dental

materials
Dental Materials 19 (2003) 212±217
www.elsevier.com/locate/dental

Zinc sulfate addition to glass-ionomer-based cements: in¯uence on

physical and antibacterial properties, zinc and ¯uoride release
Prudencio W.R. Osinaga a, Rosa Helena M. Grande b,*, Rafael Y. Ballester b,
Maria Regina L. Simionato c, CeÂlia Regina M. Delgado Rodrigues d, Antonio Muench b
a
Department of Operative Dentistry, Federal University of Pelotas, Rio Grande do Sul, Brazil
b
Department of Dental Materials, University of SaÄo Paulo, SaÄo Paulo, Brazil
c
Department of Microbiology, Institute of Biomedical Sciences, University of SaÄo Paulo, SaÄo Paulo, Brazil
d
Department of Pedodontics, University of SaÄo Paulo, SaÄo Paulo, Brazil
Received 21 May 2001; revised 8 October 2001; accepted 11 December 2001

Abstract
Objectives: The aim of this study is to evaluate the effect of ZnSO4 addition to a conventional glass ionomer and a resin-modi®ed glass
ionomer on solubility, ¯exural strength, zinc and ¯uoride (F) release, and Streptococcus mutans growth inhibition.
Methods: 5 or 10% ZnSO4 was added to Vitremer and Ketac-Fil powders. Solubility test was performed based on ISO 7489. Flexural
strength was determined by 3-point bending test based on ISO 4049. Zn release/uptake was determined by atomic emission spectrometry; F
release/uptake was measured using a F-speci®c electrode. Both release measurements were performed for 15 d before and 15 d after
recharging. Antibacterial test was conducted according to agar plate methods against S. mutans, by measuring the inhibition halos in 1-h
and 15-d specimens. Data were analyzed by ANOVA.
Results: Solubility increased with higher ZnSO4 content, but remained below the ISO 7489 limit. Flexural strength was not affected by
ZnSO4 addition, and Vitremer performed better than Ketac-Fil. The control materials released no zinc. Vitremer with 10% ZnSO4 released
the highest amount of zinc. Fluoride release was similar for Ketac-Fil and Vitremer. In both cases, the highest amounts were released in the
®rst 24 h. The growth inhibition halo of S. mutans was similar for both materials with highest content of ZnSO4 and occurred only with 1-h
specimens.
Signi®cance: Zinc addition decreased microorganisms growth and improved ¯uoride release, without signi®cantly affecting the materials'
¯exural strength and solubility.
q 2003 Published by Elsevier Science Ltd on behalf of Academy of Dental Materials.
Keywords: Glass-ionomer-based cements; Zinc and ¯uoride release; Streptococcus mutans growth inhibition; Solubility; Flexural strength

