Adhesion/decalcification mechanisms of acid interactions

with human hard tissues

M. Yoshioka,1 Y. Yoshida,2,3 S. Inoue,4 P. Lambrechts,3 G. Vanherle,3 Y. Nomura,2 M. Okazaki,2

H. Shintani,1 B. Van Meerbeek3
1
Department of Operative Dentistry, Hiroshima University Faculty of Dentistry, 1-2-3 Kasumi, Minami-Ku, Hiroshima
734-8553, Japan
2
Department of Biomaterials Science, Hiroshima University Faculty of Dentistry, 1-2-3 Kasumi, Minami-Ku,
Hiroshima 734-8553, Japan
3
BIOMAT—Department of Operative Dentistry and Dental Materials, School of Dentistry, Oral Pathology and
Maxillo-Facial Surgery, Catholic University of Leuven, Kapucijnenvoer 7, B-3000, Leuven, Belgium
4
Department of Operative Dentistry, Hokkaido University School of Dentistry, Kita13-Nishi7, Kita-Ku, Sapporo
060-8586, Japan

Received 8 January 2001; accepted 16 May 2001

Abstract: In order to study adhesion/decalcification
mechanisms of acid interactions with human hard tissues
such as bones and teeth, the chemical interaction of five
carboxylic acids (acetic, citric, lactic, maleic, and oxalic) and
two inorganic acids (hydrochloric and nitric) with enamel
and two synthetic hydroxyapatite (HAp) powders with, re-
spectively, a high and a low crystallinity were analyzed us-
ing X-ray photoelectron spectroscopy (XPS), atomic absorp-
tion spectrophotometry (AAS), and spectrophotometry (S).
X-ray diffraction revealed that the crystallinity of the highly
crystallized HAp was considerably higher than that of
enamel while the crystallinity of the poorly crystallized HAp
was similar to that of dentin and bone. XPS of acid-treated
enamel demonstrated for all carboxylic acids ionic bonding
to calcium of HAp. AAS and S showed for both HAps that
all carboxylic and inorganic acids except oxalic acid ex-
tracted Ca significantly more than P, leading to a Ca/P ratio
close to that of synthetic HAp (2.16 w/w). Oxalic acid ex-
tracted hardly any Ca, but substantially more P, leading to a
significantly smaller Ca/P ratio than that of HAp. AAS
showed that the calcium salt of oxalic acid hardly could be
dissolved, whereas the calcium salts of all the other acids
were very soluble in their respective acid solution. These
results confirm the adhesion/decalcification concept (AD-
concept) previously advanced. Depending on the dissolu-
tion rate of the respective calcium salts, acids either adhere
to or decalcify apatitic substrates. It is concluded that the
AD-concept that originally dictated the interaction of car-
boxylic acids with human hard tissues can be extended to
inorganic acids, such as hydrochloric and nitric acid. Fur-
thermore, HAp crystallinity was found not to affect the ad-
hesion/decalcification behavior of acids when interacting
with apatitic substrates, so that the AD-concept can be ap-
plied to all human hard tissues with varying HAp crystal-
linity. © 2001 John Wiley & Sons, Inc. J Biomed Mater Res 59:
56–62, 2002
Key words: acids; hydroxyapatite; human hard tissues; ad-
hesion/decalcification concept; AD-concept

INTRODUCTION tal and medical processes. It is, for instance, well

The interaction of acids with human hard tissues
such as teeth and bones is fundamental to diverse den-
Correspondence to: B. Van Meerbeek; e-mail: bart.
vanmeerbeek@med.kuleuven.ac.be
Contract grant support: Grant-in-Aid for Scientific Re-
search from the Ministry of Education, Science, Sports and
Culture of Japan; contract grant number: 08771789
Contract grant support: Grant-in-Aid for Scientific Re-
search from the Ministry of Education, Science, Sports and
Culture of Japan; contract grant number: 12771182
© 2001 John Wiley & Sons, Inc.
known that lactic acid produced by oral bacteria dis-
solves inorganic components of teeth, causing dental
caries.1–3 Other acids, such as citric, maleic, and nitric
acid, have been used to pre-treat dentin and enamel as
part of a resin-based adhesive system that is applied in
order to bond restorative composites to tooth tissue.4,5
Such acids also decalcify the apatitic component of
enamel and dentin. On the other hand, polyalkenoic
or polycarboxylic acids, such as polyacrylic acid co-
polymers in glass–polyalkenoate or glass–ionomer ce-
ments, decalcify hardly at all, but rather bond to min-
eral tissue. Direct experimental evidence elucidating
the underlying mechanism of self-adhesion of poly-
ADHESION/DECALCIFICATION CONCEPT 57

