Di

AKMHa

b

c

Sd

D

a

A

R

R

A

K

M

A

R

T

S

H

B

h0

d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 1147–1153

Available online at www.sciencedirect.com

ScienceDirect

jo ur nal home p ag e: www.int l .e lsev ierhea l th .com/ journa ls /dema

entin-smear remains at self-etch adhesiventerface

tsushi Minea,b, Jan De Muncka, Marcio Vivan Cardosoa,irsten L. Van Landuyta, André Poitevina, Annelies Van Endea,ariko Matsumotob, Yasuhiro Yoshidac, Takuo Kubokid,irofumi Yatanib, Bart Van Meerbeeka,∗

KU Leuven BIOMAT, Department of Oral Health Sciences, KU Leuven (University of Leuven), Leuven, BelgiumDepartment of Fixed Prosthodontics, Osaka University Graduate School of Dentistry, Osaka, JapanDepartment of Biomaterials and Bioengineering, Hokkaido University Graduate School of Dental Medicine,apporo, JapanDepartment of Oral Rehabilitation and Regenerative Medicine, Okayama University Graduate School of Medicine,entistry and Pharmaceutical Sciences, Okayama, Japan

r t i c l e i n f o

rticle history:

eceived 4 April 2014

eceived in revised form 6 July 2014

ccepted 11 July 2014

eywords:

ild self-etch

dhesion

esin-smear complex

EM

mear layer

ybrid layer

a b s t r a c t

Objective. The bonding potential of ‘mild’ self-etch adhesives may be compromised due to

smear interference, as they may not dissolve/penetrate the smear layer effectively due to

their relatively low acidity. We observed that the thickness of the dentin smear layer differed

depending on the surface-preparation methodology used.

Methods. The interaction of an (ultra-)mild self-etch adhesive (Clearfil S3 Bond, Kuraray

Noritake) with human dentin, prepared either using a medium-grit diamond bur (‘thick’,

clinically relevant smear layer) or 600-grit SiC-paper (‘thin’ smear layer), or just fractured

(smear-free), was evaluated using high-resolution transmission electron microscopy (TEM).

Non-demineralized/demineralized 30–100 nm interfacial cross-sections were prepared fol-

lowing common TEM-specimen processing and diamond-knife ultra-microtomy.

Results. The adhesive did not dissolve the bur-cut, nor the SiC-ground smear layer, but

impregnated it. Within this ‘resin-smear complex’, hydroxyapatite was abundantly present.

At fractured dentin, this complex was not present, while the actual layer of interaction of the

adhesive was limited to about 100 nm. Non-demineralized ‘ultra-thin’ (30–50 nm) sections

confirmed the interfacial ultra-structure to differ for the three surface-preparation meth-

ods. An electron dense band was consistently disclosed at the adhesive interface, most likely

representing the documented chemical interaction of the functional monomer 10-MDP with

Ca.

Significance. The dentin surface-preparation method significantly affects the nature of the

smear layer and the intera

© 2014 Academy

∗ Corresponding author at: KU Leuven BIOMAT, Department of Oral Hea-3000 Leuven, Belgium. Tel.: +32 16 33 75 87; fax: +32 16 33 27 52.

E-mail addresses: bart.vanmeerbeek@med.kuleuven.be, bart.vanmettp://dx.doi.org/10.1016/j.dental.2014.07.006109-5641/© 2014 Academy of Dental Materials. Published by Elsevier L

ction with the ultra-mild self-etch adhesive.

of Dental Materials. Published by Elsevier Ltd. All rights reserved.

lth Sciences, KU Leuven (University of Leuven), Kapucijnenvoer 7,

erbeek@uzleuven.be (B. Van Meerbeek).

td. All rights reserved.

