Vol 5, No 4, 2003 259

Milestones in Adhesion:
Glass-ionomer Cements
Martin J. Tyas
a
a
Professor, School of Dental Science, University of Melbourne,
Australia.
Summary: The glass-ionomer cements (GIC), when introduced in the 1970s, were initially welcomed as
the first chemically-adhesive esthetic restorative materials. However, the clinical results were disappoint-
ing, as these early products were brittle and were particularly susceptible to early water loss and uptake,
and to long-term water loss. Continued development, notably the addition of resin polymerization com-
ponents, has made the GIC an increasingly important part of clinical practice.
J Adhes Dent 2003; 5: 259266. Submitted for publication: 10.12.02; accepted for publication: 16.04.03.
Reprint requests: Prof. M. Tyas, School of Dental Science, University
of Melbourne, 711 Elizabeth Street, Melbourne 3000, Australia.
Tel: + 61-3-9341-0231. Fax: + 61-3-9341-0437. e-mail: m.tyas@
unimelb.edu.au
uch of the foundation of operative (conserva-
tive) dentistry was established in the late
1800s and early 1900s by G.V. Black. The two prin-
cipal direct restorative materials at that time were
amalgam and silicate cement, and there was a lim-
ited understanding of the caries process. As a re-
sult, Blacks cavity designs were rigidly mechanistic
and resulted in serious weakening of the remaining
tooth structure.
52
The introduction of enamel etching in 1955
7
marked the beginning of adhesive dentistry. How-
ever, it was nearly 15 years before resin composite
materials became available and allowed enamel
etching to become a clinical reality.
8
Other adhe-
sive systems have become available since 1955,
mainly based on resins, and the bonding substrate
may be enamel, dentin, metal, or ceramic. In the
context of the replacement of enamel and dentin by
direct restorative materials, adhesive techniques
allow minimal cavity preparations and result in the
consequential benefits of maintenance of tooth
strength and a reduction in pulp damage and sec-
ondary caries.
52
M
Unique among direct restorative materials (ie,
excluding liner/base materials) are the glass-iono-
mer cements, as they are able to bond chemically
to enamel and dentin, in contrast to resin compos-
ites which bond micromechanically. Glass-ionomer
cements (GIC) are fundamentally water-based ma-
terials, although many more recent products (from
about 1990 onwards) contain a water-soluble res-
in.
TERMINOLOGY
The term glass ionomer as applied to the original
cements is not strictly correct,
27
which has led to
confusion among manufacturers and users. The
correct term for the original glass ionomers is
glass polyalkenoate cements; however, this has
not received general acceptance and is incorrect
for the later cements based on poly(vinyl phospho-
nic) acid.
27
For the purposes of this review, the
glass ionomers which set only by an acid-base re-
action (see below) will be referred to as conven-
tional GICs. In some products, metal particles
have been added in an attempt to enhance the
physical properties. Silver or gold can be fused with
the glass powder, cooled, and ground to an appro-
priate particle size to produce a cermet (ceram-
ic-metal) material,
26
eg, Ketac Silver (3M/Espe,
Seefeld, Germany). Alternatively, amalgam alloy
can be mixed with the GIC powder to produce an
Tyas
260 The Journal of Adhesive Dentistry
admix, eg, Miracle Mix (GC International, Tokyo,
Japan).
The addition of a water-soluble monomer com-
monly 2-hydroxyethylmethacrylate (HEMA) to the
GIC liquid results in a resin-modified glass iono-
mer (RM-GIC).
27
Terms such as light-cured glass
ionomer, hybrid ionomer, and resin ionomer
are often used instead of resin-modified glass ion-
omer. However, the last is preferred as it is a more
accurate description.
27
COMPOSITION AND SETTING REACTION
Conventional Glass Ionomers
The first GIC became commercially available in the
mid-1970s (ASPA, DeTrey, Konstanz, Germany).
The name ASPA was an acronym for aluminosili-
cate polyacrylate, and reflects the origin of GIC as
a natural development of silicate, zinc phosphate,
and zinc polycarboxylate cements. Silicate cement,
which became available in around 1880, was
formed essentially by mixing silica powder with
phosphoric acid liquid; phosphate cement was
formed essentially by mixing zinc oxide powder with
phosphoric acid liquid. In the mid-1960s, Smith
48
developed polycarboxylate cement by combining
polycarboxylic acid and zinc oxide. This was the first
dental material to bond ionically to tooth structure,
and this was achieved by the attraction of the car-
boxyl group (COO-) in the polycarboxylic acid to the
calcium in enamel and dentin.
