RESEARCH REPORTS
Biomaterials & Bioengineering
Y. Yoshida1,2, K. Nagakane3, R. Fukuda4,
Y. Nakayama5, M. Okazaki3, H. Shintani4,
S. Inoue6, Y. Tagawa7, K. Suzuki1,2,
J. De Munck8, and B. Van Meerbeek8*
1Department of Biomaterials, Okayama University Graduate
School of Medicine and Dentistry, 2-5-1 Shikata-cho,
Okayama 700-8525, Japan; 2Research Center for Biomedical
Engineering, Okayama University, 2-5-1 Shikata-cho,
Okayama 700-8525, Japan; 3Department of Biomaterials
Science, Hiroshima University Graduate School of Bio-
materials Science, 1-2-3 Kasumi, Minami-Ku, Hiroshima 734-
8553, Japan; 4Department of Operative Dentistry and Dental
Materials, Hiroshima University Graduate School of
Biomaterials Science, 1-2-3 Kasumi, Minami-Ku, Hiroshima
734-8553, Japan; 5Research Planning Department, Toray
Research Center, Inc., Sonoyama 3-3-7, Otsu, Shiga 520-8567,
Japan; 6Division for General Dentistry, Hokkaido University
Dental Hospital, Kita 13 Nishi 7, Kita-ku, Sapporo 060-8586,
Japan; 7Institute of Experimental Animals, Shinshu University
School of Medicine, 3-1-1 Asahi, Matsumoto, Nagano 390-
8621, Japan; and 8Leuven BIOMAT Research Cluster, Depart-
ment of Conservative Dentistry, School of Dentistry, Oral
Pathology and Maxillo-Facial Surgery, Catholic University of
Leuven, Kapucijnenvoer 7, B-3000 Leuven, Belgium; *cor-
responding author, bart.vanmeerbeek@med.kuleuven.ac.be
J Dent Res 83(6):454-458, 2004
ABSTRACT
Mild self-etch adhesives demineralize dentin only
partially, leaving hydroxyapatite around collagen
within a submicron hybrid layer. We hypothesized
that this residual hydroxyapatite may serve as a
receptor for chemical interaction with the functional
monomer and, subsequently, contribute to adhesive
performance in addition to micro-mechanical
hybridization. We therefore chemically characterized
the adhesive interaction of 3 functional monomers
with synthetic hydroxyapatite, using x-ray
photoelectron spectroscopy and atomic absorption
spectrophotometry. We further characterized their
interaction with dentin ultra-morphologically, using
transmission electron microscopy. The monomer 10-
methacryloxydecyl dihydrogen phosphate (10-MDP)
readily adhered to hydroxyapatite. This bond
appeared very stable, as confirmed by the low
dissolution rate of its calcium salt in water. The
bonding potential of 4-methacryloxyethyl trimellitic
acid (4-MET) was substantially lower. The monomer
2-methacryloxyethyl phenyl hydrogen phosphate
(phenyl-P) and its bond to hydroxyapatite did not
appear to be hydrolytically stable. Besides self-
etching dentin, specific functional monomers have
additional chemical bonding efficacy that is expected
to contribute to their adhesive potential to tooth tissue.
KEY WORDS: adhesion, functional monomer,
XPS, TEM, dissolution rate.
Received January 21, 2003; Last revision April 7, 2004;
Accepted April 14, 2004
454
Comparative Study
on Adhesive Performance
of Functional Monomers
INTRODUCTION
S elf-etch adhesives are user-friendly because they do not require a rinse
phase. This substantially reduces application time as well as technique
sensitivity. In addition to their application after a two- vs. a single-step
approach, a further distinction can be made between "mild" and "strong"
current self-etch adhesives (Inoue et al., 2000; Tay and Pashley, 2001; Van
Meerbeek et al., 2003). Like etch-and-rinse adhesives, strong self-etch
adhesives completely remove hydroxyapatite from dentin, resulting in
relatively deep dentin hybridization several micrometers thick. However,
mild self-etch adhesives form only submicron-thick hybrid layers, in which
hydroxyapatite partially remains around exposed collagen (Nakabayashi and
Saimi, 1996; Inoue et al., 2000).