s 3 0

1148 d e n t a l m a t e r i a l

1. Introduction

In many laboratory studies investigating dentin bondingeffectiveness, SiC-ground surfaces are used, for reasons ofstandardization and ease of preparation [1–3]. Nevertheless,a smear layer created during clinical procedures with a dia-mond or tungsten bur is rather thick and compact, and somay compromise the bonding effectiveness to dentin, in par-ticular when a self-etch adhesive is used [4]. A thick smearlayer may also be porous and thus have weak mechanicalproperties [5]. In this respect, the documented compromiseddentin bonding effectiveness of contemporary one-step self-etch adhesives, which combine conditioning, priming andapplication of adhesive resin, could to a certain extent beattributed to interference of bur debris smeared across dentinduring cavity preparation. However, there is very little detailedmorphological data on the effect of smear interposition atthe adhesive-dentin interface produced by so-called ‘ultra-mild’ self-etch adhesives, which are significantly less acidic(pH ≈ 2.7). Especially completeness of resin penetration andadequate resin envelopment of potential residual smear ishighly important, as insufficient resin impregnation mayresult in more rapid interface degradation [6]. Therefore,high-resolution ultra-morphological examination is thoughtto give a better insight what impact potential interfacialinterference of bur smear may have on the bond durability[7].

The purpose of this study was to examine the effect ofdifferent surface preparation methods on the interfacial ultra-structure of an ultra-mild self-etch adhesive bonded to dentin.For this, high-resolution transmission electron microscopy(TEM) is the method of choice, as the combination of resolutionand image detail is unsurpassed.

2. Materials and methods

Non-carious human third molars were stored in 0.5% chlo-ramine solution at 4 ◦C and used within one month afterextraction. The teeth were randomly divided into 3 groups:‘BUR-CUT’, ‘SiC-GROUND’ and ‘FRACTURED’ dentin (seebelow). All teeth were mounted in gypsum blocks in order toease manipulation. For ‘BUR-CUT’ dentin, the occlusal thirdof the crown was removed at the level of mid-coronal dentinusing a slow-speed diamond saw (Isomet 1000, Buehler, LakeBluff, IL, USA), after which a standard smear layer was pro-duced by a medium-grit (100 �m) diamond bur (842, Komet,Lemgo, Germany) using a water-cooled high-speed contra-angle hand piece mounted in the MicroSpecimen Former (TheUniversity of Iowa, Iowa City, IA, USA). Half of the bur-cutspecimens were further processed by wet-sanding with 600-grit silicone-carbide paper (‘SiC-GROUND’). For ‘FRACTURED’(smear-free) dentin, a shallow 1–2 mm deep groove was cutcircumferentially around the tooth at the level of mid-coronaldentin, after which the coronal part was fractured using a

forceps to produce a fractured dentin surface free of smeardebris. All dentin surfaces were carefully verified for absenceof enamel and/or pulp tissue using a stereo-microscope (WildM5A, Wild Heerbrugg AG, Heerbrugg, Switzerland).

( 2 0 1 4 ) 1147–1153

The ultra-mild one-step self-etch adhesive Clearfil S3Bond (Kuraray Noritake, Tokyo, Japan) was applied accord-ing to the manufacturer’s instructions, followed by a thinlayer of flowable composite (Clearfil Protect Liner F, KurarayNoritake). Light-curing was performed using an Optilux500 (Demetron/Kerr, Danbury, CT, USA) device with a lightoutput not less than 600 mW/cm2. After bonding proce-dures, specimens were stored for 1 day in distilled waterat 37 ◦C. The specimens were processed according to theprocedure described in detail by Van Meerbeek et al. [8]. Non-demineralized and laboratory-demineralized thin (60–100 nm)and non-demineralized ultra-thin (30–50 nm) sections werecut (Ultracut UCT, Leica, Vienna, Austria) and examinedunstained and positively stained (3% uranyl acetate for12 min/lead citrate for 13 min) using TEM (JEM-1200EX II,JEOL, Tokyo, Japan). In order to reveal potential porous zonesat the interface, additional specimens were immersed in a50 wt% ammoniacal silver nitrate solution according to ananoleakage-detection protocol previously described by Tayet al. [9].