In the 1970s, Wilson and Kent
57
modified the
silica powder to produce a calcium fluoroalumino-
silicate glass, which, when mixed with a modified
polyacrylic acid, resulted in the set glass-ionomer
cement. Since that date, significant improvements
have taken place. Other polyalkenoic acids (eg, ma-
leic acid, itaconic acid, butene dicarboxylic acid,
acrylic acid-itaconic acid copolymer, acrylic ac-
id-maleic acid copolymer) have been used in vari-
ous proprietary formulations.
41
Tartaric acid is also
commonly added in order to extend the working
time and sharpen the set.
32
Polyacrylic acid of
high molecular weight is viscous, and can increase
in viscosity during storage, which compromises ac-
curate dispensing; some formulations therefore
have the polyacrylic acid freeze-dried into the pow-
der, and the liquid consists of tartaric acid or wa-
ter.
21
Some powder formulations consist of a stron-
tium glass rather than a calcium glass.
41
The setting reaction of the conventional GICs is
a very complex acid-base reaction, and involves a
short initial setting reaction followed by a prolonged
maturation reaction. For the purposes of this re-
view, a simplified description of the setting reaction
is adequate.
32,41
When the powder and liquid are
mixed, the polyalkenoic acid attacks the fluoroalu-
minosilicate glass, resulting in surface degradation
of the glass and release of metal (eg, strontium,
calcium, and aluminum), fluoride ions, silicic acid.
These metal ions then combine with the carboxylic
acid groups of the polyacid to form a polyacid salt,
which becomes the matrix of the cement, and the
glass surface is changed to a silica hydrogel. The
unreacted glass core remains as a filler. However,
it should be noted that the presence of a filler does
not imply that the set GIC is a composite; in GICs
there is a concentration gradient of ions from filler
core to matrix, whereas in a composite there is a
distinct filler-matrix interface.
41
Although conventional GIC is clinically set within
3 to 5 min, maturation, that is, complete reaction
of cations and anions, may take months or even
years. In particular, the aluminum ions are very
slow to react.
32
The prolonged maturation phase of
the conventional GICs is a significant clinical disad-
vantage, as the cement has to be protected from
salivary (water) contamination for at least 24 h and
from water loss for several months.
32
The former
results in dissolution of the metal ions before they
can react, and the cement surface becomes
opaque and weak. Water loss results in shrinkage,
cracking and staining, and possibly debonding of
the restoration.
32
Resin-modified Glass Ionomers
The resin-modified glass ionomers (RM-GIC) were
introduced in order to lessen the sensitivity of con-
ventional GICs to water balance, and enhance their
physical properties of flexural and tensile strengths
and fracture toughness.
41
The incorporation of res-
in into the liquid can be achieved in several ways,
for instance, by the addition of a water-soluble po-
lymerizable methacrylate (eg, HEMA), or by the syn-
thesis of polyalkenoic acids with pendant methacry-
late side chains. The powder component is usually
essentially the same as for the self-curing GICs.
41
The first RM-GIC was a lining material (Vitrabond,
now Vitrebond; 3M/Espe, St Paul, MN, USA). It con-
sisted of a polycarboxylic liquid, with pendant meth-
Tyas
Vol 5, No 4, 2003 261
acryloxy groups. On mixing, the normal glass iono-
mer acid-base reaction occurred, but on exposure
to light, the methacryloxy groups cross linked and
provided immediate set. In practice, the photopoly-
merization reaction occurred before the completion
of the acid-base reaction, the latter then continuing
after clinical set.
28
The setting reaction of RM-GIC is therefore two-
fold; an acid-base reaction and a resin-polymeriza-
tion reaction. In turn, the latter can be initiated by
a photopolymerization reaction or by an oxidation-re-
duction (self-curing) reaction. Following clinical
set by resin polymerization, the acid-base reaction
continues, but much more slowly. This is because
of the reduced amount of water in an RM-GIC com-
pared to a conventional GIC,
41
and because chem-
ical reactions take place more slowly in a solid
phase compared to a liquid phase. It has been sug-
gested that the self-cure polymerization reaction
comprises only about 15% of the entire process.
41
The addition of resin is highly significant in the con-
text of water balance. Following clinical set by pho-
topolymerization, the restoration can be trimmed
and finished provided that is kept wet at all times.
Dehydration still remains a significant problem, and
some manufacturers recommend the application
and photocuring of a layer of unfilled resin prior to
dismissing the patient.
41
Like all GICs, there will be
a slight decrease in translucency over the first 24 h
after placement, followed by an increase over the
next 7 days, resulting in an enhancement of esthet-
ics.