Commonly, acidic monomers in self-etch primers/adhesives are esters
originating from the reaction of a bivalent alcohol with methacrylic acid and
phosphoric/carboxylic acid derivatives. Each self-etch adhesive contains its
specific functional monomer that, to a large extent, determines its actual
adhesive performance. Thus far, however, the interaction of functional
monomers with dental tissues has seldom been characterized by chemical
analytical techniques. We therefore comparatively characterized the
adhesive interaction of 3 functional monomers with synthetic
hydroxyapatite to test the hypothesis that the bonding mechanism of mild
self-etch adhesives involves chemical interaction of functional monomers
with residual hydroxyapatite in addition to micro-mechanical hybridization.
The functional monomers investigated originated from 3 representative two-
step self-etch adhesives, and their interaction with dentin was also ultra-
morphologically characterized.
MATERIALS & METHODS
The functional monomer 4-MET (4-methacryloxyethyl trimellitic acid) and its
calcium salt 4-METCa were provided by GC (Tokyo, Japan), whereas phenyl-P
(2-methacryloxyethyl phenyl hydrogen phosphate) and 10-MDP (10-
methacryloxydecyl dihydrogen phosphate), and their respective calcium salts
(phenyl-PCa, 10-MDPCa), were provided by Kuraray (Osaka, Japan).
X-ray Photoelectron Spectroscopy (XPS)
From each functional monomer, we prepared 15% (w/w) solutions including
45% (w/w) ethanol and 40% (w/w) water. Hydroxyapatite plates (APP-101,
Asahi Optical, Tokyo, Japan) were treated with each solution at 37°C for 30 sec
and 30 min, followed by ultrasonic rinsing twice in 52.9% (w/w) ethanol for 20
min prior to XPS analysis (AXIS-HS, Kratos, Manchester, UK) in vacuo of less
than 10-7 Pa. We used Al-K␣ monochromatic x-ray with a source power of 150
W. Wide and narrow scans were measured at a pass energy of, respectively, 80
and 40 eV. Quantitative data were obtained from peak areas, and identification
of chemical states was made from detailed measurement of peak positions and
separations. Significant differences in adhesive performance among the
functional monomers were analyzed by Student's t test (␣ = 0.01).
J Dent Res 83(6) 2004 Adhesive Properties of Functional Monomers 455

Dissolution Rate of Ca Salts

To determine the stability of potential chemical bonding, we
immersed the calcium salts of the acid monomers (0.5 g 4-METCa,
1.0 g phenyl-PCa, and 0.5 g 10-MDPCa) in 10 mL ultrapure water;
samples were then shaken for 1 wk (1 Hz, 37°C). The supernatant
liquid was used as a sample for experiments after insoluble
substances had been removed by centrifugation (3000 rpm, 10
min) and subsequently filtered through a polytetrafluoroethylene
membrane (pore size = 0.20 ␮m; Samprep-LCR25-LG, Millipore
Corporation, Bedford, MA, USA). The solution was then analyzed
for calcium by means of Atomic Absorption Spectrophotometry
(AAS; AA-670, Shimadzu, Kyoto, Japan). The dissolution rate was
quantified as the amount of calcium extracted from the calcium
salts in ultrapure water (n = 5). The data obtained were analyzed
by one-way ANOVA and Scheffé's multiple comparison test (␣ =
0.05).