alkenoic acids to synthetic hydroxyapatite (HAp) re- MATERIALS AND METHODS

cently has been provided.6
In an attempt to better understand the mechanisms
underlying this contrasting adhesion/decalcification
behavior of carboxylic acids when interacting with
HAp-based substrates, we recently proposed a model
that has been referred to as the “Adhesion-Decalcification
Concept,” or AD-concept.7 The mechanism of acidic
interaction with HAp was found to involve two
phases (Fig. 1). In the first phase, carboxylic acids
bond to calcium of HAp. Depending on the diffusion
rate of the calcium–acid complexes into solution, the
acid will in the second phase either remain attached to
the HAp surface with only limited decalcification in-
volved, or the calcium–acid complex will debond, re-
sulting in a substantial decalcification effect. However,
the model so far is valid only for explaining the inter-
action of specifically 1 mono-, 2 di-, 1 tri- and 2 poly-
carboxylic acids with synthetic HAp that possesses a
particular crystallinity.7 With regard to the latter, hu-
man hard tissues are known to consist of apatitic com-
ponents with various crystallinities.
Therefore, the objectives of this study were to find
out (1) if the AD-concept also applies for other carbox-
ylic and even noncarboxylic acids, and (2) when they
interact with HAp substrates with varying crystallin-
ity. We used X-ray photoelectron spectroscopy (XPS)
to gather chemical information from acids when they
interacted with enamel. X-ray diffraction (XRD) was
employed to determine the crystallinity of diverse hu-
man hard tissues and synthetic HAps. Atomic absorp-
tion spectrophotometry (AAS) and spectrophotometry
(S) were used to analyze the concentration of calcium
and phosphorus extracted by the acids from enamel
and two HAps of either a high or a low crystallinity.
Finally, field-emission scanning electron microscopy
(Fe-SEM) was used to ultramorphologically character-
ize the interaction of the acids with human enamel.
In order to analyze chemically the mechanisms of adhe-
sion to and decalcification of HAp, the chemical interaction
of one carboxylic acid, acetic acid, and two noncarboxylic
acids, hydrochloric and nitric acid, with synthetic hydroxy-
apatite and enamel were analyzed and compared to four
other carboxylic acids previously investigated (Table I).7 The
acids investigated were used in a concentration relevant to
their use for a specific dental purpose, that is, as a compo-
nent of a dental adhesive material, or with relevance to the
process of dental caries. Citric (CA), maleic (MA), and nitric
acid (NA) were tested in their respective concentrations
(Table I), like when they are used as the initial conditioning
step in the application procedure of several commercial ad-
hesives.4 Acetic (AA) and lactic acid (LA) were used in a
0.1M concentration, as they commonly are used in a cario-
genic solution to produce artificial caries or to test the car-
iostatic potential of restorative materials.1–3,8–11 Hydrochlo-
ric acid (HCA) was used in a 0.2M solution, as it has been
used to investigate the relationship between the develop-
ment of enamel erosions and caries lesions.12 Finally, oxalic
acid (OA) was used in a 1M solution; it often in a salt for-
mulation has been used to treat dentin hypersensitivity.13
As apatitic substrates, human enamel, dentin, bone, syn-
thetic HAp plates (APP-101, Asahi Optical, Tokyo, Japan),
poorly and highly crystallized (heated at 1000°C for 5 h), and
HAp powder (AP-20C, Sekisui Plastics, Osaka, Japan) were
investigated.
X-Ray Diffraction (XRD)
X-ray diffraction was employed to determine the crystal-
linity of all apatitic substrates. Measurements were made on
an X-ray diffractometer (XD-D1, Shimadzu, Kyoto, Japan)
with a Cu target, 0.1-mm receiving slit, and scanning speeds
of 2° and 1/32°/min at 30 kV and 30 mA.
X-Ray Photoelectron Spectroscopy (XPS)

Enamel caps were obtained from extracted caries-free and

unrestored human molars stored in an aqueous solution of
0.5% chloramine at 4°C for a maximum of 1 month after
extraction. The occlusal enamel was removed parallel to the