3. Results

TEM of non-demineralized sections revealed clear differencesin substrate irregularity and smear-layer thickness for the dif-ferent surface-preparation methods applied. At FRACTUREDdentin, a smear layer could not be observed, but distinctresin tags were clearly formed (Fig. 1a and b). At SiC-GROUNDdentin, the interface appeared relatively flat and smooth witha regularly thin smear layer (up to 1 �m) deposited by theshearing/pushing motion during grinding (Fig. 2a and b). Dueto the smear plugs in the dentin tubules, resin tags wereclearly less prominent than at fractured dentin. At BUR-CUTdentin, the surface was clearly more irregular with a muchthicker (up to 5 �m), but also much more in thickness vary-ing smear layer deposited (Fig. 3a and b). The adhesive wasnot able to visibly dissolve the underlying dentin substrateat FRACTURED dentin (Fig. 1b), nor able to dissolve the SiC-GROUND (Fig. 2b) and the BUR-CUT smear layer (Fig. 3b).In the two latter conditions, a ‘resin-smear complex’ wasformed, in which hydroxyapatite (HAp), often fragmented, wasabundantly present. AgNO3-immersed specimens revealedrelatively small deposits of silver within the adhesive and atthe adhesive-dentin interface at FRACTURED dentin (Fig. 1c),most likely due to absence of surface smear. In contrast, avarying pattern of both spot- and cluster-like appearance ofnano-leakage was revealed within the resin-smear complexand within dentin for the SiC-GROUND or BUR-CUT dentinspecimens.

The inorganic component of dentin, and especiallyperitubular dentin, was demineralized by the 36-h lab-oratory demineralization (formic acid, formaldehyde).Laboratory-demineralized sections more clearly revealedthe resin-tag structure (Figs. 1d and 2d) and smear-layerthickness (Figs. 2d and 3d), as compared to what could

be observed using non-demineralized sections. Moreover,laboratory-demineralized sections disclosed better thedifference between resin-impregnated smear and acid-resistant hybridized dentin. At FRACTURED dentin, dentin

d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 1147–1153 1149

Fig. 1 – TEM photomicrographs of an ‘ultra-mild’ self-etch adhesive bonded to FRACTURED dentin. (a) Non-demineralized,unstained section, disclosing a tight, void-free interface and distinct resin-tags (black arrowheads). (b) Correspondinghigher magnification of (a): no morphologic features of interaction or demineralization were revealed in thisnon-demineralized section at this high magnification (original magnification = 100 k). Hydroxyapatite (HAp) is abundantlypresent (white arrowheads). (c) AgNO3-immersed specimens for nano-leakage screening revealed relatively few deposits ofsilver within the adhesive (arrow #1) and at the adhesive-dentin interface (arrow #2). (d) Demineralized, unstained section,illustrating a 10-�m long resin tag. (e) Demineralized, stained section: a clear electron dense hybrid layer (HL) was detected,r of ab

haigir(wpBcr

sbNntomiv

epresenting an acid-resistant hybrid layer with a thickness

ybridization was mostly limited to a depth ranging up to few hundreds of nanometres, while in most areas thenfiltration did not extend beyond the width of one colla-en fibril (about 100 nm) (Fig. 1e). Clearly, a much thickernteraction layer, consisting of a true hybrid layer and aesin-smear complex on top, was observed at SiC-GROUNDFig. 2e) and BUR-CUT dentin (Fig. 3e). Collagen fibrilsere hardly observed within the resin-smear complex. Aartial loss of the resin-smear complex was observed atUR-CUT dentin (Fig. 3e), suggesting that the resin-smearomplex contained areas that were not fully impregnated byesin.

Ultra-thin sections clearly confirmed the difference inubstrate irregularity and HAp presence, as was disclosedy the conventional thickness non-demineralized sections.evertheless, although these ultra-thin sections were alsoon-demineralized sections, they appeared somewhat similar

o laboratory-demineralized sections regarding the absence

f peritubular dentin and the resin-tag structure. The ultra-orphological differences were, however, more pronounced

n the ultra-thin sections than in case of both the con-entional non-demineralized and laboratory-demineralized

out the width of one collagen fibril.

sections (Fig. 4). The twofold hybrid layer/resin-smear com-plex and the partial loss of the resin-smear complex atBUR-CUT dentin (Fig. 4f) could be distinguished without anystaining.

4. Discussion

Three types of dentin surface-preparation methods were usedand their effect on the interfacial interaction of a represen-tative 10-MDP-based ‘ultra-mild’ one-step self-etch adhesive(Clearfil S3 Bond, Kuraray Noritake; pH ≈ 2.7) with dentinwas ultra-morphologically characterized using TEM. Previ-ously, a so-called ‘nano-interaction zone’ without distinctdemineralization was described [10]. The thickness of sucha ‘nano-interaction zone’ was measured to be below 300 nm,when an experimental 10-MDP-based precursor of Clearfil S3Bond (SSB-200, Kuraray Noritake) was employed. In that study,

solely non-demineralized unstained sections were examined.We additionally prepared and examined in this study stainedlaboratory-demineralized and ultra-thin non-demineralizedwhich enabled to more clearly reveal this nano-scale