1. Introduction risk of caries should receive a comprehensive treatment that

Caries disease still remains a major public health problem
despite the widespread use of ¯uoride and the decline in
caries prevalence observed in many countries. Minimally
invasive dentistry adopts a philosophy that includes preven-
tion, remineralization and invasive surgical approach. Such
integrated approach is necessary because restoring the tooth
does not cure the disease, and `permanent' restorations often
need replacement [1,2].
Patients having active caries lesions or who are at high
* Corresponding author. Address: Departamento de Materiais DentaÂrios,
Faculdade de Odontologia da USP, Av. Prof. Lineu Prestes, 2.227ÐCidade
UniversitaÂria, 05508-900 SaÄo Paulo, SP, Brazil. Tel./fax: 155-11-
30917840.
E-mail address: grande@usp.br (R.H.M. Grande).
0109-5641/03/$30.00 + 0.00 q 2003 Published by Elsevier Science Ltd on behalf of Academy of Dental Materials.
PII: S01 09- 5641(02)0003 2-5
arrests all the etiological factors of the disease (e.g. hygiene
orientation, diet counseling, ¯uoride and/or antibacterial
use, increase of salivary ¯ow). The initial procedure also
must include a treatment of the cavities with ¯uoride-
releasing ®lling materials, in order to apply a strong initial
`shot' on the cariogenic micro¯ora [3], and to prevent the
restoration cycle that often leads to recurrent caries lesions
[2].
The development of glass-ionomer cement (GIC), during
the 70's, offered a restorative material with long-term adhe-
sion to tooth structure, combined with signi®cant ¯uoride
(F) release. More recently, resin-modi®ed glass-ionomer
cement (RMGIC) was introduced. Besides presenting a
similar level of F release, they have also overcome some
unfavorable physical properties of GIC [4].
According to some authors, F release is the main factor
 P.W.R. Osinaga et al. / Dental Materials 19 (2003) 212±217
involved in the antibacterial activity of GIC/RMGIC [5,6].
However, for other authors F may not be responsible for the
anticariogenic effect, since greater antibacterial property
was found when the liquid of a RMGIC was tested alone
[7]. On the other hand, there are those who believe that the
zinc (Zn) present in the GIC could in¯uence the antibacterial
activity through its interaction with F [5±8].
Considering that recurrent caries lesions remain the major
reason for restoration replacement [9], during the last
decade conservative techniques have been enforced, such
as tunnel and slot preparations, as well as the use of
atraumatic restoration technique (ART) in the developing
countries [1,3]. Some attempts at adding substances that
interfere with metabolic activity or in the adhesion of dental
plaque to dental materials, have been reported in the litera-
ture (e.g. ¯uoride in the form of a ®ller, silver-supported
agents to UDMA/TEGDMA composite resin [9], or chlor-
hexidine to composite or GIC [10]).
It is known that Zn occurs in enamel, plaque and saliva. In
enamel, Zn has a similar distribution to ¯uoride and lead,
with a higher concentration on its surface. When Zn is
present in the culture medium, it may inhibit Streptococcus
mutans growth and plaque formation [11]. When incorpo-
rated into dentifrices and oral rinse solutions, Zn interferes
with the following mechanisms: (1) enzymatic metabolism
of S. mutans; (2) plaque acidogenicity (transient action); (3)
inhibition of plaque mineralization by controlling dental
calculus formation (adsorption of a Zn-monolayer on apatite
crystal), without counteracting the F bene®ts [12].
The aim of this study was to analyze the effects of zinc
sulfate addition to GIC and RMGIC on their physical prop-
erties, inhibition of S. mutans growth, and F and Zn release
(with and without recharge).
2. Material and methods
The cements used in this study were a conventional glass-
ionomer cement, Ketac-Fil (ESPE, Seefeld, Germany; batch
FW0042155 and FW0042158), and a resin-modi®ed glass-
ionomer cement, Vitremer (3M Dental Products, St Paul,
MN, USA; batches 19970521 and 19960604). The standard
cements (K0 and V0) were modi®ed by the addition of 5-
wt% (K5 and V5) or 10-wt% (K10 and V10) of ZnSO4 (Qeel
Industria Quimica, SaÄo Paulo, SP, Brazil).
2.1. Solubility test
The cements were mixed according to a 1:1 powder/
liquid ratio (in weight) and dispensed into a polyethylene
split ring mold (20 mm £ 1.5 mm) which included a piece of
stainless steel wire. The specimens of RMGIC were irra-
diated for 40 s (450 mW/cm 2). The other procedures for this
test followed the ISO 7489 [13]. Three specimens were
prepared using each one of the six materials (18 in total).
213
2.2. Flexural test
The materials were mixed according to the manufac-
turers' instructions and dispensed into a stainless steel split
mold (10 mm £ 2 mm £ 1 mm). Except for coating the speci-
mens with varnish [14], and storage at 37 8C for 2 days in
arti®cial saliva, the other procedures were based on the ISO
4049 [15]. The 3-point loading test was performed at a rate
of 1 N/min in a testing machine (model E2R-30, Wolpert,
Ludwigshafen, Germany). The ¯exural strength was calcu-
lated in MPa. Six specimens were prepared for each one of
the six materials (36 in total).
2.3. Test of ¯uoride and zinc release
To evaluate zinc and ¯uoride release/uptake, 72 speci-
mens were prepared, 36 for each ion measured. The mixed
cement was dispensed into a cylindrical polyethylene mold
(4 mm £ 3 mm) and kept in a 100% humid environment for
1 h. After that, the specimens were removed from the molds,
immersed in individual plastic vials containing arti®cial
saliva (1 ml) and incubated at 37 8C. Fluoride and zinc
released were measured every 24 h over the 15-day test
period. Arti®cial saliva was changed daily. Every time the
arti®cial saliva was changed, the specimens were removed
from the vials, rinsed with deionized water for 4 s, dried
with absorbent paper and re-immersed in new arti®cial
saliva at 37 8C.
Fluoride was measured by a F-ion selective electrode
(model 96-09, Orion Inc., Boston, MA, USA). The system
was calibrated prior to each evaluation with ¯uoride stan-
dard solutions. One milliliter of TISAB II (Orion Inc.,
Boston, MA, USA) was added to each one of the vials con-
taining 1 ml of arti®cial saliva. Zinc release was determined
by inductively coupled argon plasma atomic emission
spectrometry (Spectro¯ame, Spectro Co., Klecve, Germany)
under the following conditions: forward power 1.2 kW;
refrigerant gas ¯ow 12 ml/min; sample ¯ow rate 1.5 ml/
min; observation height 12 mm above the coil.
Recharge of the specimens was conducted on the 15th
day. The specimens were removed from the vials, rinsed
with deionized water, dried and then randomly immersed
in one of two solutions: 1500 ppm NaF for 4 min or
1500 ppm NaF plus 10 mM ZnSO4 for 4 min. After the
immersion period, the samples were rinsed, placed in vials
containing arti®cial saliva, incubated and subjected to the
same procedure described above, for another 15-day period.
2.4. Antibacterial activity
The antibacterial activity was evaluated against the S.
mutans ATCC 25175 (American Type Culture Collection,
Rockville, MD, USA). All procedures were carried out
under aseptic conditions in a laminar air¯ow cabinet.
Ninety-six specimens …n ˆ 4† were prepared as for the
zinc and ¯uoride release tests previously described.
Microorganisms were grown from frozen stock cultures
214 P.W.R. Osinaga et al. / Dental Materials 19 (2003) 212±217