32
However, the shade may darken slightly.
41
Resin-modified glass-ionomer restorative and lut-
ing cements are now also available. Examples of
the former include Vitremer (3M/Espe), Photac-Fil
(3M/Espe) and Fuji II LC (GC International, Tokyo,
Japan), and examples of luting cements include
Rely X (3M/Espe) and Fuji Plus (GC International).
CLASSIFICATION
The GICs can be classified in a variety of ways, how-
ever, the most practical is based on the clinical us-
age of the materials.
32,58
Type I GICs are luting ce-
ments, and are characterized by rapid set and low
film thickness. They are available as both self-cur-
ing and resin-modified products, but as described
above, the latter do not have a photopolymerization
setting reaction.
Type II GICs are the restorative GICs, ie, they
are used as a final restorative material in prepared
cavities. Two subtypes are recognized; Type II.1
comprises the esthetic GICs with appropriate color
and translucency, and Type II.2 comprises the re-
inforced GICs. Type II.1 are available as both con-
ventional and resin-modified versions, and the
former are characterized by the prolonged matura-
tion reaction described above. Type II.2 GICs are
the metal-containing products described above, yet
despite the term reinforced, they are not signifi-
cantly stronger than their Type II.1 counterparts.
38
They do have better wear resistance than Type II.1
conventional GICs,
26
probably because the metal
reduces the coefficient of surface friction and thus
the potential for plucking out of glass particles,
and they are fast setting and resistant to early wa-
ter uptake.
In the last few years, high powder:liquid ratio
conventional GICs have been introduced, which are
sometimes called packable or high viscosity
GICs.
41
These products (eg, Fuji IX, GC; Ketac Mo-
lar, 3M/Espe; Chemflex, Dentsply) are self-curing
and relatively fast setting, and are promoted mainly
for small cavities in deciduous teeth,
40
temporary
restorations in permanent teeth, base and lining
applications, and for the Atraumatic Restorative
Treatment (ART) technique.
12
BONDING MECHANISM
As described earlier, zinc polycarboxylate cement
was the first dental material to exhibit a chemical
bond to the calcified tissues. On ionization, each of
the various GIC acids yields a carboxyl (COO-)
group, two of which can theoretically bond ionically
to a calcium ion (Ca++) in enamel or dentin. How-
ever, this is a simplistic explanation of the bonding
of conventional GIC, and additional bonding mech-
anisms may occur for RM-GIC.
32,55
Conventional Glass Ionomers
Theoretical mechanisms of adhesion include ad-
sorption, involving chemical bonding (primary bonds:
ionic, covalent; secondary bonds: hydrogen, Van der
Waals) and diffusion (bonding between mobile mol-
ecules across the interface).
55
A fundamental requirement of effective adhesion
in any adhesive technology is a clean substrate sur-
face.
55
In addition, the surface energy of the sub-
strate (enamel or dentin) should exceed the sur-
Tyas
262 The Journal of Adhesive Dentistry
face tension of the adhesive (the GIC). In the con-
text of GIC, the tooth surface must therefore be
free from biofilm, blood, saliva, and other contami-
nants. Cleaning of unprepared cavities (eg, noncar-
ious cervical lesions) is best achieved by a brief ex-
posure to a slurry of pumice and water on a slowly
rotating brush or rubber cup. Commercial polishing
pastes are not recommended as they may create a
smear layer
32
or leave an oily residue. All cavities
should then be conditioned with an appropriate
conditioner. Several organic acid conditioners have
been suggested over the last 25 years, such as cit-
ric acid. However, the current consensus is that
polyacrylic acid is the most appropriate, at a con-
centration of ca 10%. Polyacrylic acid conditioner
will remove surface smear layer, leaving the smear
plugs in the dentin tubules essentially intact.
55
Ad-
ditional benefits of conditioning with polyacrylic
acid are that it increases the energy of the tooth
surface and therefore enhances wettability by the
freshly mixed GIC.
Placement of the freshly mixed GIC onto condi-
tioned calcified tissue results in an initial polar at-
traction, with weak hydrogen bonds predominat-
ing.
55
The free acid in the cement dissolves any re-
sidual smear layer, yet there is minimal demineral-
ization because of the buffering action of the phos-
phate ions from hydroxyapatite and because poly-
acrylic acid is quite weak.
55
As phosphate ions
(negative) are displaced from the hydroxyapatite,
calcium ions (positive) are also displaced in order
to maintain electrolytic balance, and both calcium
and phosphate ions are adsorbed into the cement.
This results in an ion-exchange
59
layer, which is
strongly bound to the hard tissue and to the ce-
ment.
