- Email: [email protected]
Fractured surface characterization: wet versus dry bonding
dental materials Dental Materials 18 (2002) 95±102
www.elsevier.com/locate/dental
Fractured surface characterization: wet versus dry bonding M. Hashimoto a,*, H. Ohno b, M. Kaga a, H. Sano c, K. Endo b, H. Oguchi a a
Department of Pediatric Dentistry, School of Dentistry, Hokkaido University, North 13, West 17, Kita-ku, Sapporo 060-8586, Hokkaido, Japan b Department of Dental Materials Science, School of Dentistry, Health Sciences University of Hokkaido, Hokkaido, Japan c Department of Operative Dentistry, School of Dentistry, Hokkaido University, Hokkaido, Japan Received 27 July 2000; received in revised form 31 October 2000; accepted 30 January 2001
Abstract Objective: Fractographic analysis was conducted to evaluate the resin±dentin bond structures made under wet and dry conditions. Methods: Resin±dentin bonded specimens were prepared using two adhesive resin systems (Single Bond/SB; 3M and All Bond 2/AB2; Bisco Inc) under wet and dry conditions. The specimens were sectioned perpendicular to the adhesive interface to produce a square barshaped specimen (adhesive area: 0.9 mm 2) by means of a diamond saw. The mean bond tensile test was then conducted at a crosshead speed of 1.0 mm/min. The mean bond strengths were statistically compared with two-way ANOVA and Fisher's PLSD test (p , 0.05). Subsequently, the fractured surfaces of all specimens were examined using SEM and the area fractions of failure modes (%) were measured using an image analyzer on SEM microphotographs. Results: No signi®cant differences in tensile-bond strength were observed between SB (60.1 ^ 16.4 MPa) and AB2 (69.8 ^ 17.4 MPa) (p . 0.05) under wet conditions. However, the bond strength either of SB or AB2 made under wet conditions was signi®cantly greater than those made under dry conditions (SB: 26.2 ^ 12.5 MPa and AB2: 6.8 ^ 3.3 MPa) (p , 0.05). Under fractographic analysis, the major portion at the fractured surface was occupied by the cohesive failure of bonding resin and the resin composite for the wet conditions, and the top of the hybrid layer for the dry conditions in both systems. Signi®cance: The interaction between the top of the hybrid layer and the bonding resin in¯uenced the bond integrity. q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. Keywords: Wet bonding technique; Fractography; Micro-tensile bond test; Hybrid layer
1. Introduction It has been speculated that the integrity of the hybrid layer is affected by the degree of resin in®ltration of the exposed collagen network after acid application [1]. Several studies have shown that the amount of resin impregnation into the demineralized dentin gradually decreases toward the bottom of the hybrid layer by micro-Raman analysis [2,3]. Many studies have reported that the bond strength obtained with a wet bonding technique is greater than that with a dry one [4±6]. It is thought that the shrinkage of collagen ®brils is induced by the air blast, leading to the prevention of resin in®ltration into the collagen web. However, little is known about the differences in the test-fractured surfaces obtained after bonding under wet and dry conditions. Incomplete resin impregnation of the collagen network leaves an exposed demineralized dentin zone beneath the hybrid layer [7,8]. Moreover, it has been suggested that the exposed * Corresponding author. Tel.: 181-11-706-4292; fax: 181-11-706-4307. E-mail address: [email protected] (M. Hashimoto).
collagen web is susceptible to hydrolytic degradation over a long period, leading to the reduction of bond strength both in vivo [9] and in vitro [10±12]. Thus, the presence of the demineralized dentin zone within the bond structure is a matter of a special interest in adhesive dentistry. We postulated that the demineralized dentin was easily created under dry conditions. Therefore, the purpose of this study was to evaluate the relation between the demineralized dentin and the bond integrity, using specimens made under wet and dry conditions. 2. Materials and methods 2.1. Tooth preparation Twelve non-carious human premolars extracted for orthodontic reasons with the informed consent of the patients were employed. The teeth were stored in normal saline solution at 48C for less than 1 month after extraction. Flat dentin surfaces of the coronal portion were prepared perpendicular to the long axis of the tooth to expose a ¯at
0109-5641/02/$22.00 + 0.00 q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. PII: S 0109-564 1(01)00023-9
96
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
Table 1 Chemical formulations of acid-conditioner, primer, bonding resin, and resin composite from the two dentin adhesive systems investigated (Lot number) Material (manufacturer)
Acid-conditioner
Primer
All Bond 2 (Bisco Inc.,Schaumburg, IL)
Uni-Etch Primer A (9800001637) NTG(9800001774) GMA acetone, ethanol, water 32.0% phosphoric acid Primer B (9800001638) BPDM, acetone Single Bond (3M Dental Scotchbond Etching Products Division, St Paul, MN) Gel (9AL) 35.0% phosphoric acid
dentin surface using a model trimmer with a water coolant. The surface was then ground with 600-grit silicon carbide paper under wet conditions for 30 s. 2.2. Bonding procedure Two commercially available adhesive resin systems; Single Bond(SB; 3M Dental Products Division, St Paul, MN) and All Bond 2(AB2; Bisco Inc., Schumburg, IL) were used in this study (Table 1). Specimens were randomly assigned to the following two acid-conditioned dentin surfaces. The prepared dentin surfaces were acid conditioned for 15 s and were rinsed off with distilled water for 10 s. For wet bonding, excess water was blot dried from the dentin surface with a cotton pellet, leaving the surface visibly moist. For dry bonding, the acidconditioned dentin surfaces were dried for 5 s with oil-free compressed air using an air syringe, keeping the syringe tip 10 cm from the dentin surface. The following bonding procedures were conducted according to the manufacturers' instructions. After applying the bonding resin, six 1-mm increments of a resin composite were built up and light cured for 60 s with a light-curing unit (Curing Light XL 3000; 3M Dental Products Division) for each increment.