Fig. 1. Solubility average values. Means with the same letters are statistically similar …p . 0:05†:

in TSB (Tryptic Soy Broth, Difco Laboratories, Detroit, MI,
USA) at 37 8C for 48 h in a candle jar. Cells were washed in
PBS by centrifuging at 3000 £ g, 4 8C, for 5 min (Eppen-
dorf, model 5804 R, Hamburg, Germany). Inocula were
prepared by the re-suspension of washed cells to a pre-
determined optical density in a spectrophotometer (Spectro-
nic Instruments, model 20 Genesys, Rochester, NY, USA) at
a wavelength of 550 nm, to relate to known concentrations
of 2 £ 10 10 cfu/ml.
Aliquots of 1.0 ml of the standardized bacterial suspen-
sion were mixed with 25 ml of sterile TSA (Tryptic Soy
Agar, Difco Laboratories, Detroit, MI, USA), cooled to
45 8C, in sextuplicate. The resultant suspensions were
poured into six petri dishes and 24 1-h specimens were
placed at pre-destined areas before the solidi®cation.
The 72 remaining specimens were immersed in arti®cial
saliva for 15 days under a daily change regimen. On the
15th day, 24 specimens were distributed into agar plates, as
described earlier. Before being placed in petri dishes and incu-
bated, the remaining 48 specimens were recharged in one of
the two test solutions (NaF or NaF plus ZnSO4). After incuba-
tion on micro-aerophilic condition, the agar plates were exam-
ined for bacterial growth inhibition. The diameters of the
inhibition halos were measured using a dial caliper.