32,55
However, one of the problems of investi-
gating the ion-exchange layer is that conventional
scanning electron microscopy requires dehydration
of the specimen, resulting in shrinkage and crack-
ing.
36
New environmental imaging techniques
are now increasing our understanding of GIC adhe-
sive mechanisms.
36
The nature of the ion-exchange layer
59
has re-
ceived attention from various authors, and several
names have been ascribed to it. This has led to
some confusion in the literature with mechanisms
of resin composite bonding, particularly to dentin.
Using a scanning electron microscope equipped
with a high-resolution field emission gun and a cry-
ostage, in order to stablise water in the GIC, Ngo et
al
36
described a zone of interaction between GIC
and enamel/dentin. This layer was 1 to 2 m thick,
resistant to acid etching, but of unknown nature. It
seems probable that it was equivalent to the ion
exchange layer of Wilson et at.
59
Subsequently,
Ferrari and Davidson
10
noted a similar acid-resis-
tant layer between GIC and enamel/dentin, approx-
imately 6 m thick, which they termed the inter-dif-
fusion zone. The nature of this zone was not dis-
cussed.
Hosoya and Garcia-Godoy
17
employed SEM to
study the interface between two conventional GICs
and enamel/dentin, following pretreatment with
20% polyacyrlic acid/3% aluminium chloride solu-
tion. They were unable to demonstrate the interdif-
fusion zone of Ferrari and Davidson,
10
but indicat-
ed that the interdiffusion zone could also be
termed the hybrid layer, analogous to that formed
between resin-modified GIC and dentin
37
(see be-
low). This has given rise to some confusion regard-
ing bonding systems (GIC and resin composite),
and it is recommended that the term hybrid layer
should not be used in relation to conventional GIC;
ion-exchange layer is preferred.
The nature of the ion-exchange layer was investi-
gated by Sennou
43
using x-ray photoelectron spec-
trometry (XPS), and the name interphase given.
The interphase was found to consist of reciprocal
diffusion of ions from GIC and dentin. The dentin
yielded calcium and phosphate ions, and the glass
ionomer yielded aluminium, silica and fluoride ions,
together with calcium and/or strontium, depending
on the glass composition.
A further XPS study
61
clearly demonstrated a
chemical bond between the carboxyl groups of poly-
alkenoic acid and the calcium component of hy-
droxyapatite (HA). These authors used HA instead
of enamel/dentin, in order to validate the instru-
mentation and thus the findings. The carboxyl
groups were found to replace the phosphate ions of
the HA in creating these ionic bonds, and about
68% of the carboxyl groups became bonded.
Tays group
49
further investigated the ion ex-
change layer, and proposed that various zones
and layers described in the literature
10,36,43
were
equivalent. Geiger,
15
using Fourier-transformed in-
frared spectroscopy, identified this layer as fluori-
dated carbonatoapatite. Tay et al
49
confirmed the
presence of the interphase or intermediate lay-
er, as containing elements of bidirectional diffu-
sion of ions between GIC and dentin. A further
study
60
using SEM and transmission electron mi-
croscopy measured the intermediate layer at 0.5
to 1.5 m thick (in agreement with Ngo et al
36
but
Tyas
Vol 5, No 4, 2003 263
not with Ferrari and Davidson
10
). In addition, na-
nometre-sized plate-like particles were identified
within the intermediate layer, which they speculat-
ed may be the reprecipitation of calcium and phos-
phate.
In addition to the invigorating bonding mecha-
nisms described, chemical bonding of GIC to col-
lagen has also been proposed.
34
This possibility
does not appear to have been further investigated.
The cohesive strength of the ion-exchange layer
is unknown,
41,55
since in laboratory bond strength
testing, cohesive failure of the cement is usually re-
sponsible for bond failure.
49
This finding greatly
complicates the evaluation and the significance of
bond strength testing of GICs, since measured
bond strengths are usually in the range of 3 to
10 MPa (ie, the cohesive strength of the GIC),
whereas for resin-dentin bonds adhesive failure of
> 30 MPa is frequently reported.
4
Despite these
laboratory values, GIC has excellent clinical reten-
tion in noncarious cervical lesions.
Resin-modified GICs
Since resin-modified glass-ionomer cements (RM-GIC)
contain a polymerizable resin, commonly 2-hydroxy-
ethyl methacrylate (HEMA), the theoretical opportuni-
ty exists for bonding to occur through micromechani-
cal retention in the same way as resin composite.