TABLE I
List of Acids Investigated
Acids Code Concentration pH MW
Acetic acid AA 0.1 M 2.9 60
Citric acida CA 10 w/w% 1.5 192
Hydrochloric acid HCA 0.2 M 0.7 36
Lactic acida LA 0.1 M 2.4 90
Maleic acida MA 10 w/w% 0.9 116
Nitric acida NA 2.5 w/w% 1.0 63
Oxalic acid OA 1M 0.6 90
Figure 1. Schematic presentation of the “Adhesion/
Decalcification Concept” previously advanced.7 a
Carboxylic acids that previously have been investigated.6
58
occlusal surface using an EXAKT cutting–grinding system
(EXAKT, Norderstedt, Germany). The enamel surface was
polished using 600-grit SiC sandpaper for 1 min, followed
by ultrasonic rinsing with distilled water for 20 min.
The acid solutions investigated were applied on enamel as
previously described in detail by Yoshida et al.7 For AA, CA,
LA, HCA, MA, and NA, the acid solutions were applied on
enamel with any excess solution immediately removed from
the surface with an absorbent paper and a gentle N2-gas
blast. The specimens then were mildly rinsed with organic
solvent/ultrapure water (Milli-Q water: > 18 M⍀cm) in or-
der to remove the excess acid and calcium salt that had
formed through reaction with Ca extracted from enamel. For
OA, enamel caps were put in the 1M OA solution at 37°C for
30 min, followed by ultrasonic rinsing with ultrapure water
for 15 min.
XPS then was performed using an AXIS-HS spectrometer
(Kratos, Manchester, UK), under a base pressure of less than
10−7 Pa. Al-Ka monochromatic X ray was obtained using a
15-kV accelerating voltage and 10-mA filament current.
Wide and narrow scans were measured at pass energies of,
respectively, 80 and 40 eV. Quantitative data were obtained
from peak areas, and chemical interactions were identified
from detailed measurement of peak positions and interpeak
distances.
Atomic Absorption Spectrophotometry (AAS)
and Spectrophotometry (S)
The HAp powder samples and the calcium salts of the
acids were immersed in 100 mL of each respective acid so-
lution and stirred for 15 s at 300 rpm and 37°C. About 2 mL
of solution were withdrawn at the top of the recipient and
subsequently filtered through a polytetrafluoroethylene
membrane with a pore size of 0.20 ␮m (Samprep-LCR25-LG,
Millipore). After 30 min of stirring, the remaining solution
was centrifuged at 3000 rps for 10 min and then again fil-
tered through a polytetrafluoroethylene membrane (pore
size of 0.20 ␮m). The solution then was analyzed for calcium
using an atomic absorption spectrophotometer (AAS; AA-
670, Shimadzu, Kyoto, Japan) and for phosphorus using a
spectrophotometer (S; UV mini-1240, Shimadzu, Kyoto, Ja-
pan). The amount of calcium and phosphate ions extracted
from HAp powder by the acids and the dissolution rate of
the calcium salts in the respective acid solutions were quan-
tified.
For spectrophotometry, we adopted the “Eastoe”
method14 for quantitative analysis of phosphorus. One mL
of the respective HAp powder and calcium salt solutions
was added to a mixture of 5 mL of distilled water, 0.4 mL of
ammonium molybdate/sulfate reagent, 0.2 mL of ami-
nonaphtholsulfonic acid/sodium sulfite reagent, and fur-
ther added up to 10 mL with distilled water. Subsequently,
the solutions were immersed successively in water at 100°C
for 10 min and in water at 15°C for 5 min prior to analysis at
815 nm.
Field-Emission Scanning Electron
Microscopy (FE-SEM)
Enamel caps were prepared in the same way as has been
described for the XPS analysis. The polished enamel surface
YOSHIOKA ET AL.
then was treated with the respective acids at 37°C, followed
by ultrasonic rinsing with distilled water for 5 min. The
specimens immediately were immersed in 2.5% glutaralde-
hyde in 0.1M of sodium cacodylate buffer at pH 7.4 for 24 h
at 4°C. After fixation, the specimens were rinsed with 20 mL
of 0.2M sodium cacodylate buffer at pH 7.4 for 1 h, with
three changes, followed by distilled water for 1 min. They
then were dehydrated in ascending grades of ethanols (25%
for 20 min, 50% for 20 min, 75% for 20 min, 95% for 30 min,
and 100% for 60 min). Finally, the specimens were immersed
in hexamethyldisilazane (HMDS) for 10 min and left to dry
in atmosphere.5,15,16 The enamel surfaces, treated with acids
as well as untreated (control), eventually were ultramorpho-
logically characterized using a Fe-SEM (S-4100, Hitachi, To-
kyo, Japan).
RESULTS
XRD (Fig. 2) revealed that the crystallinity of a HAp
plate was higher than that of enamel. The poorly crys-
tallized HAp powder had a crystallinity similar to
those of dentin and bone. After heating at 1000°C for
5 h, the crystallinity of the HAp powder considerably
increased and was significantly higher than that of
enamel.
XPS (Figs. 3–5) of enamel treated with MA disclosed
a twofold C 1s peak at a binding energy of approxi-
mately 285 eV [Fig. 3(b)]. Although XPS of untreated
enamel already presented a C 1s peak [Fig. 3(a)], ap-
plication of MA significantly increased its intensity.
The narrow−scan spectrum of the C 1s region of MA
(Fig. 4) showed that the twofold peak consists of one
peak at a binding energy of 284.6 eV, mainly repre-
senting the C=C and the C-H bindings of MA, and
another peak at 288.5 eV, representing the two car-
boxyl groups. When MA was applied to enamel, the
latter peak significantly (t Test: p < 0.01) shifted to a
lower binding energy (Fig. 4). This shifted peak at
288.2 eV represents the formation of an ionic bond
between the carboxyl groups of MA and calcium of
enamel. Application of OA on enamel significantly in-
creased the intensity of the C 1s peak and reduced the
intensity of the P peaks [Fig. 3(c)]. The narrow-scan
spectra in Figures 4 and 5 reveal that the major peak at
288.2 eV must be attributed to the formation of an
ionic bond between the carboxyl groups of, respec-
tively, MA and OA, and Ca of enamel HAp. Other
peaks most likely originate from the initial untreated
enamel surface.
AAS and S (Table II) revealed that when HAp was
immersed during 30 min in the acid solutions, it re-
sulted in slightly higher amounts of extracted Ca and
P than when they were immersed in acid for only 15 s.
The dissolution rate of the highly crystallized HAp
was substantially lower than that of the poorly crys-
tallized HAp. For poorly as well as highly crystallized
ADHESION/DECALCIFICATION CONCEPT 59