Transmission Electron Microscopy (TEM)
Extracted non-carious human third molars (gathered after we
obtained informed consent according to a protocol approved by the
Commission for Medical Ethics of the Catholic University of
Leuven) were used within 1 mo of extraction (stored in 0.5%
chloramine/water, 4°C). After we removed the occlusal crown
third using an Isomet diamond saw (Isomet 1000, Buehler, Lake
Bluff, IL, USA), we wet-sanded the exposed dentin (60 sec, 600-
grit silicon-carbide paper) to produce a standard smear layer. All
specimens were randomly divided into 3 groups of 2 teeth each,
and were subjected to a bonding treatment according to the
manufacturer's instructions, with either UniFil Bond (UF-B, GC),
Clearfil Liner Bond 2 (C-LB2, Kuraray), or Clearfil SE Bond (C-
SE, Kuraray). UF-B, C-LB2, and C-SE contain 4-MET, phenyl-P,
and 10-MDP, respectively.
Following adhesive treatment, the resin-bonded dentin
specimens were processed for TEM according to a protocol
previously described (Van Meerbeek et al., 1998). Demineralized
samples were also prepared by immersion in a 10% formaldehyde-
formic acid solution (36 hrs). Sections (70-90 nm thick) were cut
by means of a diamond knife (Diatome, Bienne, Switzerland) in an
ultramicrotome (Ultracut UCT, Leica, Vienna, Austria), and were
observed unstained and positively stained (5% uranyl acetate [UA]
for 20 min/saturated lead citrate [LC] for 3 min) by TEM (Philips
CM10, Eindhoven, The Netherlands).

RESULTS
XPS wide-scan spectra of untreated hydroxyapatite and spectra
of hydroxyapatite treated with 4-MET, phenyl-P, and 10-MDP,
respectively, are alike, except for the C 1s peak at a binding
energy of approximately 285 eV that appeared when
hydroxyapatite was exposed to functional monomers (Fig. 1).
The intensity of the C 1s peak of untreated hydroxyapatite (Fig.
1a) increased slightly when hydroxyapatite was treated for 30
sec with 4-MET (Fig. 1b), while it increased considerably when Figure 1. XPS wide-scan spectra of untreated hydroxyapatite (a), of
hydroxyapatite treated with 15% (w/w) 4-MET for 30 sec (b), of
hydroxyapatite was exposed to 10-MDP for 30 sec (Fig. 1c). hydroxyapatite treated with 15% (w/w) 10-MDP for 30 sec (c), of
Application of 4-MET to hydroxyapatite for 30 min further hydroxyapatite treated with 15% (w/w) 4-MET for 30 min (d), of
increased the intensity of the C 1s peak in comparison with hydroxyapatite treated with phenyl-P for 30 min (e), and of
intensity from the 30-second application (Fig. 1d). A slightly hydroxyapatite treated with 10-MDP for 30 min (f).
increased C 1s peak appeared when hydroxyapatite was
exposed to phenyl-P for 30 min (Fig. 1e). No increase in
intensity of the C 1s peak was recorded when hydroxyapatite
was exposed to 10-MDP for 30 min (Fig. 1f), as compared with Application of 4-MET to hydroxyapatite (Fig. 2a) resulted in
the short 30-second application (Fig. 1c). a significant shift of the peak representing carboxyl groups and
456 Yoshida et al.
esters at 288.6 eV, to a lower binding energy and an increase of
its FWHM (full width at half-maximum). Deconvolution of the
shifted peak disclosed two components, representing esters and
unreacted carboxyl groups at 288.6 eV, and carboxyl groups
reacted with Ca of hydroxyapatite at 288.2 eV.
Application of phenyl-P to hydroxyapatite resulted in a
two-fold C 1s peak with a shape totally different from that of
phenyl-P itself (Fig. 2b). Application of 10-MDP to
hydroxyapatite resulted in a single C 1s peak, resembling that
of 10-MDP itself (Fig. 2c). Deconvolution disclosed: an intense
sub-peak at 284.6 eV, representing C-C/C-H/C=C bindings; a
second, much smaller peak at 286.0 eV, representing C-O
bindings; and the smallest peak at 288.6 eV, representing
esters.