1150 d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 1147–1153

Fig. 2 – TEM photomicrographs of an ‘ultra-mild’ self-etch adhesive bonded to SiC-GROUND dentin. (a) Non-demineralized,unstained section, disclosing a tight interface. No clear resin tag was observed, compared to what was observed at fractureddentin, most likely due to the presence of a smear plug (black arrowhead). (b) Corresponding higher magnification of (a): noclear morphologic features of interaction or demineralization were revealed (original magnification = 100 k). HAp wasabundantly present (white arrowheads). (c) AgNO3-immersed specimens revealed relatively small deposits of silver withinthe adhesive (arrow #1), in the smear layer (arrow #2) and inside dentin underneath (arrow #3). (d) Demineralized,unstained section: porous resin tags were observed, most likely because of demineralization of residual smear particles. (e)Demineralized, stained section: a clearly electron dense hybrid layer (HL) can be detected, with on top a resin-smear

sin.

complex (RSC), consisting of residual smear enveloped by re

interaction with dentin and thus to investigate potentialsmear interference.

As nor the SiC-GROUND, nor the BUR-CUT smear layerwas dissolved, but impregnated by the adhesive, a ‘resin-smear complex’ was produced, which we have describedbefore in our previous study regarding the interaction ofthe adhesive with likewise differently prepared enamel sur-faces [11]. Hydroxyapatite was abundantly present within thisresin-smear complex, that differed in thickness and unifor-mity depending on the surface-preparation methodology used(Figs. 2a,b and 3a,b). The increasing gradation of HAp crys-tals with depth, as Koshiro et al. reported [10], could not beobserved to the same degree in our study.

The underlying mechanism of bonding of mild self-etch adhesives was shown to be based upon submicronmicro-mechanical interlocking [12], supplemented by primarychemical interaction of the functional monomer 10-MDP with

HAp that remained around the partially exposed collagen[13,14]. According to the ‘Adhesion-Decalcification concept’(AD-concept), 10-MDP chemically bonds to HAp and therebykeeps HAp as the natural shelter around collagen [15]. While

The collagen fibril structure can hardly be seen in the RSC.

such chemical interaction did not raise the ‘immediate’ bondstrength, studies testing the biodegradation resistance ofadhesive interfaces have shown that it improved the bond sta-bility [16,17]. Additionally, the demineralization capability ofultra-mild self-etch adhesives is limited; the adhesive showedhardly any sign of demineralization at FRACTURED dentin(Fig. 1a and b).

The nano-scale interaction with dentin was clearlyevidenced by the stained laboratory-demineralized sections.The densely stained layer observed at FRACTURED dentin, waslimited to about 100 nm (Fig. 1e). Since this substrate lackeda smear layer, the densely stained layer must be identifiedas ‘true’ hybrid layer. The much thicker densely stained layerobserved at SiC-GROUND (Fig. 2e) and BUR-CUT dentin (Fig. 3e)must thus be attributed to resin having infiltrated residualsurface smear. The resultant ‘resin-smear complex’ is pos-itioned on top of the true hybrid layer. Some authors have

already before speculated that separation of the hybridizedsmear layer from the true hybrid layer may occur [18,19]. Inthe event that the underlying dentin is demineralized andhybridized, there is a further concern that the only material

d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 1147–1153 1151

Fig. 3 – TEM photomicrographs of an ‘ultra-mild’ self-etch adhesive bonded to BUR-CUT dentin. (a) Non-demineralized,unstained section: the surface irregularity and thickness of the smear layer (SL) differed regionally. The thickness of thesmear layer varied from almost 0 to 3 �m. (b1) Corresponding higher magnification of (a), disclosing a ‘shallow’ smear layerof less than 1 �m. Hydroxyapatite was abundantly present (white arrowheads). (b2) Corresponding higher magnification of(a), disclosing a ‘thick’ smear layer of more than 4 �m. (c) AgNO3-immersed specimens revealed a varying pattern of bothspot- and cluster-like appearance of nano-leakage within the adhesive (arrow #1), in the smear layer (arrow #2) and insidedentin underneath (arrow #3). Only a few spots of silver deposition were found in the resin-smear complex (RSC). (d)Demineralized, unstained section: the irregularity of the smear layer was more evident. (e1) Demineralized, stained section,disclosing a ‘shallow’ smear layer. A clearly electron dense hybrid layer (HL) became visible, with on top a resin-smearcomplex, consisting of residual smear enveloped by resin. (e2) Demineralized, stained section, disclosing a ‘thick’ smearlayer. Not only the thickness of the resin-smear complex, but also that of the hybrid layer was different from that atS sin-