Figure 3. XPS wide-scan spectra of (a) polished enamel, (b)

enamel exposed to MA, and (c) enamel exposed to OA.

polyalkenoate cements to mineralized tissues is based

upon electrostatic reaction of the carboxyl groups of
the polyalkenoic acid with the calcium of synthetic
HAp.6 A similar methodology later was applied to
analyze the chemical interaction of diverse carboxylic
Figure 2. X-ray diffraction patterns of (a) human enamel, acids with synthetic HAp.7 Based on the results of the
(b) dentin, (c) bone, (d) HAp plate, (e) poorly crystallized latter study, we advanced the AD-concept that pro-
HAp powder, and (f) highly crystallized HAp powder. Note
that the crystallinity of the HAp plate and the highly crys- vides an explanation as to why carboxylic acids either
tallized HAp powder is higher than the crystallinity of adhere to or merely decalcify HAp. Both previous
enamel. The crystallinity of the poorly crystallized HAp studies clearly demonstrated the potential and accu-
powder is similar to the crystallinity of dentin and bone. racy that can be obtained through the correlative use
HAp powders, all carboxylic and inorganic acids, ex-
cept oxalic acid, extracted Ca significantly more than
they extracted P, leading to a Ca/P ratio close to that
of HAp (2.16 w/w) (Table II). OA extracted hardly any
Ca but substantially more P, leading to a significantly
smaller Ca/P than that of HAp.
AAS also revealed that the salts of oxalic acid could
hardly be dissolved whereas the calcium salts of all
other acids were very soluble in their respective acid
solutions (Table III).
FE-SEM of enamel exposed to 10% MA disclosed a
typical enamel etch-pattern (Fig. 6)17 while enamel
treated with 1M of OA (Fig. 7) showed a surface quite
like the untreated enamel control (Fig. 8).