Bonding resin
Resin composite
D/E Bonding resin (099218) HEMA, UDMA, Bis-GMA
Aelite-Fil: A2 (9900005824) Bis-GMA TEGDMA Filler
Single Bond (9DE) HEMA, Bis-GMA polyalkenoic acid copolymer, ethanol, water
Z100: A2 (9PE) Bis-GMA TEGDMA Filler
of the four groups. The bond strengths obtained were subjected to two-way ANOVA, followed by Fisher's PLSD test at p , 0.05. 2.4. Fractographic analysis After the micro-tensile test, all fractured surfaces of the dentin side of each specimen were sputter coated with gold (Ion sputter E-1030; Hitachi Ltd, Tokyo, Japan) and observed with a ®eld-emission scanning electron microscope (S-4000; Hitachi Ltd, Tokyo, Japan). The adhesive areas on the dentin side of the fractured surface were measured with an image analyzer (Digitizer, KD4030B; Graphtec, Tokyo, Japan) on SEM photomicrographs. To evaluate the failure pattern for each adhesive system, the area fractions of the failure modes per total fractured surfaces (%) of all specimens were calculated in the SEM
2.3. Micro-tensile bond test After the bonded specimens had been stored in sterilized water at 378C for 24 h, those to undergo microtensile testing were sectioned perpendicular to the adhesive interface with a diamond saw (Isomet; Buehler Ltd, Lake Bluff, IL). Subsequently, these slices of resin±dentin bonded slabs were sectioned with a diamond saw to produce a square bar-shaped specimen (adhesive area: 0.9 mm 2), as shown in Fig. 1(a) [13]. Four square barshaped specimens were obtained per tooth. These specimens were then attached to a testing apparatus with a cyanoacrylate adhesive (Model Repair II blue; Sankin Industry Co. Ltd, Tokyo, Japan) and a tensile load was applied with a table-top material tester (EZ Test; Shimadzu Co., Kyoto, Japan) at a crosshead speed of 1.0 mm/min. Twenty specimens were tested from each
Fig. 1. (a) Specimen preparation for the micro-tensile bond test. A: Sectioned perpendicular to the adhesive interface using a diamond saw. C: Sectioned perpendicular to the adhesive interface to produce a square bar-shaped specimen using a diamond saw. (b) Classi®cation of failure modes.
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
97
photomicrographs using the image analyzer. The failure modes were classi®ed into the following ®ve groups by failure in: (I) the resin composite; (II) the bonding resin; (III) the top of the hybrid layer; (IV) the middle of the hybrid layer; (V) the demineralized dentin zone (Fig. 1(b)). The percentage of each failure mode on the dentin side of the specimen was then measured with the image analyzer on SEM photomicrographs. 3. Results The adhesive areas at the dentin side of the fractured surface, calculated from SEM photomicrographs using the image analyzer, were 0.902 ^ 0.051 mm 2 (80 specimens). In Fig. 2, it can be seen that there were no signi®cant differences in tensile-bond strengths of the specimens made under wet conditions between SB (60.1 ^ 16.4 MPa) and AB2 (69.8 ^ 16.4 MPa) (p . 0.05). However, the bond strength of SB (26.2 ^ 12.5 MPa) was signi®cantly greater than that of AB2 (6.8 ^ 3.3 MPa), when dry conditions were employed (p , 0.05). Cohesive failure of the resin composite or bonding resin accounted for 61.9% of the fractured surface after wet bond of SB. Failure of the top of the hybrid layer was shown as the major portion of the fractured surface for AB2 (52.0%). In contrast, various failure patterns were observed, such as at the top and middle of the hybrid layer and the demineralized dentin zone for SB. The area fraction of cohesive failure (resin composite and bonding resin) after wet bonding was greater than that after dry conditioning for both systems. The major failure was at the top of the hybrid layer (SB: 49.9%, AB2: 91.6%) after dry bonding. The area fraction of the failure mode of the top of the hybrid layer of the bonds made under dry conditions was greater than that for wet conditions in both systems. Cohesive failure of the dentin was not observed at the fractured surface of the 80 specimens investigated in the present study. Fig. 3(a) shows the overall fractured surface on the dentin side of a specimen of AB2 after wet bonding. The sections marked (b)±(d) in the ®gure are shown at higher magni®cations as SEM photomicrographs in Fig. 3(b)±(d). Fig. 3(b) shows the boundary between the top and the middle of the hybrid layer. A typical feature of the middle of the hybrid layer was that it was localized beneath the top of the hybrid layer. Fig. 3(c) and (d) show the failures of the top and the middle of the hybrid layer, respectively. Fig. 3(e) shows a schematic illustration of the area fractions of the failure modes in the specimen of Fig. 3(a), calculated by means of the image analyzer. These were 1.1% in the bonding resin, 70.5% in the top of hybrid layer, and 28.4% in the middle of the hybrid layer. Fig. 4 shows the failures of the top of the hybrid layer in the dentin sides of specimens made under wet and dry conditions of the two adhesive systems tested (Fig. 4(a): wet AB2; (b) dry AB2; (c) wet SB; and (d) dry SB).
Fig. 2. Tensile bond strength (a) and area fractions of failure modes on fractured surfaces (b) for each group. Identical letters indicate means not signi®cantly different at p , 0.05. n 20 for each group.
Scratches were observed at the surface of the top of the hybrid layer made under wet and dry conditions. Fig. 5(a) shows the overall fractured surface on the dentin side of a specimen of SB made under dry conditions. The sections marked (b) and (c) in the ®gure are shown at higher magni®cations as SEM photomicrographs in Fig. 5(b) and (c), respectively. In this specimen, the demineralized dentin zone occupied the major portion of the fractured surface (92.4%). Exposed collagen ®brils with the open space within the collagen web were characteristic of the demineralized dentin. 4. Discussion From SEM observations of the fractured surface on the dentin side, the area fraction of failure mode for each specimen was calculated in the manner illustrated in Fig. 3(e). It has been assumed that the cohesive failure at the fractured surface is reduced when using the micro-tensile
98
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
Fig. 3. Total fractured surface on the dentin side of a specimen of All Bond 2 prepared under wet conditions (a). (b)±(d): Higher magni®cation of the sections of the fractured surfaces marked (b)±(d) in (a). Bonding resin (B), top of hybrid layer (TH), middle of hybrid layer (MH). (e) The distribution and area fractions of the failure modes shown in (a).
test compared to the shear test or traditional tensile testing [14,15]. However, the cohesive failure of the bonding resin and the resin composite accounted for the major portion on the fractured surface of the wet bond as follows: 61.9% of SB, 30.6% of AB2 (Fig. 2(b)). Several reports have suggested that perfect adhesion is
achieved at the resin±dentin when cohesive failure of the dentin is seen at the fractured surface, which implies that the bond integrity is greater than the physical properties of the dentin [16]. In fact, cohesive failure of the dentin was usually observed on the fractured surface using the shear bond test [17,18], traditional tensile test [19] and
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
99
Fig. 4. The failure of the top of the hybrid layers of bonds made under wet and dry conditions of the two adhesive systems tested: (a) wet AB2; (b) dry AB2; (c) wet SB; and (d) dry SB.