cigiwsiiHrTrri

amo

iC-GROUND dentin (see Fig. 2e). A partial loss within the re

onnecting the resin-smear complex with the true hybrid layers resin that had to diffuse around the globular particle aggre-ates within the actual smear layer, and then onwards into thenterfibrillar spaces of the underlying intertubular dentin, asell as into the dentinal tubules, the latter forming hybridized

mear plugs. However, given the abundant presence of HApn this resin-smear complex and the effective chemical bind-ng potential of the 10-MDP functional monomer with thisAp, it is very likely that these residual smear particles may

einforce the resin-smear complex. Moreover, for none of ourEM sections observed, the hybrid layer separated from theesin-smear complex, corroborating our hypothesis that wellesin-enveloped smear globules may reinforce the adhesiventerface.

At FRACTURED dentin, the acid-resistant hybrid layer,s observed in the laboratory-demineralized sections, wasostly limited to a depth ranging up to a few hundreds

f nanometers, while in most areas the infiltration did not

smear complex was observed (hand point).

extend beyond the width of one collagen fibril (about 100 nm).The micro-fibrillar arrangement of collagen type I, whichresults from the pentameric staggered organization of themolecules in a higher order supra-molecular supertwistednanostructure, has been shown to result in dentinal collagenwith an intermolecular spacing of about 1.3 nm. This zone istoo small to accommodate one single molecule of commondental monomers (∼2 nm) [20]. Therefore, monomers can onlyinfiltrate more deeply in case of considerable demineralizationor fragmentation as within the smear layer.

There is a common consensus that the final goal of dentinbonding is complete infiltration of resin monomers into dem-ineralized collagen fibrils exposed by separate acid-etching orby self-etch adhesives [21]. Any discrepancy between dentin

demineralization and resin infiltration may result in silver-nitrate- or fluorescent-dye-detectable nanoleakage withinwater-rich zones of the hybrid layer [22,23], and the adhe-sive layer [24]. A partial loss of the resin-smear complex was

1152 d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 1147–1153

Fig. 4 – TEM photomicrographs of non-demineralized, non-stained ‘ultra-thin’ sections of the interface between anultra-mild self-etch adhesive and dentin covered with different smear layers. (a) Overview image of the adhesive–dentininterface at FRACTURED dentin, disclosing a tight, void-free interface. A clear resin-tag was observed (black arrowhead). Thehardest dentinal tissue, namely peritubular dentin, was absent in this thin section (asterisks). (b) Overview image of theadhesive–dentin interface at SiC-GROUND dentin, disclosing more porous resin tags most likely due to residual smear(black arrowheads). Peritubular dentin could not be observed in this thin section (asterisks). (c) Overview image of theadhesive–dentin interface at BUR-CUT dentin, disclosing a smear layer (SL) regionally differing in thickness. (d) Highermagnification of the interface at FRACTURED dentin, disclosing HAp crystals scattered throughout the hybrid layer anddentin (white arrowheads). A more electron dense band was consistently observed, which might represent the interactionof the functional monomer 10-MDP with Ca. (e) Higher magnification of the interface at SiC-GROUND dentin, disclosing aresin-smear complex (RSC) on top of the hybrid layer (HL). (f) Higher magnification of the interface at BUR-CUT dentin,disclosing a fragile part of the resin-smear complex with some voids (hand pointer) that may represent areas not infiltrated

pare

well by resin. The true hybrid layer is not as distinct, as com

observed in this study at BUR-CUT dentin (Fig. 3e); areas withinthe resin-smear complex were identified not to have beenadequately filled with resin. This is perfectly in line withour nano-leakage assessment, where the adhesive resin wasable to infiltrate the SiC-GROUND dentin surface more effec-tively than the BUR-CUT surface. The silver-tracer methodthat has been widely employed as adhesive interface evalua-tion method for the past 15 years, actually underestimates theretention of water within apparently well-infiltrated collagenmatrices in hybrid layers after bonding with etch-and-rinseand self-etch adhesives. On the other hand, using biomimetichybrid-layer remineralization techniques a progressive reduc-tion in silver-tracer uptake over time and, hence, in waterdistribution within adhesive-dentin interfaces was reported

for etch-and-rinse hybrid layers [25]. Moreover, besides incom-plete resin infiltration, such regions of silver deposition mayalso represent sites of retained water [26], especially in case of

d to that observed at FRACTURED and SiC-GROUND dentin.