With regard to chemical bonding efficacy, quantitative
J Dent Res 83(6) 2004
analysis revealed that the carbon concentration detected on
hydroxyapatite exposed to 15% (w/w) 4-MET for 30 min (17.2
+ 0.6%) was not significantly different (p > 0.05) from that of
hydroxyapatite exposed to 15% (w/w) 10-MDP for 30 min
(17.3 + 0.6%). However, 4-MET and 10-MDP have 15 and 14
carbon atoms, respectively, in a single molecule. Thus, for
comparison of bonding potentials, the atom% should be
adjusted by the number of carbon atoms in a single molecule.
The atom%/14 value for hydroxyapatite exposed to 10-MDP
(1.23 + 0.04%) was significantly (p < 0.05) higher than the
atom%/15 value of hydroxyapatite exposed to 4-MET (1.15 +
0.04%).
AAS (n = 5) revealed that phenyl-PCa was significantly
more soluble ([Ca 2+] in water = 1.91 + 0.14 g/L) than 4-
METCa ([Ca2+] in water = 1.36 + 0.27 g/L). The calcium salt
of 10-MDP (10-MDPCa) was the least soluble in water ([Ca2+]
in water = 6.79 + 0.43 mg/L).
TEM revealed that all 3 self-etch adhesives formed a
shallow (0.5-1 ␮m) hybrid layer (Fig. 3). Unstained, non-
demineralized sections revealed that all systems demineralized
dentin only partially, leaving hydroxyapatite around collagen
within the submicron hybrid layer. Staining disclosed a typical
hybrid-layer ultrastructure with cross-banded collagen
separated by electronlucent interfibrillar spaces. A typical 'shag
carpet' pattern often appeared at the top of the hybrid-layer.
Stained, demineralized sections confirmed the acid-resistance
of the hybrid layer.
DISCUSSION
High-resolution microscopic analysis showed that all 3 self-etch
adhesives hybridized dentin, indicating that their bonding
mechanism depended, at least in part, on micro-mechanical
interlocking of resin into a micro-porous surface (Nakayabashi
et al., 1982; Nakabayashi and Saimi, 1996). In contrast to etch-
and-rinse adhesives that involve phosphoric-acid etching, the
Figure 2. (a) XPS narrow-scan spectra of the C 1s region of 4-META (4-
methacryloxyethyl trimellitate anhydride) powder (top spectrum) and of
hydroxyapatite treated with 15% (w/w) 4-MET for 30 min (bottom
spectrum). Peak deconvolution revealed that almost all carbon
originated from 4-MET, with a peak at 284.6 eV representing C-C, C-H,
and C=C bindings, a peak at 286.1 eV representing C-O bindings, and
a peak representing ester and carboxyl groups. As compared with 4-
META powder (288.6 eV), the carboxyl peak had shifted to a lower
binding energy (288.3 eV). Deconvolution of the shifted peak disclosed
a peak at 288.6 eV, representing the ester function and unreacted
carboxyl groups, and a peak at 288.2 eV that represents carboxyl
groups that reacted with the Ca of hydroxyapatite. It should be
mentioned that the peak position and intensity of the C-O component
(top spectrum) for 4-META powder are slightly different from what
theoretically would be expected, and from the C-O subpeak in the
bottom spectrum. This must be attributed to impurities within the 4-META
powder sample that were readily removed during the adhesive
treatment. (b) XPS narrow-scan spectra of the C 1s region of untreated
phenyl-P (top spectrum) and of hydroxyapatite treated with 15% (w/w)
phenyl-P (bottom spectrum). (c) XPS narrow-scan spectra of the C1s
region of untreated 10-MDP (top spectrum) and of hydroxyapatite
treated with 15% (w/w) 10-MDP (bottom spectrum). Peak deconvolution
revealed that almost all carbon originated from 10-MDP on
hydroxyapatite, with a peak at 284.6 eV representing C-C, C-H, and
C=C bindings, a peak at 286.1 eV representing C-O bindings, and a
peak representing ester at 288.6eV. In high-resolution XPS C 1s spectra,
aliphatic carbon shows a large asymmetry due to vibrational effect,
while aromatic carbon is symmetrical (Beamson and Briggs, 1992).