micro-tensile test for hourglass-shaped specimens [9]. No cohesive failure of the dentin was observed on the fractured surfaces of the 80 specimens in the present study. Based on these facts, the failure mode on the fractured surface was the result of the bond testing method or the shape of the resin±dentin bonded slabs. Further work is required to clarify the effects of the bond-test method and the specimen's shape on the bond strength results and failure modes. The typical morphological phase of the top of the hybrid layer showed no porosity of inter®bular space with scratches at the fractured surface. The SEM microphotograph in Fig. 3(b) indicated that the middle of the hybrid layer was localized below the top of the hybrid layer. These morphological differences are shown in Fig. 3(c) and (d). Exposed collagen ®brils with the presence of voids between the collagen ®brils were typical of the demineralized dentin (Fig. 5). For AB2, the failure pattern, except for cohesive failure of the resin composite and bonding resin, was mainly classi®ed into the top of the hybrid layer (39.5%) with wet bonding. However, for SB, failure occurred in various portions such as the middle and top of the hybrid layer
and the demineralized dentin zone. Fig. 5 shows the failure of a specimen made under dry conditions with the SB system, where the major region at the fractured surface was occupied by the demineralized dentin zone. However, the total percentage of the failure of demineralized dentin was small on the fractured surfaces for the wet and the dry conditions. Thus, the exposed collagen ®brils within the demineralized dentin zone might not play a major role in the reduction of the bond strength of the two adhesive systems tested. Nakabayashi et al. [7] found, using the 4META MMA/TBB resin system, that the demineralized dentin zone was responsible for the bond failure under tensile loading. In contrast, several reports using the OneStep (Bisco Inc.) demonstrated that the fracture surfaces of the specimens that showed lower bond strength were at the top of the hybrid layer [20,21]. Based on those reports, it appears that the failure pattern of the fractured surface might depend on the adhesive system. The results in this study were obtained in a short-term period (24 h after bonding). Even though high bond strength of resin to dentin was achieved for the short term, concern still remains about its long-term durability. It has been suggested that the bond
100
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
Fig. 5. (a) Total fractured surface on the dentin side of a specimen of Single Bond made under dry conditions; (b) and (c) show higher magni®cation of the sections of the fractured surfaces marked (b) and (c) in (a).
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
strength of resin±dentin decreases because of the hydrolysis of collagen ®brils within the demineralized dentin occurring with time in the oral environment [9]. Hence, the demineralized dentin zone is related to the longevity of resin±dentin bond structures. Recently, several SEM [5,22], TEM [23,24] and AFM examinations [25] showed using a polyalkenoic acid±resin system, that a primer layer is formed at the top of the hybrid layer due to the chemical action of polyalkenoic acid copolymer. Polyalkenoic acid copolymer is also included in the bonding resin of the SB resin (Table 1). Thus, it is possible that the chemical union of calcium±polyalkenoic acid at the top of the hybrid layer prevented fracturing in this region. Hence, the area fraction of the failure mode of the top of the hybrid layer using SB is lower than that of AB2 for bonds made under wet conditions. Moreover, for dry conditions, the bond strength of SB was signi®cantly greater than that of AB2 (p , 0.05). The failure of the top of the hybrid layer accounted for the major portion of the fractured surface for AB2 (91.6%), and was greater than that for SB (49.9%). These results suggested that the chemical union of polyalkenoic acid and the top of the hybrid layer prevented failure in this zone. We speculated that the chemical binding of polycarbonic acid acted not only in wet conditions but also dry conditions. PerdigaÄo et al. [6] reported that the water included in the adhesive resin acts as a re-wetting agent when applied on dry dentin. Water is included in the composition of AB2 and SB (Table 1). Thus, another explanation was that the re-wetting effect of SB was greater than that of AB2. Hence, the decrease of bond strength and increase in the failure of the top of the hybrid layer for SB might be lower than that for AB2. Analysis of the results of this study indicated that the failure of the top of the hybrid layer accounted for the major portion of the fractured surface bonded under dry conditions for both systems. Thus, the bond integrity might be determined by the amount of in®ltration of the adhesive resin into the exposed collagen ®brils, especially the interaction at the top of the hybrid layer and the bonding resin layer. The morphological appearance at the top of the hybrid layer between AB2 and SB was not so different for bonds made under wet and dry conditions, while a morphological difference was recognized between wet and dry conditions for both systems (Fig. 4). A possible explanation is that the mineral matrix of the dentin was removed and the exposed super®cial collagen ®brils were entangled with micro®brils at the top part of the collagen web after chemical acidi®cation at the mineralized dentin surface [26,27]. It has been speculated that the intermolecular cross-linking of collagen ®brils breaks down, and that the unbinding of the triple helix of the collagen ®brils might create the membrane structure at the top part of the collagen network [28]. Moreover, brief air drying (5 s in this study) might cause shrinkage of the collagen network, thus inducing a decrease of the size of the microspace between the ®brils [29], leading to the formation of the membrane structure at the top part of
101