HEMA-containing adhesives that tend to absorb water moreeasily [27]. Consequently, it is not yet evident whether the par-tial loss of the resin-smear complex (Fig. 3e) and the fragilepart (Fig. 4f) at BUR-CUT dentin are of significant concern withregard to bond durability.

TEM of ultra-thin sections revealed clear differencesin terms of substrate irregularity, smear-layer thickness,resin tag/smear plug formation and hybrid layer/resin-smearcomplex ultrastructure for the three surface-preparationmethods applied. The observations are complementary tothose of conventional thickness non-demineralized andlaboratory-demineralized TEM sections, as especially the trueinteraction with underlying dentin can be disclosed, whichusing conventional thickness sections requires laboratory-

demineralization and staining. However, as these ultrathinTEM sections were not demineralized, the crystal structureat the interface can still be detected. Since these sections

0 ( 2

wot

stbtoshcaam

C

Tr

r

d e n t a l m a t e r i a l s 3

ere not stained, we hypothesize that the electron dense partbserved at the interaction layer, represents the interaction ofhe functional monomer 10-MDP with Ca.

It is concluded that the dentin surface-preparation methodignificantly affects the nature of the smear layer and thushe interaction of in particular (ultra-)mild self-etch adhesives,eing more uniform at SiC-GROUND dentin as compared tohe more clinically relevant BUR-CUT dentin surface. In lightf the increased clinical use of (ultra-)mild self-etch adhe-ives, the effect of different surface-preparation methods (e.g.and excavation, bur preparation, laser ablation, air-abrasion,hemo-mechanical methods, air-polishing, ultrasonicationnd sono-abrasion) on the surface receptiveness of dentin fordhesion using self-etch adhesives needs to be investigatedore.

onflicts of interest

he authors declare that they have no conflicts of interestegarding the products investigated.

e f e r e n c e s

[1] Amano S, Yamamoto A, Tsubota K, Rikuta A, Miyazaki M,Platt JA, et al. Effect of thermal cycling on enamel bondstrength of single-step self-etch systems. Oper Dent2006;31:616–22.

[2] Ando S, Watanabe T, Tsubota K, Yoshida T, Irokawa A,Takamizawa T. Effect of adhesive application methods onbond strength to bovine enamel. J Oral Sci 2008;50:181–6.

[3] Watanabe T, Tsubota K, Takamizawa T, Kurokawa H, RikutaA, Ando S, et al. Effect of prior acid etching on bondingdurability of single-step adhesives. Oper Dent2008;33:426–33.

[4] Senawongse P, Srihanon A, Muangmingsuk A, HarnirattisaiC. Effect of dentine smear layer on the performance ofself-etching adhesive systems: a micro-tensile bondstrength study. J Biomed Mater Res B: Appl Biomater2010;94:212–21.

[5] Kenshima S, Reis A, Uceda-Gomez N, Tancredo Lde L, FilhoLE, Nogueira FN, et al. Effect of smear layer thickness and pHof self-etching adhesive systems on the bond strength andgap formation to dentin. J Adhes Dent 2005;7:117–26.

[6] Sano H, Takatsu T, Ciucchi B, Horner JA, Matthews WG,Pashley DH. Nanoleakage: leakage within the hybrid layer.Oper Dent 1995;20:18–25.

[7] Van Meerbeek B, Vargas M, Inoue S, Yoshida Y, Perdigao J,Lambrechts P, et al. Microscopy investigations. Techniques,results, limitations. Am J Dent 2000;13:3–18.

[8] Van Meerbeek B, Yoshida Y, Lambrechts P, Vanherle G, DukeES, Eick JD, et al. A TEM study of two water-based adhesivesystems bonded to dry and wet dentin. J Dent Res

1998;77:50–9.