Therefore, the peak at 284.6 eV representing C-C, C-H, and C=C
bindings of 10-MDP is expressed with an asymmetrical shape.
J Dent Res 83(6) 2004 Adhesive Properties of Functional Monomers 457

self-etch adhesives demineralized dentin only partially, leaving

hydroxyapatite partially attached to collagen. Since
hydroxyapatite remains available for interaction, the main
objective of this study was to characterize any potential
chemical interaction between the functional monomers and this
residual hydroxyapatite. Therefore, we used synthetic
hydroxyapatite to exclude any influence from other organic
tooth components that would have made unambiguous
interpretation of the XPS spectra impossible. We selected 1
carboxyl- and 2 phosphate-based monomers that constitute the
functional monomers of 3 representative two-step self-etch
adhesives (Van Meerbeek et al., 2003). We used XPS to
investigate interfacial interaction of adhesive molecules with
hydroxyapatite (Yoshida et al., 2000, 2001); it showed that all 3
functional monomers interacted differently with hydroxyapatite.
As can be derived theoretically from the chemical formula of
4-MET, the ratio of C-C/C-H/C=C bindings to C-O bindings and
to -COO- bindings is 9 to 2 to 4. Peak area analysis (n = 13) of
the peak that appeared when 4-MET interacted with
hydroxyapatite revealed areas of, respectively, 9.1 + 0.2 for the
C-C/C-H/C=C peak and 2.0 + 0.0 for the C-O peak, when the
area of the -COO- bindings was taken as 4. The theoretical and
measured peak ratios are alike, thus indicating that the recorded
C 1s peak represented 4-MET that remained attached to
hydroxyapatite. This C 1s peak cannot represent any carbon
contamination, since it would have resulted in totally different
peak area ratios. Previously, we demonstrated—for polyalkenoic
acids (Yoshida et al., 2000) as well as for some mono-, di-, and
tri-carboxylic acids (Yoshida et al., 2001)—that the shift of the
carboxyl peak to a lower binding energy could be explained by
the formation of an ionic bond between the carboxyl groups and
the Ca of hydroxyapatite. Likewise, the peak shift resulting from
the interaction of 4-MET with hydroxyapatite indicates that the
carboxyl groups formed an ionic bond with Ca of hydroxy-
apatite. While a short 30-second exposure resulted in rather weak
interaction intensity, 4-MET interacted more intensely with
hydroxyapatite after a 30-minute exposure.
The interaction of phenyl-P with hydroxyapatite cannot be
explained by simple bonding to hydroxyapatite. As can
theoretically be derived from the chemical formula of phenyl-P,
the ratio of C-C/C-H/C=C bindings to C-O bindings and to -
COO- bindings is 8 to 3 to 1. Unreacted phenyl-P (Fig. 2b)
showed a peak shape that reflects the theoretical peak ratio,
which differs completely from the measured peak area ratio of
the C 1s peak of phenyl-P on hydroxyapatite (n = 32).
Therefore, the recorded spectrum should most likely be
ascribed to phenyl-P that was hydrolytically broken into several
sub-components following ultrasonic rinsing. In addition,
among the three monomer solutions investigated, the 15%
(w/w) phenyl-P solution was the most highly concentrated in
molarity, because the molecular weights of 4-MET, phenyl-P, Figure 3. TEM photomicrographs illustrating the resin-dentin interfaces
produced by Unifil Bond (GC) (a,b), Clearfil Liner Bond 2 (Kuraray) (c),
and 10-MDP are 322.3, 286.2, and 322.3, respectively. and Clearfil SE (Kuraray) (d,e). The photomicrographs in (a) and (d)
Nevertheless, even a 30-minute application resulted in only a represent unstained, non-demineralized sections, those in (b) and (e)
rather weak C 1s peak that would be indicative of chemical stained, demineralized sections, and that in (c) a stained, non-
interaction with hydroxyapatite. demineralized section. A = adhesive resin; H = hybrid layer; L = lab-
demineralized unaffected dentin; R = resin-impregnated smear plug; U
As can theoretically be derived from the chemical formula
= unaffected dentin.