the exposed collagen web [26]. The smooth surface of the top of the hybrid layer bonded under dry conditions shown in Fig. 4(b) and (d) was similar to the membrane structure previously reported [26,27,29]. This result suggested that the volume fraction of the adhesive resin within the top part of the collagen web decreased [26]. Hence, the top of the hybrid layer was increased proportionally at the fractured surface because failure was liable to develop at this portion. Moreover, the membrane structure might have prevented the in®ltration of the resin into the collagen web. Then, a nanoscale void within the hybrid layer [1,30,31] and the demineralized dentin zone beneath the hybrid layer could be readily created within the bond structure. Comparing the top of the hybrid layer and the demineralized dentin zone after bonding under dry conditions, the bond integrity between the bonding resin layer and the top of the hybrid layer might be weaker than the mechanical strength of the exposed collagen ®brils within the demineralized dentin. Therefore, it is possible that the failure of the top of the hybrid layer concealed the presence of the demineralized dentin zone on the fractured surface under morphological observation. In conclusion, at the time of the resin application, great care must be paid to the surface condition of the dentin after the acidi®cation that in¯uences both the short- and the long-term integrity of the resin±dentin bonds in the clinical situation. Acknowledgements This work was supported in part by a Grant-in-Aid for Scienti®c Research No. 11470401 from the Ministry of Education, Science and Culture, Japan. References [1] Hashimoto M, Ohno H, Endo K, Kaga M, Sano H, Oguchi H. The effect of hybrid layer thickness on bond strength: demineralized dentin zone of hybrid layer. Dent Mater 2000;16:406±11. [2] Suzuki M, Kato H, Wakumoto S. Vibrational analysis by Raman spectroscopy of the interface between dental adhesive resin and dentin. J Dent Res 1991;70:1092±7. [3] Van Meerbeek H, Mohrbacher H, Celis JP, Roos JR, Braem M, Lambrechts P, Vanherle G. Chemical characterization of the resin± dentin interface by micro-Raman spectroscopy. J Dent Res 1993; 72:1423±8. [4] Kanca 3rd J. Improving bond strength through acid etching of dentin and bonding to wet dentin surfaces. J Am Dent Assoc 1992;123:35± 43. [5] Nakajima M, Sano H, Zheng L, Tagami M, Pashley DH. Effect of moist versus dry bonding to normal vs. caries-affected dentin with Scotchbond Multi-Purpose Plus. J Dent Res 1999;78:1298±303. [6] PerdigaÄo J, Van Meerbeek B, Lopes MM, Ambrose WW. The effect of a re-wetting agent on dentin bonding. Dent Mater 1999;15:285±95. [7] Nakabayashi N, Watanabe A, Arao T. A tensile test to facilitate identi®cation of defects in dentine bonded specimens. J Dent 1998; 26:379±85. [8] Spencer P, Swafford JR. Unprotected protein at the dentin±adhesive interface. Quint Int 1999;30:501±7. [9] Hashimoto M, Ohno H, Kaga M, Endo K, Sano H, Oguchi H. In vivo
102
[10] [11] [12] [13] [14]
[15] [16] [17] [18] [19] [20]
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102 degradation of resin±dentin bonds in humans over 1 to 3 years. J Dent Res 2000;79:1385±91. Burrow MF, Satoh M, Tagami J. Dentin bond durability after three years using a dentin bonding agent with and without priming. Dent Mater 1996;12:302±7. Kato G, Nakabayashi N. The durability of adhesion to phosphoric acid etched, wet dentin substrates. Dent Mater 1998;14:347±52. Kitasako Y, Burrow MF, Nikaido T, Tagami J. The in¯uence of storage on dentin bond durability of resin cement. Dent Mater 2000;16:1±6. Shono Y, Ogawa T, Terashita M, Carvalho RM, Pashley EL, Pashley DH. Regional measurement of resin±dentin bonding as an array. J Dent Res 1999;78:699±705. Sano H, Shono T, Sonoda H, Takatsu T, Ciucchi B, Carvalho RM, Pashley DH. Relationship between surface area for adhesion and tensile bond strength. Evaluation of a micro-tensile bond test. Dent Mater 1994; 10:236±40. Pashley DH, Sano H, Ciucchi B, Yoshiyama M, Carvalho RM. Adhesion testing of dentin bonding agents: a review. Dent Mater 1995; 11:117±25. Davidson CL, Abdalla AL, de Gee AJ. An investigation into the quality of dentine bonding systems for accomplishing a durable bond. J Oral Rehabil 1993;20:291±300. Versluis A, Tantbirojn D, Douglas WH. Why do shear bond tests pull out dentin? J Dent Res 1997;76:1298±307. Tantbirojn D, Cheng YS, Versluis A, Hodhges JH, Douglas WH. Nominal shear or fracture mechanics in the assessment of composite±dentin adhesion? J Dent Res 2000;79:41±48. Van Noort K, Cardew GE, Howard IC, Noroozi S. The effect of local interfacial geometry on the measurement of the tensile bond strength to dentin. J Dent Res 1991;70:889±93. Pereira PNR, Okuda M, Sano H, Yoshikawa T, Burrow MF, Tagami J. Effect of intrinsic wetness and regional difference on dentin bond strength. Dent Mater 1999;15:46±53.
[21] Yoshikawa T, Sano H, Burrow MF, Tagami J, Pashley DH. Effects of dentin depth and cavity con®guration on bond strength. J Dent Res 1999;78:898±905. [22] Nakajima M, Sano H, Burrow MF, Tagami M, Yoshiyama M, Ebisu S, Ciucchi B, Russel CM, Pashley DH. Tensile bond strength and SEM evaluation of caries-affected dentin using dentin adhesives. J Dent Res 1995;74:1679±88. [23] Van Meerbeek B, Yoshida Y, Lambrechts P, Vanherle G, Duke ES, Eick JD, Robinson SJ. A TEM study of two water-based adhesive systems bonded to dry and wet dentin. J Dent Res 1998;77:50±59. [24] Tay FR, Gwinnet JA, Wei SHY. Micromorphological spectrum of acid-conditioned dentin following the application of a water-based adhesive. Dent Mater 1998;14:329±38. [25] Yoshida Y, Van Meerbeek B, Snauwaert J, Hellemans L, Lambrechts P, Vanherle G, Wakasa K, Pashley DH. A novel approach to AFM characterization of adhesive tooth±biomaterial interfaces. J Biomed Mater Res 1999;47:85±90. [26] Pashley DH, Ciucchi B, Yoshiyama M, Carvalho RM. Permeability of dentin to adhesive agents. Quint Int 1995;24:618±31. [27] PerdigaÄo J, Lambrechts P, Van Meerbeek B, Tome AR, Vanherle G, Lopes AB. Morphological ®eld emission-SEM study of the effect of six phosphoric acid etching agents on human dentin. Dent Mater 1996;12:262±71. [28] Maciel KT, Carvalho RM, Ringle RD, Preston CD, Russell CM, Pashley DH. The effects of acetone, ethanol, HEMA, and air on the stiffness of human decalci®ed dentin matrix. J Dent Res 1996;75: 1851±8. [29] Pashley DH, Zhang Y, Agee KA, Rouse CJ, Carvalho RM, Russell CM. Permeability of demineralized dentin to HEMA. Dent Mater 2000;16:7±14. [30] Sano H, Shono T, Takatsu T, Hosoda H. Microporous dentin zone beneath resin-impregnated layer. Oper Dent 1994;19:59±64. [31] Li H, Burrow MF, Tyas MJ. Nanoleakage patterns of four dentin bonding systems. Dent Mater 2000;16:48±56.