DISCUSSION AND CONCLUSIONS

By means of XPS, we previously could provide di- Figure 4. XPS narrow-scan spectrum of the C 1s region of
rect evidence that the self-adhering potential of glass– MA and MA on enamel.
60 YOSHIOKA ET AL.

mineral substrates with different crystallinities, as ex-

Figure 5. XPS narrow-scan spectrum of the C 1s region of
OA on enamel.
of XPS, AAS, and S in the investigation of chemical
interactions of acids with HAp. The same methodol-
ogy was employed in this study, not only to test the
validity of the AD-concept for the interaction of car-
boxylic acids with human hard tissue such as enamel,
but also to find out if the AD-concept is valid for clari-
fying the interaction behavior of noncarboxylic acids
such as hydrochloric and nitric acid with apatitic tis-
sues. The secondary objective of this study was to in-
vestigate whether or not the adhesion-decalcification
interaction of acids with mineralized tissue depends
on the crystallinity of the apatitic substrate. Alto-
gether, this study was designed to allow us to draw
conclusions as to the validity of the AD-concept not
only for diverse types of acids but also for diverse
ist in enamel, dentin, and bone.
All carboxylic acids investigated were found to re-
veal a significant shift of the carboxyl peak to a lower
binding energy, indicating the formation of ionic
bonds between the carboxyl groups of the acids with
the Ca of HAp enamel. Among these acids, AA, CA,
LA, and MA formed only a weak bond to enamel, with
a gentle rinsing with distilled water sufficing to almost
completely remove the acids (below detection limit of
XPS). On the contrary, OA formed a rather strong
bond to enamel and still could be detected by XPS
even after thorough ultrasonic rinsing with distilled
water. This tendency to either bond to or to decalcify
enamel was confirmed by AAS and spectrophotom-
etry, which showed that calcium and phosphorus
were extracted by AA, CA, LA, and MA at a rate
equivalent to the Ca/P ratio of HAp whereas OA ex-
tracted mainly phosphorus and hardly any calcium.
These findings eventually were corroborated by the
Fe-SEM analysis.
Consequently, both the adhesion and the decalcifi-
cation phenomena resulting from mono-, di-, and tri-
carboxylic acids interacting with apatitic substrates
still can be explained by the “adhesion/decalcification
concept” (AD-concept) that we advanced previously
(Fig. 1).7 Following this concept, the carboxyl groups
of the acids form in a first-phase ionic bonds to Ca at
the HAp surface. This step may be determined largely
by the pKa of the acids. At the same time, PO43− and
OH− are extracted by H3O+ from the HAp surface and
brought into solution. These processes occur with
maintenance of electron neutrality at both the HAp
surface and into solution. In the second phase, the
acids either remain attached to the Hap surface (OA)

TABLE II
AAS and S Results of the Amount of Calcium and Phosphorus Extracted from HAp Powder by the Acids Investigated
Highly crystallized HAp powder
in acid for 15 s (ppm) AA CA HCA LA MA NA OA
Ca 2560 9730 9180 880 11800 5110 2.73
P 979 5180 3190 417 5610 2350 1550
Ca/P (Ca/PHAp = 2.16 w/w) 2.61 1.88 2.88 2.11 2.10 2.17 0.00176
Highly crystallized HAp powder
in acid for 30 min (ppm) AA CA HCA LA MA NA OA
Ca 2870 12300 16000 1200 13500 5190 3.07
P 1310 6920 5660 427 6900 2380 1440
Ca/P (Ca/PHAp = 2.16 w/w) 2.19 1.78 2.83 2.81 1.96 2.18 0.00213
Poorly crystallized HAp powder
in acid for 15 s (ppm) AA CA HCA LA MA NA OA
Ca 4340 12400 9800 1200 12700 5790 2.84
P 1810 6100 3640 619 5960 2870 1260
Ca/P (Ca/PHAp = 2.16 w/w) 2.40 2.03 2.69 1.94 2.13 2.02 0.00225
Poorly crystallized HAp powder
in acid for 30 min (ppm) AA CA HCA LA MA NA OA
Ca 4940 13600 17100 1530 14600 7680 3.41
P 2020 7120 5960 719 7070 3980 1690
Ca/P (Ca/PHAp = 2.16 w/w) 2.45 1.91 2.87 2.13 2.07 2.01 0.00202
All data were obtained to three significant figures.
ADHESION/DECALCIFICATION CONCEPT 61