of 10-MDP, the ratio of C-C/C-H/C=C bindings to C-O
bindings and to -COO- bindings is 11 to 2 to 1. Peak area
analysis (n = 11) of the peak that appeared upon interaction of
10-MDP with hydroxyapatite (Fig. 2c) revealed areas of, 0.1 for the C-O peak when the area of the -COO- bindings was
respectively, 10.7 ± 0.6 for the C-C/C-H/C=C peak and 2.0 + taken as 1. The C 1s peak of 10-MDP on hydroxyapatite is very
458 Yoshida et al. J Dent Res 83(6) 2004

similar to that of unreacted 10-MDP (Fig. 2c). Like 4-MET, but Sports and Culture of Japan and by funds from the Uehara
unlike phenyl-P, the theoretical and measured peak ratios are Memorial Foundation, The Nakatomi Foundation, and the
alike, indicating that the recorded C 1s peak represented 10- Toshio Nakao Chair for Adhesive Dentistry, inaugurated at the
MDP that remained strongly attached to hydroxyapatite. Catholic University of Leuven (B. Van Meerbeek and P.
When we compared the chemical bonding efficacy of the 3 Lambrechts, Chairholders). We thank GC and Kuraray for
functional monomers investigated, quantitative determination providing the monomers and calcium salts. We also thank J.
of carbon concentration on hydroxyapatite revealed that the Yearn (GC Europe) for critically reviewing the manuscript.
bonding potential of 10-MDP to hydroxyapatite is significantly
stronger than that of 4-MET, even after a 30-minute exposure. REFERENCES
Moreover, the fact that even a short 30-second exposure of
Beamson G, Briggs D (1992). Lineshapes. In: High resolution XPS of
hydroxyapatite to 10-MDP produced a significant C 1s peak,
organic polymers. The Scienta ESCA 300 Database. Chichester:
and that the peak intensity did not increase further when the
John Wiley & Sons Ltd., pp. 33-36.
exposure time was extended, strongly suggests that 10-MDP
Inoue S, Van Meerbeek B, Vargas M, Yoshida Y, Lambrechts P,
has a high chemical bonding potential to hydroxyapatite within
Vanherle G (2000). Adhesion mechanism of self-etching
a clinically reasonable application time. The chemical bonding
adhesives. In: Advanced adhesive dentistry. Proceedings of the
capacity of 4-MET is weaker, and it is doubtful that 4-MET,
3rd International Kuraray Symposium, Dec. 3-4, 1999, Granada,
even within a short application time, is capable of chemically
Spain. Tagami J, Toledano M, Prati CA, editors. Cirimido, Italy:
bonding to hydroxyapatite. This confirms the rather weak
Grafiche Erredue, pp. 131-148.
adsorption of 4-META from ethanol and dichloromethane onto
Inoue S, Vargas MA, Van Meerbeek B, Abe Y, Yoshida Y,
synthetic hydroxyapatite reported in a study by Misra (1989).
Lambrechts P, et al. (2001). Micro-tensile bond strength of eleven
The chemical bonding efficacy of phenyl-P is clearly the lowest
modern adhesives to dentin. J Adhes Dent 3:237-246.
among the 3 functional monomers investigated.
Misra DN (1989). Adsorption of 4-methacryloxyethyl trimellitate
However, chemical bonding potential on its own is
anhydride (4-META) on hydroxyapatite and its role in composite
insufficient to contribute to bonding performance. The ionic
bonding. J Dent Res 68:42-47.
bindings formed should be stable in an aqueous environment.