www.elsevier.com/locate/dental
Fractured surface characterization: wet versus dry bonding M. Hashimoto a,*, H. Ohno b, M. Kaga a, H. Sano c, K. Endo b, H. Oguchi a a
Department of Pediatric Dentistry, School of Dentistry, Hokkaido University, North 13, West 17, Kita-ku, Sapporo 060-8586, Hokkaido, Japan b Department of Dental Materials Science, School of Dentistry, Health Sciences University of Hokkaido, Hokkaido, Japan c Department of Operative Dentistry, School of Dentistry, Hokkaido University, Hokkaido, Japan Received 27 July 2000; received in revised form 31 October 2000; accepted 30 January 2001
Abstract Objective: Fractographic analysis was conducted to evaluate the resin±dentin bond structures made under wet and dry conditions. Methods: Resin±dentin bonded specimens were prepared using two adhesive resin systems (Single Bond/SB; 3M and All Bond 2/AB2; Bisco Inc) under wet and dry conditions. The specimens were sectioned perpendicular to the adhesive interface to produce a square barshaped specimen (adhesive area: 0.9 mm 2) by means of a diamond saw. The mean bond tensile test was then conducted at a crosshead speed of 1.0 mm/min. The mean bond strengths were statistically compared with two-way ANOVA and Fisher's PLSD test (p , 0.05). Subsequently, the fractured surfaces of all specimens were examined using SEM and the area fractions of failure modes (%) were measured using an image analyzer on SEM microphotographs. Results: No signi®cant differences in tensile-bond strength were observed between SB (60.1 ^ 16.4 MPa) and AB2 (69.8 ^ 17.4 MPa) (p . 0.05) under wet conditions. However, the bond strength either of SB or AB2 made under wet conditions was signi®cantly greater than those made under dry conditions (SB: 26.2 ^ 12.5 MPa and AB2: 6.8 ^ 3.3 MPa) (p , 0.05). Under fractographic analysis, the major portion at the fractured surface was occupied by the cohesive failure of bonding resin and the resin composite for the wet conditions, and the top of the hybrid layer for the dry conditions in both systems. Signi®cance: The interaction between the top of the hybrid layer and the bonding resin in¯uenced the bond integrity. q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. Keywords: Wet bonding technique; Fractography; Micro-tensile bond test; Hybrid layer
1. Introduction It has been speculated that the integrity of the hybrid layer is affected by the degree of resin in®ltration of the exposed collagen network after acid application [1]. Several studies have shown that the amount of resin impregnation into the demineralized dentin gradually decreases toward the bottom of the hybrid layer by micro-Raman analysis [2,3]. Many studies have reported that the bond strength obtained with a wet bonding technique is greater than that with a dry one [4±6]. It is thought that the shrinkage of collagen ®brils is induced by the air blast, leading to the prevention of resin in®ltration into the collagen web. However, little is known about the differences in the test-fractured surfaces obtained after bonding under wet and dry conditions. Incomplete resin impregnation of the collagen network leaves an exposed demineralized dentin zone beneath the hybrid layer [7,8]. Moreover, it has been suggested that the exposed * Corresponding author. Tel.: 181-11-706-4292; fax: 181-11-706-4307. E-mail address: [email protected] (M. Hashimoto).
collagen web is susceptible to hydrolytic degradation over a long period, leading to the reduction of bond strength both in vivo [9] and in vitro [10±12]. Thus, the presence of the demineralized dentin zone within the bond structure is a matter of a special interest in adhesive dentistry. We postulated that the demineralized dentin was easily created under dry conditions. Therefore, the purpose of this study was to evaluate the relation between the demineralized dentin and the bond integrity, using specimens made under wet and dry conditions. 2. Materials and methods 2.1. Tooth preparation Twelve non-carious human premolars extracted for orthodontic reasons with the informed consent of the patients were employed. The teeth were stored in normal saline solution at 48C for less than 1 month after extraction. Flat dentin surfaces of the coronal portion were prepared perpendicular to the long axis of the tooth to expose a ¯at
0109-5641/02/$22.00 + 0.00 q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. PII: S 0109-564 1(01)00023-9
96
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
Table 1 Chemical formulations of acid-conditioner, primer, bonding resin, and resin composite from the two dentin adhesive systems investigated (Lot number) Material (manufacturer)
Acid-conditioner
Primer
All Bond 2 (Bisco Inc.,Schaumburg, IL)
Uni-Etch Primer A (9800001637) NTG(9800001774) GMA acetone, ethanol, water 32.0% phosphoric acid Primer B (9800001638) BPDM, acetone Single Bond (3M Dental Scotchbond Etching Products Division, St Paul, MN) Gel (9AL) 35.0% phosphoric acid
dentin surface using a model trimmer with a water coolant. The surface was then ground with 600-grit silicon carbide paper under wet conditions for 30 s. 2.2. Bonding procedure Two commercially available adhesive resin systems; Single Bond(SB; 3M Dental Products Division, St Paul, MN) and All Bond 2(AB2; Bisco Inc., Schumburg, IL) were used in this study (Table 1). Specimens were randomly assigned to the following two acid-conditioned dentin surfaces. The prepared dentin surfaces were acid conditioned for 15 s and were rinsed off with distilled water for 10 s. For wet bonding, excess water was blot dried from the dentin surface with a cotton pellet, leaving the surface visibly moist. For dry bonding, the acidconditioned dentin surfaces were dried for 5 s with oil-free compressed air using an air syringe, keeping the syringe tip 10 cm from the dentin surface. The following bonding procedures were conducted according to the manufacturers' instructions. After applying the bonding resin, six 1-mm increments of a resin composite were built up and light cured for 60 s with a light-curing unit (Curing Light XL 3000; 3M Dental Products Division) for each increment.