Lim et al
34
used a range of imaging techniques
(SEM, confocal microscopy, XPS and secondary ion
mass spectrometry) to investigate the interface be-
tween an RM-GIC and dentin. They reported the
presence of resin tags in the dentinal tubules, and
also that an ion-exchange process had taken
place, but no resin-dentin hybrid layer was found.
Friedl et al
14
were also unable to demonstrate hy-
brid layer formation, and confirmed the presence
of resin tags about 3 m long in the dentin tu-
bules.
Titley et al
50
investigated separately the liquid
only and the mixed cement, and their effect on den-
tin. The liquid initiated a mild peritubular etch and
the formation of tubular plugs. Continued expo-
sure to the liquid (> 30 s) resulted in progressive
disintegration of the plugs, possibly due to an ef-
fervescent reaction. The authors concluded that
bonding is primarily chemical, but that the surface
demineralization caused by the liquid component
prior to setting may allow improved penetration of
HEMA and later mechanical interlocking. Howev-
er, it is not clear whether the authors meant inter-
locking into the dentinal tubules or into the col-
lagen of the tubules.
Pereira et al
37
used SEM to examine the
RM-GIC/enamel and dentin interfaces. Following
treatment of the substrate with 20% polyacrylic ac-
id/3% aluminium chloride, resin tags were found in
the dentinal tubules. In addition, a resin-rich layer
and an indistinct zone were observed between
the RM-GIC and the dentin. The authors implied,
but did not specifically state, that the resin-rich lay-
er was the penetration of resin into the collagen lay-
er, ie, the classical hybrid layer in resin compos-
ite bonding. Regarding enamel, no results of the
SEM examination were given. However, the authors
implied that they identified micromechanical inter-
locking as being responsible for bonding.
Ramos and Perdigo
39
used field-emission SEM
in a study of the bonding of an RM-GIC to dentin.
They reported no penetration into dentin, and no
resin-dentin interdiffusion zone (hybrid layer). Col-
lagen fibres exposed by the 10% citric acid condi-
tioner remained exposed, and the authors suggest-
ed that the measured bond strength (2.9 MPa) was
due to ionic bonding.
Sidhu and Watson
46
and Sidhu et al
44
examined
the interface between RM-GIC and tooth tissue us-
ing confocal microscopy and 2-photon imaging. The
authors described a structureless or absorp-
tion layer in the cement adjacent to cut dentin tu-
bules, up to 25 m thick, although one of the four
RM-GICs tested did not show this feature. It was
also absent in association with enamel. There was
no suggestion by the authors that this absorption
layer was relevant to the bonding mechanism. No
evidence for resin tag formation into enamel or den-
tin was found.
Currently, therefore, there is no overwhelming ev-
idence that RM-GICs bond by the formation of a hy-
brid layer (ie, the infiltration of the resin component
into collagen exposed by acid pretreatment). Saito
et al
41
have published scanning electron micro-
graphs claiming to show hybrid layer formation for
RM-GICs but not for conventional GICs, but the ex-
perimental details were not given.
BIOLOGICAL PROPERTIES
The biocompatibility of GICs has been extensively
reviewed by Sidhu and Schmalz,
45
and the reader
is referred to this excellent paper for more detail.
Tyas
264 The Journal of Adhesive Dentistry
CLINICAL PERFORMANCE
In terms of adhesion in non-undercut noncarious
cervical lesions, the GICs have shown excellent
performance.
25,29,51
The technique is easy and
quick, in contrast to the more demanding resin-den-
tin bonding systems.
19
However, the conventional
GICs do suffer from a number of disadvantages as
discussed above; low fracture toughness, subopti-
mum polish and sensitivity to water balance for sev-
eral months. Despite the benefits gained by incor-
porating resin to produce the RM-GIC, the problems
of strength and polish remain.
32
Consequently, the
ideal situations for GICs are in non-stress-bearing
areas, ie, anterior approximal cavities and cervical
cavities. Although they can be used in occlusal and
posterior approximal cavities, this should only be
done after a careful assessment of the anticipated
load. The clinical procedure for the placement of
conventional and resin-modified GICs in all the
above cavity types has been described in detail
elsewhere,
19,32
and successful clinical perfor-
mance for 10 years
25
and 15 years
31
for conven-
tional GICs has been reported. However, clinical
studies on RM-GICs are fewer and short term, be-
cause of the more recent introduction of these
products.
1-3,5,6,9,11,18,23,33,53
To date, there is no
real consistency among the results, both for differ-
ent brands of RM-GIC and for comparative studies
of RM-GIC and polyacid-modified resin composites.
One criticism which has been made against the
RM-GICs is the potential for water uptake and swell-
ing due to the HEMA content.
