use. That is probably one reason why the ranking of bond strengths differs between laboratories. Another more obvious reason is differences in test methods, which have not yet been standardized. Although the clinicalperformanceof products based on the adhesive bonding system described here, as well as others that have found their way to the dental materials marketplace, are worthy of use, they are still not as strong, dependable, and technique insensitive as acid-etched bonding to enamel. Therefore, continued extensive work and studies on adhesive bonding to dentin are stronglyjustified. There is good reason to be optimistic that with continued improvement of the tools of research, the developmentof more clinically satisfactory adhesive bonding to dentin probably will be forthcoming before the beginning of the 2 1st century.
covered that the smear-removing capabilities of ferric oxalate were due primarily to the presence of small amounts of nitric acid left over from the synthesis of the oxalate.% Controlled additions of nitric acid to the aqueous oxalate solution were done to determine the optimum acid concentration for this s o l ~ t i o n .A~small ' increase in the concentration of nitric acid to about 2.5 percent HNO, by weight also improved the simultaneous etching of instrumented enamel. A side effect of the application of the ferric oxalate solution was discovered: the occasional appearance of black staining at the adhesive interface in early animal trials.52This could be reproduced in the laboratory by applying a sodium sulfide solution to femc oxalatetreated dentin. The cause of this staining in uiuo is probably (although not proven to be) the reduction of ferric to ferrous ions by sulfide-forming anaerobic microorganisms, resulting in the formation of black ferrous sulfide pigments. To eliminate this, acidic aluminum oxalate was substituted, and it produced no staining on dentin.53 Aqueous solutions of aluminum oxalate and nitric acid were then applied in animal trials with no staining problems." There was some in uitro evidence that aluminum oxalate did not produce as much of the reaction products plugging the dentin tubules as had ferric oxalate. This technology was transferred from the laboratory to the dental materials marketplace in 1986, when a product incorporating aluminum oxalate and based on the patents assigned to the ADA Health Foundation became commercially available. In the continuing laboratory research, it was found that the aluminum oxalate could be eliminated entirely from the experimental system without loss in adhesion, providing that the dilute nitric acid solution was retained. None of the other acids, evaluated in a wide range of concentrations. were as good or better than the dilute nitric acid (which should be distinguished from concentrated nitric acid, a strong oxidizing agent). It was then surprisingly discovered that NPG (N-phenyl glycine) could be substituted from NPG-GMA or "EGMA. The experimental system was then reduced to the three components of (1) dilute nitric acid, (2)an NPGacetone solution, and (3)a PMDM-acetone solution. The three components still had to be applled individually in sequence to achieve adhesion, and efforts were concentrated on ways to simplify application. It was then suspected and verified that NPG would be soluble in the dilute aqueous nitric acid solution. This permitted a simplification of the procedure to the application of two solutions: (1) an acidic NFG solution and (2) a PMDM-acetone solution. However. preparation and storage of the Arst solution was ditricult because of the reactivity of the NPG molecule to atmospheric oxygen. Storage times were very short if the add-NPG solution was exposed to air. If used shortly afler mbdng. adhesion was effective. Methods were then developed for preparingthe solution under an inert atmosphere and protecting it from
REFERENCES 1. Buonocore MG. A simple method of increasing the adhesion of acrylic fllling materials to enamel surfaces. J Dent Res 1955:W. 849-853. 2. Newman GV,SnyderWH. WilsonCJr. Acrylic adhesives for bondingattachments to tooth surfaces.Angle Orthod 1968; 3812. 3. Laswell HR, Welk DA, Regenos J W . Attachment of resin restorations to acid pretreated enamel. J Am Dent Assoc 197 1;82:558-563.
aa
Bonding to Hard Tooth Tissues
4. Doyle WA. Operative dentistry. In: Goldman HM. Forrest
25. Bowen RL, Cleek CW. A new seriesofX-ray-opaquereinforcing fillers for composite materials. J Dent Res 1972:51: 177-182. 26. Bowen RL. Reed LE. Semiporous reinforcing fillers for com-
SP, Byrd DL, McDonald RE. eds. Current therapy in dentistry, Vol. 111. St. Louis: CV Mosby, 1968: 823-849. 5. Doyle WA. Acid etching in pedodontics. Dent Clin North1973: 17: 93-104. 6. Erickson RL. Adhesive dental materials. In: OkabeT, Taka-
27.
shashi s. eds. Transactions. International Congress on Dental Materials. Joint Meeting of The Academy of Dental Materials and The Japanese Society for Dental Materials and Devices, Honolulu, HI: November 1-4, 1989: 55-69. 7. Pashley DH. Interactions of Dental Materials with Dentin. In: Setcos JC, ed. Transactions, Academy of Dental Materials, Proceedings of Conference on Enamel-Dentin-PulpBone-Periodontal Tissue Interactions with Dental Materials. Chicago, IL. February 9. 1990: 55-73. 8. Swartz ML. Phillips RW. A method of measuring the adhesive characteristics of dental cement. J Am Dent h s o c
28. 29.