Bonding resin
Resin composite
D/E Bonding resin (099218) HEMA, UDMA, Bis-GMA
Aelite-Fil: A2 (9900005824) Bis-GMA TEGDMA Filler
Single Bond (9DE) HEMA, Bis-GMA polyalkenoic acid copolymer, ethanol, water
Z100: A2 (9PE) Bis-GMA TEGDMA Filler
of the four groups. The bond strengths obtained were subjected to two-way ANOVA, followed by Fisher's PLSD test at p , 0.05. 2.4. Fractographic analysis After the micro-tensile test, all fractured surfaces of the dentin side of each specimen were sputter coated with gold (Ion sputter E-1030; Hitachi Ltd, Tokyo, Japan) and observed with a ®eld-emission scanning electron microscope (S-4000; Hitachi Ltd, Tokyo, Japan). The adhesive areas on the dentin side of the fractured surface were measured with an image analyzer (Digitizer, KD4030B; Graphtec, Tokyo, Japan) on SEM photomicrographs. To evaluate the failure pattern for each adhesive system, the area fractions of the failure modes per total fractured surfaces (%) of all specimens were calculated in the SEM
2.3. Micro-tensile bond test After the bonded specimens had been stored in sterilized water at 378C for 24 h, those to undergo microtensile testing were sectioned perpendicular to the adhesive interface with a diamond saw (Isomet; Buehler Ltd, Lake Bluff, IL). Subsequently, these slices of resin±dentin bonded slabs were sectioned with a diamond saw to produce a square bar-shaped specimen (adhesive area: 0.9 mm 2), as shown in Fig. 1(a) [13]. Four square barshaped specimens were obtained per tooth. These specimens were then attached to a testing apparatus with a cyanoacrylate adhesive (Model Repair II blue; Sankin Industry Co. Ltd, Tokyo, Japan) and a tensile load was applied with a table-top material tester (EZ Test; Shimadzu Co., Kyoto, Japan) at a crosshead speed of 1.0 mm/min. Twenty specimens were tested from each
Fig. 1. (a) Specimen preparation for the micro-tensile bond test. A: Sectioned perpendicular to the adhesive interface using a diamond saw. C: Sectioned perpendicular to the adhesive interface to produce a square bar-shaped specimen using a diamond saw. (b) Classi®cation of failure modes.
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
97
photomicrographs using the image analyzer. The failure modes were classi®ed into the following ®ve groups by failure in: (I) the resin composite; (II) the bonding resin; (III) the top of the hybrid layer; (IV) the middle of the hybrid layer; (V) the demineralized dentin zone (Fig. 1(b)). The percentage of each failure mode on the dentin side of the specimen was then measured with the image analyzer on SEM photomicrographs. 3. Results The adhesive areas at the dentin side of the fractured surface, calculated from SEM photomicrographs using the image analyzer, were 0.902 ^ 0.051 mm 2 (80 specimens). In Fig. 2, it can be seen that there were no signi®cant differences in tensile-bond strengths of the specimens made under wet conditions between SB (60.1 ^ 16.4 MPa) and AB2 (69.8 ^ 16.4 MPa) (p . 0.05). However, the bond strength of SB (26.2 ^ 12.5 MPa) was signi®cantly greater than that of AB2 (6.8 ^ 3.3 MPa), when dry conditions were employed (p , 0.05). Cohesive failure of the resin composite or bonding resin accounted for 61.9% of the fractured surface after wet bond of SB. Failure of the top of the hybrid layer was shown as the major portion of the fractured surface for AB2 (52.0%). In contrast, various failure patterns were observed, such as at the top and middle of the hybrid layer and the demineralized dentin zone for SB. The area fraction of cohesive failure (resin composite and bonding resin) after wet bonding was greater than that after dry conditioning for both systems. The major failure was at the top of the hybrid layer (SB: 49.9%, AB2: 91.6%) after dry bonding. The area fraction of the failure mode of the top of the hybrid layer of the bonds made under dry conditions was greater than that for wet conditions in both systems. Cohesive failure of the dentin was not observed at the fractured surface of the 80 specimens investigated in the present study. Fig. 3(a) shows the overall fractured surface on the dentin side of a specimen of AB2 after wet bonding. The sections marked (b)±(d) in the ®gure are shown at higher magni®cations as SEM photomicrographs in Fig. 3(b)±(d). Fig. 3(b) shows the boundary between the top and the middle of the hybrid layer. A typical feature of the middle of the hybrid layer was that it was localized beneath the top of the hybrid layer. Fig. 3(c) and (d) show the failures of the top and the middle of the hybrid layer, respectively. Fig. 3(e) shows a schematic illustration of the area fractions of the failure modes in the specimen of Fig. 3(a), calculated by means of the image analyzer. These were 1.1% in the bonding resin, 70.5% in the top of hybrid layer, and 28.4% in the middle of the hybrid layer. Fig. 4 shows the failures of the top of the hybrid layer in the dentin sides of specimens made under wet and dry conditions of the two adhesive systems tested (Fig. 4(a): wet AB2; (b) dry AB2; (c) wet SB; and (d) dry SB).
Fig. 2. Tensile bond strength (a) and area fractions of failure modes on fractured surfaces (b) for each group. Identical letters indicate means not signi®cantly different at p , 0.05. n 20 for each group.
Scratches were observed at the surface of the top of the hybrid layer made under wet and dry conditions. Fig. 5(a) shows the overall fractured surface on the dentin side of a specimen of SB made under dry conditions. The sections marked (b) and (c) in the ®gure are shown at higher magni®cations as SEM photomicrographs in Fig. 5(b) and (c), respectively. In this specimen, the demineralized dentin zone occupied the major portion of the fractured surface (92.4%). Exposed collagen ®brils with the open space within the collagen web were characteristic of the demineralized dentin. 4. Discussion From SEM observations of the fractured surface on the dentin side, the area fraction of failure mode for each specimen was calculated in the manner illustrated in Fig. 3(e). It has been assumed that the cohesive failure at the fractured surface is reduced when using the micro-tensile
98
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
Fig. 3. Total fractured surface on the dentin side of a specimen of All Bond 2 prepared under wet conditions (a). (b)±(d): Higher magni®cation of the sections of the fractured surfaces marked (b)±(d) in (a). Bonding resin (B), top of hybrid layer (TH), middle of hybrid layer (MH). (e) The distribution and area fractions of the failure modes shown in (a).
test compared to the shear test or traditional tensile testing [14,15]. However, the cohesive failure of the bonding resin and the resin composite accounted for the major portion on the fractured surface of the wet bond as follows: 61.9% of SB, 30.6% of AB2 (Fig. 2(b)). Several reports have suggested that perfect adhesion is
achieved at the resin±dentin when cohesive failure of the dentin is seen at the fractured surface, which implies that the bond integrity is greater than the physical properties of the dentin [16]. In fact, cohesive failure of the dentin was usually observed on the fractured surface using the shear bond test [17,18], traditional tensile test [19] and
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
99
Fig. 4. The failure of the top of the hybrid layers of bonds made under wet and dry conditions of the two adhesive systems tested: (a) wet AB2; (b) dry AB2; (c) wet SB; and (d) dry SB.