32
However, the clini-
cal significance, if any, is unclear.
The recent trend towards minimum-intervention
dentistry
52
has focused increasing interest on the
GICs.
32
One of the main tenets of minimum-inter-
vention dentistry is the concept of removing only
infected dentin when restoring a carious lesion,
and leaving the affected dentin.
24
Affected dentin
is demineralized, but the collagen remains intact.
Such dentin is clinically stained, but firm and dry,
and has the potential to remineralize.
13
Recent pre-
liminary studies suggest that the fluoride released
from GIC, in conjunction with the metal ion originat-
ing from the glass (eg, calcium or strontium), may
promote such remineralisation.
35
This has particu-
lar relevance in the Atraumatic Restorative Treat-
ment (ART) Technique,
13
which is being strongly
promoted in environments where conventional den-
tal facilities are not available. Such environments
can include developing countries, remote rural
communities, nursing homes, and for the home-
bound, where, for example, rotary instruments and
mixing machines are not available. The ART tech-
nique also plays an important role in the treatment
of young children, dental phobics and the medi-
cally compromized, and has additional potential in
conjunction with chemomechanical caries removal
systems.
22,42
The fundamentals of the ART tech-
nique are removal of caries with hand instruments
(thereby leaving the affected dentin), and restora-
tion of the resulting cavity with GIC.
TYPE I AND III GLASS-IONOMER CEMENTS
This review has concentrated mainly on Type II (re-
storative) GICs. Space constraints do not allow a de-
tailed evaluation of Type I (luting) or Type III (liner,
base, and fissure sealant) GICs. Nevertheless, GICs
are important in these applications. The cervical
lining technique was recommended in 1984,
26
and
clinical trials are demonstrating its usefulness in
maintaining the integrity of the cervical margin of ap-
proximal posterior cavities, provided that an appro-
priate material is used.
20,30,54,56
Glass-ionomer ce-
ments are also useful as fissure sealants where the
strict moisture control for resin-based sealants can-
not be obtained, eg, for partly erupted teeth.
47
Al-
though the clinical retention rate of GICs is less than
that of resin sealants, the prevention of fissure car-
ies appears comparable.
47
The use of GICs in the
deciduous dentition has been discussed extensive-
ly elsewhere.
16
REFERENCES
1. Abdalla AI, Alhadainy HA. Clinical evaluation of hybrid ionomer
restoratives in Class V abrasion lesions: Two-year results.
Oper Dent 1997;28:255-258.
2. Avila G, Navarro MF, Palma RG, Franco EB. Clinical evaluation
of a conventional versus a resin-modified glass ionomer ce-
ment [abstract 381]. J Dent Res 1996;75:65.
3. Barnes DM, Blank LW, Gingell JC, Gilner PP. A clinical evalu-
ation of a resin-modified glass ionomer restorative material.
J Am Dent Assoc 1995;126:1245-1253.
4. Bouillaguet S, Gysi P, Wataha JC, Ciucchi B, Cattani M, Godin
C, Meyer JM. Bond strength of composite to dentin using con-
ventional, one-step, and self-etching adhesive systems. J
Dent 2001;29:55-61.
5. Brackett MG, Dib A, Brackett WW, Estrada BE, Reyes AA.
One-year clinical performance of a resin-modified glass ion-
omer and a resin composite restorative material in unpre-
pared Class V restorations. Oper Dent 2002;27:112-116.
Tyas
Vol 5, No 4, 2003 265
6. Brackett WW, Browning WD, Ross JA, Brackett MG. Two-year
clinical performance of a polyacid-modified resin composite
and a resin-modified glass-ionomer restorative material.
Oper Dent 2001;26:12-16.
7. Buonocore M. A simple method of increasing the adhesion
of acrylic filling materials to enamel surfaces. J Dent Res
1955;34:849-853.
8. Chandler HH, Bowen RL, Paffenbarger GC, Mullineaux AL.
Clinical investigation of a radiopaque composite restorative
material. J Am Dent Assoc 1970;81:935-940.
9. Donly KJ, Segura A, Kanellis M, Erickson RL. Clinical perfor-
mance and caries inhibition of resin-modified glass ionomer
cements and amalgam restorations. J Am Dent Assoc 1999;
10:1459-1466.
10. Ferrari M, Davidson CL. Interdiffusion of a traditional glass
ionomer cement into conditioned dentin. Am J Dent 1997;10:
295-297.
11. Folwaczny M, Loher C, Mehl A, Kunzelmann KH, Hickel R.
Tooth-colored filling materials for the restoration of cervical
lesions: A 24-month follow-up study. Oper Dent 2000;25:
251-258.