TABLE III
AAS Results of the Solubility of Calcium Salts in the Respective Acid Solutions
AAS (ppm) 15 sec
amount of Ca extracted HCA NA MA CA LA AA OA H2O
Calcium chloride 501000 452000
Calcium nitrite 189000 188000
Ca(C4H2O4) 20200 4650
Ca(C4H3O4)2 48800 27500
Calcium citrate 12700 174
Calcium lactate 11700 10400
Calcium acetate 53600 52100
Calcium oxalate 6.79 0.278
AAS (ppm) 30 min
amount of Ca extracted HCA NA MA CA LA AA OA H2O
Calcium chloride 513000 462000
Calcium nitrite 241000 245000
Ca(C4H2O4) 24300 4900
Ca(C4H3O4)2 53400 28500
Calcium citrate 18600 183
Calcium lactate 15500 11900
Calcium acetate 64200 54200
Calcium oxalate 6.86 0.323
All data were obtained to three significant figures.

with only a limited decalcification step, or they
debond with a significant decalcification effect (AA,
CA, LA, and MA).
The dissolution rate measurements using AAS
show that the calcium salts of AA, CA, LA, and MA
were very soluble in their respective acid solutions
whereas the salts of OA hardly could be dissolved.
The fact that acids either adhere to or decalcify HAp
appears to depend largely upon the dissolution rate of
the respective calcium salts in their respective acidic
solutions. Although XPS showed a similar bonding
effect to HAp for all acids, the less soluble their re-
spective salts, the more stable their bond to HAp and
the less they decalcify HAp. This second phase there-
fore was termed the adhesion/decalcification-
determining step (Fig. 1).
Besides mono-, di-, and tricarboxylic acids, polycar-
boxylic acids such as polyalkenoic acids that form the
basic component of glass–polyalkenoate cements18–20
interact with apatitic substrates, following the AD-
concept. These acids are responsible for the self-
adhesiveness of glass–polyalkenoate cements and be-
come strongly attached to the calcium of apatitic sub-
strate through ionic bonding, as previously was
reported.6,7 It also is apparent that due to immediate
gel formation, calcium–polyalkenoate complexes
hardly dissolve in the polyalkenoic acid solution itself.
Finally, the present study has shown that the interac-
tion of noncarboxylic acids such as HCA and NA also
occurs, following the AD-concept.