micro-tensile test for hourglass-shaped specimens [9]. No cohesive failure of the dentin was observed on the fractured surfaces of the 80 specimens in the present study. Based on these facts, the failure mode on the fractured surface was the result of the bond testing method or the shape of the resin±dentin bonded slabs. Further work is required to clarify the effects of the bond-test method and the specimen's shape on the bond strength results and failure modes. The typical morphological phase of the top of the hybrid layer showed no porosity of inter®bular space with scratches at the fractured surface. The SEM microphotograph in Fig. 3(b) indicated that the middle of the hybrid layer was localized below the top of the hybrid layer. These morphological differences are shown in Fig. 3(c) and (d). Exposed collagen ®brils with the presence of voids between the collagen ®brils were typical of the demineralized dentin (Fig. 5). For AB2, the failure pattern, except for cohesive failure of the resin composite and bonding resin, was mainly classi®ed into the top of the hybrid layer (39.5%) with wet bonding. However, for SB, failure occurred in various portions such as the middle and top of the hybrid layer
and the demineralized dentin zone. Fig. 5 shows the failure of a specimen made under dry conditions with the SB system, where the major region at the fractured surface was occupied by the demineralized dentin zone. However, the total percentage of the failure of demineralized dentin was small on the fractured surfaces for the wet and the dry conditions. Thus, the exposed collagen ®brils within the demineralized dentin zone might not play a major role in the reduction of the bond strength of the two adhesive systems tested. Nakabayashi et al. [7] found, using the 4META MMA/TBB resin system, that the demineralized dentin zone was responsible for the bond failure under tensile loading. In contrast, several reports using the OneStep (Bisco Inc.) demonstrated that the fracture surfaces of the specimens that showed lower bond strength were at the top of the hybrid layer [20,21]. Based on those reports, it appears that the failure pattern of the fractured surface might depend on the adhesive system. The results in this study were obtained in a short-term period (24 h after bonding). Even though high bond strength of resin to dentin was achieved for the short term, concern still remains about its long-term durability. It has been suggested that the bond
100
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
Fig. 5. (a) Total fractured surface on the dentin side of a specimen of Single Bond made under dry conditions; (b) and (c) show higher magni®cation of the sections of the fractured surfaces marked (b) and (c) in (a).
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102
strength of resin±dentin decreases because of the hydrolysis of collagen ®brils within the demineralized dentin occurring with time in the oral environment [9]. Hence, the demineralized dentin zone is related to the longevity of resin±dentin bond structures. Recently, several SEM [5,22], TEM [23,24] and AFM examinations [25] showed using a polyalkenoic acid±resin system, that a primer layer is formed at the top of the hybrid layer due to the chemical action of polyalkenoic acid copolymer. Polyalkenoic acid copolymer is also included in the bonding resin of the SB resin (Table 1). Thus, it is possible that the chemical union of calcium±polyalkenoic acid at the top of the hybrid layer prevented fracturing in this region. Hence, the area fraction of the failure mode of the top of the hybrid layer using SB is lower than that of AB2 for bonds made under wet conditions. Moreover, for dry conditions, the bond strength of SB was signi®cantly greater than that of AB2 (p , 0.05). The failure of the top of the hybrid layer accounted for the major portion of the fractured surface for AB2 (91.6%), and was greater than that for SB (49.9%). These results suggested that the chemical union of polyalkenoic acid and the top of the hybrid layer prevented failure in this zone. We speculated that the chemical binding of polycarbonic acid acted not only in wet conditions but also dry conditions. PerdigaÄo et al. [6] reported that the water included in the adhesive resin acts as a re-wetting agent when applied on dry dentin. Water is included in the composition of AB2 and SB (Table 1). Thus, another explanation was that the re-wetting effect of SB was greater than that of AB2. Hence, the decrease of bond strength and increase in the failure of the top of the hybrid layer for SB might be lower than that for AB2. Analysis of the results of this study indicated that the failure of the top of the hybrid layer accounted for the major portion of the fractured surface bonded under dry conditions for both systems. Thus, the bond integrity might be determined by the amount of in®ltration of the adhesive resin into the exposed collagen ®brils, especially the interaction at the top of the hybrid layer and the bonding resin layer. The morphological appearance at the top of the hybrid layer between AB2 and SB was not so different for bonds made under wet and dry conditions, while a morphological difference was recognized between wet and dry conditions for both systems (Fig. 4). A possible explanation is that the mineral matrix of the dentin was removed and the exposed super®cial collagen ®brils were entangled with micro®brils at the top part of the collagen web after chemical acidi®cation at the mineralized dentin surface [26,27]. It has been speculated that the intermolecular cross-linking of collagen ®brils breaks down, and that the unbinding of the triple helix of the collagen ®brils might create the membrane structure at the top part of the collagen network [28]. Moreover, brief air drying (5 s in this study) might cause shrinkage of the collagen network, thus inducing a decrease of the size of the microspace between the ®brils [29], leading to the formation of the membrane structure at the top part of
101