30.
31.
1955;50: 172. 9. Schouboe PJ, Paffenbarger GC. Sweeney WT. Resin ce-
10. 11.
12.
13. 14.
15.
16. 17.
32.
ments and posterior-type direct filling resins. J Am Dent Assoc 1956;52:584. Bowen RL. Use of epoxy resins in restorative materials. J Dent Res 1956:35:360-369. Buonocore M, Wileman W, Brudevold F. A report of a resin composition capable of bonding to human dentin surfaces. J Dent Res 1956;35:846. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. VI. Forces developing in direct filling materials during hardening. JAm Dent Assoc 1967:74:439-445. Turner PS. Thermal-expansion stresses in reinforced plastics. J Res NBS 1946:37:239. Smith DL, SchoonoverIC. Direct filling resins: dimensional changes resulttng h m polymerization shrinkage and water sorption.J Am Dent Assoc 195346:540. Paffenbarger GC, SchoonoverIC, Souder W. Dental silicate cements: physical and chemical properties and a specification. J Am Dent Assoc 1938;25:32-87. Henschel CJ. Observations concerning in vivo disintegrationofsillcatecementrestorations.J Dent Res 1949:28:528. Bowen RL, Paffenbarger GC, Mullineaux AL. A laboratory and clinical comparison of silicate cements and a directflllingresin:aprogressreport. JProsthetDent 1968:20426-
33.
34.
35.
36.
37.
38.
437. 18. v l m a n SC.Peyton FA. Acrylics and other synthetic resins used in dentistry. Philadelphia: J B Uppincott, 1946. 19. Paffenbarger GC, Nelsen RJ. Sweeney WT. Direct and in-
39.
direct filIing resins: a review of some physical and chemical properties. J Am Dent h s o c 1953:4751&524. 20. Coy HD. Directresinfillings.JAmDentAssoc 1953:47:532-
40.
537. 21. U.S.Patent3.066,112:DentalFillingMaterials Comprising Vinyl Silane Treated Fused Silica and A Binder Consisting
41.
of the Reaction Product of Bis Phenol and GlyddylAcrylate. 1962. 42.
22. Bowen RL,Rodriguez MS. Tensile strength and modulus of elasticity of tooth structure and several restorative materials. J Am Dent Assoc 1962;64:378. 23. Bowen RL. Effect of particle shape and size distribution in a reinforced polymer. J Am Dent Assoc 1964;69:481-495. 24. Chandler HH, Bowen RL. Paffenbarger GC,MullineauxAL. Clinical investigation of radiopaque composite restorative materials. J Am Dent Assoc 1971;50:935-940.
43.
posite resins. 1. Preparation of provisional glass formulations. J Dent Res 1976;55:738-747. Bowen RL, Reed LE. Semiporous reinforcing fillers for composite resins. 11. Heat treatment and etching characteristics. J Dent Res 1976;55:748-756. U.S. Patent 4,215,033: Composite Dental Materials. 1980. Bowen RL, Cobb EN, Rapson JE. Adhesive bonding ofvarious materials to hard tooth tissues. X X V . Improvement in bond strength to dentin. J Dent Res 1982;61:1070-1076. Bowen RL. Cobb EN, Setz LE. Adhesive bonding of various materials to hard tooth tissues. XXVII. Adhesive bonding to dentin and enamel. Dentistry 82 1982;2: 11-13. Bowen RL. Cobb EN. A method for bonding to dentin and enamel. J Am Dent Assoc 1983;107:734-736. Bowen RL, Cobb EN, Misra DN. Adhesive bonding of various materials to hard tooth tissues. XXVIlLAdhesivebonding by surface initiation of polymerization. Ind Eng Chem Prod Res Dev 1984:23:78-8 1. Bowen RL. Cobb EN. Setz LE. Recently developed concepts in adhesive bonding of composites to dentin and enamel. In: Proceedings of the James A. English Symposium, State University of New York at Buffalo. Buffalo Dent Rev 1984;1:10-12. Bowen RL, Eick J D , Henderson DA. Anderson DW. Adhesivebondingofvariousmaterials to hard tooth tissues.XXIX. Smear layer: removal and bonding considerations. Oper Dent Suppl 1984:3:30-34. US. Patent 4,521,550: Method for ObtainingStrongAdhesive Bonding of Composites to Dentin. Enamel and Other Substrates, 1985. U.S. Patent 4,588.576: Multi-step Method for Obtaining Strong Adhesive Bonding of Composites to Dentin. Enamel and Other Substrates, 1986. U.S. Patent 4,659,751: Simplified Method for Obtaining Strong Adhesive Bonding of Composites to Dentin, Enamel and Other Substrates, 1987. Bowen RL. Investigation of the surfaces of hard tooth tissues by a surface activity test. In Phillips RW. Ryge G . eds.: proceedings.Workshop on adhesiverestorative dental materials at Indiana University. Sept. 28-29. 