Figure 7. Fe-SEM photomicrograph of enamel exposed to

OA. Note that the morphology of enamel treated with 1M of
Figure 6. Fe-SEM photomicrograph illustrating the typical OA resembles that of the polished, untreated enamel surface
enamel etch-pattern exposed by MA. Bar = 3 ␮m. shown in Figure 8. Bar = 3 ␮m.
62
Figure 8. Fe-SEM photomicrograph of 600-grit SiC wet-
sanded enamel, followed by ultrasonic rinsing with distilled
water for 20 min. Note the scars produced by polishing,
which make enamel prisms hardly observable. Bar = 3 ␮m.
In order to be able to generalize the AD-concept to
all types of human hard tissues, the crystallinity of the
apatitic structure was hypothesized to be a factor of
major importance that may influence its interaction
with the acid. The inorganic phase of enamel, dentin,
and bone consist of biologic apatites that differ
strongly in crystallinity. As shown by the XRD,
enamel mineral has a significantly higher crystallinity
than that of both dentin and bone. These results are in
line with data out of the literature.21–26 In this study,
the interaction of all acids with enamel and two syn-
thetic HAp powders with respectively higher crystal-
linity than enamel and crystallinities similar to dentin
and bone (lower crystallinity than enamel) were ana-
lyzed. The acid interaction was found not to depend
on the crystallinity of HAp.
Consequently, from this study can be concluded
that the AD-concept can be generalized to organic and
inorganic acids and to all human hard tissues with
varying HAp crystallinity.
References
1. Allenspach-Petrzilka GE, Guggenheim B, Lutz F. Morphologi-
cal aspects of experimental dentinal caries in rats. Caries Res
1987;21:47–61.
2. van Dorp CSE, Exterkate RAM, ten Cate JM . Mineral loss
during etching of enamel lesions. Caries Res 1990;24:6–10.
3. Kleter GA, Damen JJM, Everts V, Niehof J, ten Cate JM. The
influence of the organic matrix on demineralization of bovine
root dentin in vitro. J Dent Res 1994;73:1523–1529.
4. Nakabayashi N, Pashley DH. Hybridization of dental hard tis-
sues. Tokyo: Quintessence; 1998.
5. Perdigao J. An ultra-morphological study of human dentine
YOSHIOKA ET AL.
exposed to adhesive systems. [PhD Thesis]. Belgium: Catholic
University of Leuven; 1995.
6. Yoshida Y, Van Meerbeek B, Nakayama Y, Snauwaert J, Hel-
lemans L, Lambrechts P, Vanherle G, Wakasa K. Evidence of
chemical bonding at biomaterial–hard tissue interface. J Dent
Res 2000;79:709–714.
7. Yoshida Y, Van Meerbeek B, Nakayama Y, Yoshioka M, Snau-
waert J, Abe Y, Lambrechts P, Vanherle G, Okazaki M. Adhe-
sion to and decalcification of hydroxyapatite by carboxylic ac-
ids. 2001. Forthcoming.
8. Margolis HC, Moreno EC. Kinetics of hydroxyapatite dissolu-
tion in acetic, lactic, and phosphoric acid solutions. Calcif Tis-
sue Int 1992;50:137–143.
9. Klont B, Damen JJ, ten Cate JM. Degradation of bovine incisor
root collagen in an in vitro caries model. Arch Oral Biol 1991;
36:299–304.
10. Klont B, ten Cate JM. Remineralization of bovine incisor root
lesions in vitro: The role of the collagenous matrix. Caries Res
1991;25:39–45.
11. Klont B, ten Cate JM. Susceptibility of the collagenous matrix
from bovine incisor root to proteolysis after in vitro lesion
formation. Caries Res 1991;25:46–50.
12. Larsen MJ. On the chemical and physical nature of erosions
and caries lesions in dental enamel. Caries Res 1991;25:323–
329.
13. Pashley DH, Outhwaite WC. Desensitizing oxalate dental com-
position and method of treatment. US patent no. 4,057,621;
1977.
14. Eastoe JE. Method for the determination of phosphate calcium
and protein in small portions of mineralized tissues [PhD The-
sis]. Belgium: Université de Liège; 1965.
15. Perdigao J, Lambrechts P, Van Meerbeek B, Vanherle G, Lopes
ALB. A field emission SEM comparison of four postfixation
drying techniques for human dentin. J Biomed Mater Res 1995;
29:1111–1120.
16. Perdigao J, Lambrechts P, Van Meerbeek B, Tome AR, Van-
herle G, Lopes AB. Morphological field emission–SEM study of
the effect of six phosphoric acid etching agents on human den-
tin. Dent Mater 1996;12:262–271.
17. Hermsen RJ, Vrijhoef MMA. Loss of enamel due to etching
with phosphoric or maleic acid. Dent Mater 1993;9:332–336.
18. Smith DC. A new dental cement. Br Dent J 1968;62:590–592.
19. Wilson AD, Kent BE. A new translucent cement for dentistry,
the glass ionomer cement. Br Dent J 1972;132:133–135.
20. Wilson AD, Prosser HJ, Powis DM. Mechanism of adhesion of
polyelectrolyte cements to hydroxyapatite. J Dent Res 1983;62:
590–592.
21. Miles AE. Structural and chemical organization of teeth. Vol. II.
New York: Academic Press; 1967.
22. Okazaki M, Aoba T, Doi Y, Takahashi J, Moriwaki Y. Solubility
and crystallinity in relation to fluoride content of fluoridated
hydroxyapatites. J Dent Res 1981;60:845–849.
23. Okazaki M, Moriwaki Y, Aoba T, Doi Y, Takahashi J. Solubility
behavior of CO3 apatites in relation to crystallinity. Caries Res
1981;15:477–483.
24. Okazaki M, Ohmae H, Hino T. Insolubilization of apatite–
collagen composites by UV irradiation. Biomaterials 1989;10:
564–568.
25. Okazaki M, Ohmae H, Takahashi J, Kimura H, Sakuda M.
Insolubilized properties of UV-irradiated CO3 apatite–collagen
composites. Biomaterials 1990;11:568–572.
26. LeGeros RZ, Sakae T, Bautista C, Retino M, LeGeros JP. Mag-
nesium and carbonate in enamel and synthetic apatites. Adv
Dent Res 1996;10:225–231.