the exposed collagen web [26]. The smooth surface of the top of the hybrid layer bonded under dry conditions shown in Fig. 4(b) and (d) was similar to the membrane structure previously reported [26,27,29]. This result suggested that the volume fraction of the adhesive resin within the top part of the collagen web decreased [26]. Hence, the top of the hybrid layer was increased proportionally at the fractured surface because failure was liable to develop at this portion. Moreover, the membrane structure might have prevented the in®ltration of the resin into the collagen web. Then, a nanoscale void within the hybrid layer [1,30,31] and the demineralized dentin zone beneath the hybrid layer could be readily created within the bond structure. Comparing the top of the hybrid layer and the demineralized dentin zone after bonding under dry conditions, the bond integrity between the bonding resin layer and the top of the hybrid layer might be weaker than the mechanical strength of the exposed collagen ®brils within the demineralized dentin. Therefore, it is possible that the failure of the top of the hybrid layer concealed the presence of the demineralized dentin zone on the fractured surface under morphological observation. In conclusion, at the time of the resin application, great care must be paid to the surface condition of the dentin after the acidi®cation that in¯uences both the short- and the long-term integrity of the resin±dentin bonds in the clinical situation. Acknowledgements This work was supported in part by a Grant-in-Aid for Scienti®c Research No. 11470401 from the Ministry of Education, Science and Culture, Japan. References [1] Hashimoto M, Ohno H, Endo K, Kaga M, Sano H, Oguchi H. The effect of hybrid layer thickness on bond strength: demineralized dentin zone of hybrid layer. Dent Mater 2000;16:406±11. [2] Suzuki M, Kato H, Wakumoto S. Vibrational analysis by Raman spectroscopy of the interface between dental adhesive resin and dentin. J Dent Res 1991;70:1092±7. [3] Van Meerbeek H, Mohrbacher H, Celis JP, Roos JR, Braem M, Lambrechts P, Vanherle G. Chemical characterization of the resin± dentin interface by micro-Raman spectroscopy. J Dent Res 1993; 72:1423±8. [4] Kanca 3rd J. Improving bond strength through acid etching of dentin and bonding to wet dentin surfaces. J Am Dent Assoc 1992;123:35± 43. [5] Nakajima M, Sano H, Zheng L, Tagami M, Pashley DH. Effect of moist versus dry bonding to normal vs. caries-affected dentin with Scotchbond Multi-Purpose Plus. J Dent Res 1999;78:1298±303. [6] PerdigaÄo J, Van Meerbeek B, Lopes MM, Ambrose WW. The effect of a re-wetting agent on dentin bonding. Dent Mater 1999;15:285±95. [7] Nakabayashi N, Watanabe A, Arao T. A tensile test to facilitate identi®cation of defects in dentine bonded specimens. J Dent 1998; 26:379±85. [8] Spencer P, Swafford JR. Unprotected protein at the dentin±adhesive interface. Quint Int 1999;30:501±7. [9] Hashimoto M, Ohno H, Kaga M, Endo K, Sano H, Oguchi H. In vivo
102
[10] [11] [12] [13] [14]
[15] [16] [17] [18] [19] [20]
M. Hashimoto et al. / Dental Materials 18 (2002) 95±102 degradation of resin±dentin bonds in humans over 1 to 3 years. J Dent Res 2000;79:1385±91. Burrow MF, Satoh M, Tagami J. Dentin bond durability after three years using a dentin bonding agent with and without priming. Dent Mater 1996;12:302±7. Kato G, Nakabayashi N. The durability of adhesion to phosphoric acid etched, wet dentin substrates. Dent Mater 1998;14:347±52. Kitasako Y, Burrow MF, Nikaido T, Tagami J. The in¯uence of storage on dentin bond durability of resin cement. Dent Mater 2000;16:1±6. Shono Y, Ogawa T, Terashita M, Carvalho RM, Pashley EL, Pashley DH. Regional measurement of resin±dentin bonding as an array. J Dent Res 1999;78:699±705. Sano H, Shono T, Sonoda H, Takatsu T, Ciucchi B, Carvalho RM, Pashley DH. Relationship between surface area for adhesion and tensile bond strength. Evaluation of a micro-tensile bond test. Dent Mater 1994; 10:236±40. Pashley DH, Sano H, Ciucchi B, Yoshiyama M, Carvalho RM. Adhesion testing of dentin bonding agents: a review. Dent Mater 1995; 11:117±25. Davidson CL, Abdalla AL, de Gee AJ. An investigation into the quality of dentine bonding systems for accomplishing a durable bond. J Oral Rehabil 1993;20:291±300. Versluis A, Tantbirojn D, Douglas WH. Why do shear bond tests pull out dentin? J Dent Res 1997;76:1298±307. Tantbirojn D, Cheng YS, Versluis A, Hodhges JH, Douglas WH. Nominal shear or fracture mechanics in the assessment of composite±dentin adhesion? J Dent Res 2000;79:41±48. Van Noort K, Cardew GE, Howard IC, Noroozi S. The effect of local interfacial geometry on the measurement of the tensile bond strength to dentin. J Dent Res 1991;70:889±93. Pereira PNR, Okuda M, Sano H, Yoshikawa T, Burrow MF, Tagami J. Effect of intrinsic wetness and regional difference on dentin bond strength. Dent Mater 1999;15:46±53.
[21] Yoshikawa T, Sano H, Burrow MF, Tagami J, Pashley DH. Effects of dentin depth and cavity con®guration on bond strength. J Dent Res 1999;78:898±905. [22] Nakajima M, Sano H, Burrow MF, Tagami M, Yoshiyama M, Ebisu S, Ciucchi B, Russel CM, Pashley DH. Tensile bond strength and SEM evaluation of caries-affected dentin using dentin adhesives. J Dent Res 1995;74:1679±88. [23] Van Meerbeek B, Yoshida Y, Lambrechts P, Vanherle G, Duke ES, Eick JD, Robinson SJ. A TEM study of two water-based adhesive systems bonded to dry and wet dentin. J Dent Res 1998;77:50±59. [24] Tay FR, Gwinnet JA, Wei SHY. Micromorphological spectrum of acid-conditioned dentin following the application of a water-based adhesive. Dent Mater 1998;14:329±38. [25] Yoshida Y, Van Meerbeek B, Snauwaert J, Hellemans L, Lambrechts P, Vanherle G, Wakasa K, Pashley DH. A novel approach to AFM characterization of adhesive tooth±biomaterial interfaces. J Biomed Mater Res 1999;47:85±90. [26] Pashley DH, Ciucchi B, Yoshiyama M, Carvalho RM. Permeability of dentin to adhesive agents. Quint Int 1995;24:618±31. [27] PerdigaÄo J, Lambrechts P, Van Meerbeek B, Tome AR, Vanherle G, Lopes AB. Morphological ®eld emission-SEM study of the effect of six phosphoric acid etching agents on human dentin. Dent Mater 1996;12:262±71. [28] Maciel KT, Carvalho RM, Ringle RD, Preston CD, Russell CM, Pashley DH. The effects of acetone, ethanol, HEMA, and air on the stiffness of human decalci®ed dentin matrix. J Dent Res 1996;75: 1851±8. [29] Pashley DH, Zhang Y, Agee KA, Rouse CJ, Carvalho RM, Russell CM. Permeability of demineralized dentin to HEMA. Dent Mater 2000;16:7±14. [30] Sano H, Shono T, Takatsu T, Hosoda H. Microporous dentin zone beneath resin-impregnated layer. Oper Dent 1994;19:59±64. [31] Li H, Burrow MF, Tyas MJ. Nanoleakage patterns of four dentin bonding systems. Dent Mater 2000;16:48±56.