Nakabayashi N, Saimi Y (1996). Bonding to intact dentin. J Dent Res
XPS showed that chemical bonding induced, in particular, by
75:1706-1715.
10-MDP, even after a short 30-second application, resisted
Nakabayashi N, Kojima K, Masuhara E (1982). The promotion of
ultrasonic cleaning. In addition, AAS revealed that the calcium
adhesion by the infiltration of monomers into tooth substrates. J
salt of 10-MDP was hardly soluble. The calcium salt of 4-MET
Biomed Mater Res 16:265-273.
could be dissolved more easily, whereas phenyl-P was very
Sano H, Takatsu T, Ciucchi B, Horner JA, Matthews WG, Pashley DH
soluble in water. These data conform to the adhesion-
(1995). Nanoleakage: leakage within the hybrid layer. Oper Dent
decalcification (AD) concept previously presented (Yoshida et
20:18-25.
al., 2001; Yoshioka et al., 2001). According to the AD-concept,
Sano H, Yoshikawa T, Pereira PN, Kanemura N, Morigami M,
the less soluble the calcium salt of the acidic molecule, the
Tagami J, et al. (1999). Long-term durability of dentin bonds
more intense and stable the molecular adhesion to a
made with a self-etching primer, in vivo. J Dent Res 78:906-911.
hydroxyapatite-based substrate. The higher and more stable
Tay FR, Pashley DH (2001). Aggressiveness of contemporary self-
bonding performance of 10-MDP vs. that of 4-MET is also
etching systems. I. Depth of penetration beyond dentin smear
reflected in the higher micro-tensile strength to dentin of the
layers. Dent Mater 17:296-308.
10-MDP-based Clearfil SE (Kuraray) compared with that of the
Van Meerbeek B, Yoshida Y, Lambrechts P, Vanherle G, Duke ES,
4-MET-based Unifil Bond (GC) (Inoue et al., 2001). In
Eick JD, et al. (1998). A TEM study of two water-based adhesive
addition to increased bond strength, improved monomer-
systems bonded to dry and wet dentin. J Dent Res 77:50-59.
dentinal tissue interaction may also enhance sealing potential
Van Meerbeek B, De Munck J, Yoshida Y, Inoue S, Vargas M, Vijay
for the prevention of nano-leakage (Sano et al., 1995), and thus
P, et al. (2003). Buonocore memorial lecture. Adhesion to enamel
extend bonding longevity (Sano et al., 1999).
and dentin: current status and future challenges. Oper Dent
In conclusion, besides self-etching dentin, some functional
28:215-235.
monomers may, within a clinically reasonable time, interact
Yoshida Y, Van Meerbeek B, Nakayama Y, Snauwaert J, Hellemans
chemically with hydroxyapatite. Hydroxyapatite, however,
L, Lambrechts P, et al. (2000). Evidence of chemical bonding at
needs to remain available at the partially demineralized dentin
biomaterial-hard tissue interfaces. J Dent Res 79:709-714.
surface. The specific molecular formula of the functional
Yoshida Y, Van Meerbeek B, Nakayama Y, Yoshioka M, Snauwaert J,
monomer and the subsequent dissolution rate of its calcium salt
Abe Y, et al. (2001). Adhesion to and decalcification of
determine actual bonding efficacy and stability.
hydroxyapatite by carboxylic acids. J Dent Res 80:1565-1569.
Yoshioka M, Yoshida Y, Inoue S, Lambrechts P, Vanherle G, Nomura
ACKNOWLEDGMENTS Y, et al. (2001). Adhesion/decalcification mechanisms of acid
This study was supported in part by a Grant-in-Aid for interactions with human hard tissues. J Biomed Mater Res 59:56-
Scientific Research from the Ministry of Education, Science, 62.