12. Frencken J, Pilot T, Songpaisan Y, Phantumvanit P. Atraumatic
restorative treatment (ART): Rationale, technique and devel-
opment. J Pub Health Dent 1996;56:135-140.
13. Frencken JE, Holmgren CJ. Atraumatic restorative treatment
for dental caries. 1999, Nijmegen: STI Book b.v.
14. Friedl K-H, Powers J, Hiller K-A. Influence of different factors
on bond strength of hybrid ionomers. Oper Dent 1995;20:
74-80.
15. Geiger SB, Weiner S. Fluoridated carbonatoapatite in the in-
termediate layer between glass ionomer and dentin. Dent
Mater 1993;9:33-36.
16. Hickel R, Manhart J. Glass-ionomers and compomers in pe-
diatric dentistry. In: CL Davidson, IA Mjr (eds). Advances in
glass-ionomer cements. Chicago: Quintessence Publishing,
1999:201-226.
17. Hosoya Y, Garcia-Godoy F. Bonding mechanism of ketac-molar
aplicap and fuji IX gp to enamel and dentin. Am J Dent 1998;
11:235-239.
18. Jefferies SR, Marier RP. Clinical evaluation of a VLC glass ion-
omer: One year results. J Dent Res 1994;73:184.
19. Jonas P, Lasfargues J-J, Degrange M. Clinical applications of
glass-ionomer cements: Class III and Class V restorations.
In: CL Davisdson, IA Mjr (eds). Advances in glass-ionomer
cements. Chicago: Quintessence Publishing, 1999:227-252.
20. Knibbs PJ. The clinical performance of a glass polyalkenoate
(glass ionomer) cement used in a sandwich technique with
a composite resin to restore Class II cavities. Brit Dent J
1992;172:103-107.
21. Knibbs PJ, Plant CG, Pearson CG. A clinical assessment of
an anhydrous glass ionomer cement. Brit Dent J 1986;161:
99-103.
22. Kubo S, Li H, Burrow MF, Tyas MJ. Nanoleakage of dentin ad-
hesive systems bonded to Carisolv-treated dentin. Oper Dent
2002;27:387-395.
23. Maneenut C, Tyas MJ. Clinical evaluation of resin-modified
glass-ionomer restorative cements in cervical abrasion le-
sions: One-year results. Quintessence Int 1995;26:739-743.
24. Massler M. Changing concepts in the treatment of carious
lesions. Br Dent J 1967;123:547-548.
25. Matis BA, Cochran M, Carlson T. Longevity of glass-ionomer
restorative materials: Results of a 10-year evaluation. Quin-
tessence Int 1996;27:373-382.
26. McLean JW, Gasser O. Glass-cermet cements. Quintessence
Int 1985;16:333-343.
27. McLean JW, Nicholson JW, Wilson AD. Proposed nomencla-
ture for glass-ionomer dental cements and related materials.
Quintessence Int 1994;25:587-589.
28. Mitra SB. Adhesion to dentin and physical properties of a
light-cured glass-ionomer liner/base. J Dent Res 1991;70:
72-74.
29. Mount GJ. Longevity of glass ionomer cements. J Prosthet
Dent 1986;55:682-685.
30. Mount GJ. Clinical requirements for a successful sandwich
dentine to glass-ionomer cement to composite resin. Aust
Dent J 1989;34:159-165.
31. Mount GJ. Longevity in glass-ionomer restorations: Review
of a successful technique. Quintessence Int 1997;28:
643-650.
32. Mount GJ. An atlas of glass-ionomer cements. A clinicians
guide. 3rd ed. London: Martin Dunitz, 2002.
33. Neo J, Chew C. Direct tooth-colored materials for noncarious
lesions: A 3-year clinical report. Quintessence Int 1996;27:
183-188.
34. Nezu T, Winnik FM. Interaction of water-soluble collagen with
poly(acrylic acid). Biomaterials 2000;21:415-419.
35. Ngo H. Biological properties of glass-ionomers. In: GJ Mount
(ed). An atlas of glass-ionomer cements. A clinicians guide.
London: Martin Dunitz, 2002:43-55.
36. Ngo H, Mount G, Peters M. A study of glass-ionomer cement
and its interface with enamel and dentin using a low-temper-
ature, high-resolution scanning electron microscope tech-
nique. Quintessence Int 1997;28:63-69.
37. Pereira PNR, Yamada T, Tei R, Tagami J. Bond strength and
interface micromorphology of an improved resin-modified
glass ionomer cement. Am J Dent 1997;10:128-132.