1961. Spencer, Indiana: Owen Utho Senrice. 1961: 177-191. U.S. Patent 3,179,623: Method of Preparing a Monomer Having Phenoxy and Methacrylate Groups Linked by Hy droxy Glycerol Groups, 1965. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. 11. Bonding to dentin promoted by a surfaceactive comonomer. J Dent Res 1965:44:895-902. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. 111. Bonding to dentin improved by pretreatment and the use of a surface-active comonomer. J Dent Res 1965:M 903-905. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. IV. Bonding to dentin, enamel, and fluorapatite improved by the use of a surface-active comonomer. J Dent Res 1965:44:906-911. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. V. The effect of a surface-active comonomer on adhesion to diversesubstrates. J Dent Res 196544:13691373.
89
JOURNAL OF ESTHEnC DEl\mSTRY VOLUME 3. NUMBER 3 May/June 1991
50. Cobb EN, Blosser RL, Bowen RL. Johnston AD. Ferric oxalate with nitric acid a s a conditioner in an adhesive bonding system. J Adhesion 1989:28:41-49. 51. Blosser RL. Bowen RL. Effects of purified ferric oxalate/ nitric acid solutions as a pretreatment for the NTG-GMA and PMDM bonding system. Dent Mater 1988:4:225-231. 52. Stanley HR. Bowen R L, Cob b EN. Pulp responses to a dentin and enamel adhesive bonding procedure. Oper Dent 1988:13: 107-113. 53. Bowen RL, Tung MS, Blosser RL, Asmussen E. Dentine and enamel bonding agents. Int Dent J 1987:37:158-161. 54. Blosser RL, Rupp NW.Stanley HR, Bowen RL. Pulpal and micro-organismresponses to two experimentaldental bonding systems. Dent Mater 1989:5:140-144. 55. Johnston AD, Asmussen E. Bowen RL. Substitutes for NPhenylglycine in adhesive bonding to dentin. J Dent Res 1989:68: 1337-1344. 56. Johnston AD, Farahani M. Bowen RL. Effect of catalyst on restorative dental material synthesis. J Dent Res 1990: 69:233 (Abstr 995). 57. Bowen RL. Eichmiller FC, MarjenhoffWA, Rupp NW. Adhesive bonding of composites. J A m Coll Dent 1989:56: 10-13.
44. Bowen RL.Adhesive bonding of various materials to hard tooth tissues. VII. Metal salts as mordants for coupling agents. In: Moskowitz HD. Ward GT. Woolridge ED, eds. Dental adhesive materials. Proceedings for a symposium held November 8-9. 1973at the Hunter-Bellewe School of Nursing. New York: Prestige Graphic Services. 1974:205221. 45. Misra DN. Bowen RL. Wallace BM. Adhesive bonding of various materials to hard tooth tissues. VIII. Nickel and copper ions on hydroxyapatite: role of ion exchange and surface nucleation. J Colloid Interface Sci 1975:51:3643. 46. Misra DN. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. X. Initial rates of adsorption of nickel or copper ions on hydroxyapatite surface. J Biomed Mater Res 1978:12:505-515. 47. Bowen RL. Adhesive bonding of various materials to hard tooth tissues. X N . Enamel mordant selection assisted by ESCA KPS). J Dent Res 1978:57:551-556. 48. Bowen RL. McClendon LT. Gills TE. Adhesive bonding of various materials to hard tooth tissues. XV. Neutron acti vation analysis of dentin sorption of mordant salts.J Dent Res 1978;57 255-260. 49. Chen RS,Bowen RL. The use of N-phenylglycinein a dental adhesive system. J Adhesion Sd Techno1 1989;3:49-54.
90