[9] Tay FR, Pashley DH, Yoshiyama M. Two modes ofnanoleakage expression in single step adhesives. J Dent Res2002;81:472–6.

0 1 4 ) 1147–1153 1153

[10] Koshiro K, Sidhu SK, Inoue S, Ikeda T, Sano H. New conceptof resin–dentin interfacial adhesion: the nanointeractionzone. J Biomed Mater Res B: Appl Biomater 2006;77:401–8.

[11] Mine A, De Munck J, Vivan Cardoso M, Van Landuyt KL,Poitevin A, Kuboki T, et al. Enamel-smear compromisesbonding by mild self-etch adhesives. J Dent Res2010;89:1505–9.

[12] Van Meerbeek B, De Munck J, Yoshida Y, Inoue S, Vargas M,Vijay P, et al. Buonocore memorial lecture. Adhesion toenamel and dentin: current status and future challenges.Oper Dent 2003;28:215–35.

[13] Yoshida Y, Nagakane K, Fukuda R, Nakayama Y, Okazaki M,Shintani H, et al. Comparative study on adhesiveperformance of functional monomers. J Dent Res2004;83:454–8.

[14] Fu B, Sun X, Qian W, Shen Y, Chen R, Hannig M. Evidence ofchemical bonding to hydroxyapatite by phosphoric acidesters. Biomaterials 2005;26:5104–10.

[15] Yoshida Y, Van Meerbeek B, Nakayama Y, Yoshioka M,Snauwaert J, Abe Y, et al. Adhesion to and decalcification ofhydroxyapatite by carboxylic acids. J Dent Res2001;80:1565–9.

[16] Inoue S, Koshiro K, Yoshida Y, De Munck J, Nagakane K,Suzuki K, et al. Hydrolytic stability of self-etch adhesivesbonded to dentin. J Dent Res 2005;84:1160–4.

[17] Erhardt MC, Pisani-Proenca J, Osorio E, Aguilera FS, ToledanoM, Osorio R. Influence of laboratory degradation methodsand bonding application parameters on microTBS ofself-etch adhesives to dentin. Am J Dent 2011;24:103–8.

[18] Watanabe I, Nakabayashi N, Pashley DH. Bonding to grounddentin by a phenyl-P self-etching primer. J Dent Res1994;73:1212–20.

[19] Van Landuyt KL, De Munck J, Mine A, Cardoso MV, PeumansM, Van Meerbeek B. Filler debonding & subhybrid-layerfailures in self-etch adhesives. J Dent Res 2010;89:1045–50.

[20] Bertassoni LE, Orgel JP, Antipova O, Swain MV. The dentinorganic matrix – limitations of restorative dentistry hiddenon the nanometer scale. Acta Biomater 2012;8:2419–33.

[21] Breschi L, Mazzoni A, Ruggeri A, Cadenaro M, Di Lenarda R,De Stefano Dorigo E. Dental adhesion review: aging andstability of the bonded interface. Dent Mater 2008;24:90–101.

[22] Reis AF, Giannini M, Pereira PN. Long-term TEM analysis ofthe nanoleakage patterns in resin–dentin interfacesproduced by different bonding strategies. Dent Mater2007;23:1164–72.

[23] Sauro S, Watson TF, Mannocci F, Miyake K, Huffman BP, TayFR, et al. Two-photon laser confocal microscopy ofmicropermeability of resin–dentin bonds made with wateror ethanol wet bonding. J Biomed Mater Res B: ApplBiomater 2009;90:327–37.

[24] Tay FR, Pashley DH. Water treeing – a potential mechanismfor degradation of dentin adhesives. Am J Dent 2003;16:6–12.

[25] Kim J, Arola DD, Gu L, Kim YK, Mai S, Liu Y, et al. Functionalbiomimetic analogs help remineralize apatite-depleteddemineralized resin-infiltrated dentin via a bottom-upapproach. Acta Biomater 2010;6:2740–50.

[26] Van Meerbeek B. The myth of nanoleakage. J Adhes Dent

2007;9:491–2.

[27] Van Landuyt KL, Snauwaert J, Peumans M, De Munck J,Lambrechts P, Van Meerbeek B. The role of HEMA in one-stepself-etch adhesives. Dent Mater 2008;24:1412–9.