38. Peutzfeldt A. Compomers and glass ionomers: Bond strength
to dentin and mechanical properties. Am J Dent 1996;9:
259-263.
39. Ramos JC, Perdigao J. Bond strength and SEM morphology
of dentin-amalgam adhesives. Am J Dent 1997;10:152-158.
40. Rutar J, McAllan L, Tyas MJ. Three-year clinical performance
of glass ionomer cement in primary molars. Int J Paediatr
Dent 2002;12:146-147.
41. Saito S, Tosaki S, Hirota K. Characteristics of glass ionomer
cements. In: CL Davidson, IA Mjr (eds). Advances in
glass-ionomer cements. Chicago: Quintessence Publishing,
1999:15-50.
42. Sakoolnamarka R, Burrow MF, Kubo S, Tyas MJ. Morpholog-
ical study of demineralized dentine after caries removal using
two different methods. Aust Dent J 2002;47:116-122.
43. Sennou HE, Lebugle AA, Gregoire GL. X-ray photoelectron
spectroscopy study of the dentin-glass ionomer cement in-
terface. Dent Mater 1999;15:229-237.
44. Sidhu SK, Pilecki P, Cheng PC, Watson TF. The morphology
and stability of resin-modified glass-ionomer adhesive at the
dentin/resin-based composite interface. Am J Dent 2002;
15:129-136.
45. Sidhu SK, Schmalz G. The biocompatibility of glass-ionomer
cement materials. A status report for the American Journal
of Dentistry. Am J Dent 2001;14:387-396.
Tyas
266 The Journal of Adhesive Dentistry
46. Sidhu SK, Watson TF. Interfacial characteristics of resin-mod-
ified glass-ionomer materials: A study on fluid permeability
using confocal fluorescence microscopy. J Dent Res 1998;
77:1749-1759.
47. Smales RJ, Gao W, Ho FT. In vitro evaluation of sealing pits
and fissures with newer glass-ionomer cements developed
for the ART technique. J Clin Pediatr Dent 1997;21:321-323.
48. Smith DC. A new dental cement. Brit Dent J 1968;125:
381-385.
49. Tay FR, Smales RJ, Ngo H, Wei SH, Pashley DH. Effect of dif-
ferent conditioning protocols on adhesion of a GIC to dentin.
J Adhes Dent 2001;3:153-167.
50. Titley KC, Smith DC, Chernecky R. SEM observations of the
reactions of the components of a light-activated glass poly-
alkenoate (ionomer) cement on bovine dentine. J Dent 1996;
24:411-416.
51. Tyas MJ. Clinical studies related to glass ionomer cements.
Oper Dent 1992;5:191-198.
52. Tyas MJ, Anusavice KJ, Frencken JE, Mount GJ. Minimal in-
tervention dentistry a review. Int Dent J 2000;50:1-12.
53. Van Dijken JWV. 3-year clinical evaluation of a compomer, a
resin-modified glass ionomer and a resin composite in Class
III restorations. Am J Dent 1996;9:195-198.
54. Van Dijken JWV, Kieri C, Carlen M. Longevity of extensive
Class II open-sandwich restorations with a resin-modified
glass-ionomer cement. J Dent Res 1999;78:1319-1325.
55. Watson T. Bonding of glass-ionomer cements to tooth struc-
ture. In: CL Davidson, IA Mjr (eds). Advances in glass-iono-
mer cements. Chicago: Quintessence Publishing, 1999:
121-136.
56. Welbury RR, Murray JJ. A clinical trial of the glass ionomer
cement-composite resin 'sandwich' technique in Class II cav-
ities in permanent premolar and molar teeth. Quintessence
Int 1990;21:507-512.
57. Wilson AD, Kent BE. The glass-ionomer cement. A new trans-
lucent cement for dentistry. J App Chem Biotech 1971;21:
313.
58. Wilson AD, McLean JW. Glass-ionomer cement. Chicago:
Quintessence Publishing, 1988.
59. Wilson AD, Prosser HJ, Powis DM. Mechanism of adhesion
of polyelectrolyte cements to hydroxyapatite. J Dent Res
1983;62:590-592.
60. Yip HK, Tay FR, Ngo HC, Smales RJ, Pashley DH. Bonding of
contemporary glass ionomer cements to dentin. Dent Mater
2001;17:456-470.
61. Yoshida Y, Van Meerbeek B, Nakayama Y, Snauwaert J, Helle-
mans L, Lambrechts P, Vanherle G, Wakasa K. Evidence of
chemical bonding at biomaterial-hard tissue interfaces. J
Dent Res 2000;79:709-714.