Fig. 2. Flexural strength average values. Means with the same letters are statistically similar …p . 0:05†:
 P.W.R. Osinaga et al. / Dental Materials 19 (2003) 212±217 215

Fig. 3. Fluoride release average values before/after recharging for all materials. Differences are not signi®cant in regions were it is impossible to decipher
differences between lines in the graph. (a). Fluoride release average values before/after recharging for Ketac-Fil. (b). Fluoride release average values before/
after recharging for Vitremer.

For all the experiments, data obtained at ®rst, third and
15th days, were statistically analyzed by ANOVA, using
Tukey test to detect pair wise differences …p , 0:05†:
3. Results
No signi®cant difference in solubility was found between
the two cements (0.28% for Ketac-Fil and 0.29% for
Vitremer) and no signi®cant interaction was observed.
The addition of ZnSO4 increased the solubility of both
materials and it was more evident with the highest percent-
age of Zn (K10 ˆ 0:59; V10 ˆ 0:56%) (Fig. 1)
For the ¯exural strength data, a signi®cant difference
between the cements was veri®ed. Vitremer showed a
mean ¯exure strength value of 64.0 MPa, nearly twice that
of Ketac-Fil (29.4 MPa). No signi®cant effects for the addi-
tion of ZnSO4 and no interaction between the main factors
were observed (Fig. 2).
Zn release values indicated signi®cant main effects, as
well as signi®cant interactions. The specimens of K0 and
V0 (e.g. no ZnSO4 added) showed average values of
216 P.W.R. Osinaga et al. / Dental Materials 19 (2003) 212±217
Fig. 4. Average values of S. mutans growth inhibition halos (mm). For X axis, ®rst letter indicates the material (K: Ketac-Fil; V: Vitremer); part between `-'
and `/' indicates the time elapsed between mixing and placing in contact with the culture in petri dishes (1 h: one hour; 15 d: 15 days); part after `/' indicates the
ions of recharge (F: ¯uoride; F 1 Zn: ¯uoride plus zinc).
0.0 ^ 0.0 ppm. Zn release different from zero occurred with
K5; V5 and K10 released similar amounts (0.9 ^ 0.5 and
0.8 ^ 0.7 ppm, respectively), and V10 ˆ 7:5 ^ 0:4 ppm;
only in the ®rst day. No increase in Zn release was observed
after recharging.
The analysis of the F release values indicated signi®cant
main effects, as well as the interactions. As shown in Fig. 3,
both materials released high amounts of ¯uoride, especially
up to the third day and from specimens with the highest
percentage of ZnSO4. Effect of recharging was observed
only on the ®rst day, especially in K0 group.
Overall S. mutans growth inhibition values indicated
signi®cant main effects. The highest inhibition effect was
observed in 1-h aged specimens of both materials, which
increased with 10% ZnSO4 addition, about 2.2 mm diameter
halos. No inhibitory effect could be found in the 15-day
specimens. After recharging, the inhibition increase was
not signi®cant.
4. Discussion
Fluoride-releasing esthetic materials that provide positive
results on microbial inhibition have attracted considerable
attention, and led to efforts towards the improvement of
their physical properties [1,4]. Some restorative techniques
using GIC imply an incomplete removal of the carious
tissues and, consequently, the risk of secondary caries is
increased [2,3]. Therefore, any additive with the ability to
enhance the antimicrobial effect of these materials without
counteracting their physical and F release properties would
be of great value.
The addition of 10% ZnSO4 to the cement powder
increased its solubility nearly ®ve times (Fig. 1). However,
the highest value (0.59%) remained below the maximum
value (0.70%) acceptable by ISO 7489 speci®cation [14].
Similar solubility values for GIC and RMGIC were
previously reported [4].
The strength of the cements was assessed by ¯exural test
because it was considered to be the most appropriate test for
this kind of material [14]. The higher ¯exural strength for
Vitremer (64.0 MPa) compared to Ketac-Fil (29.4 MPa) is
in agreement with previously reported studies [14,16].
The ZnSO4 addition, either 5 or 10%, did not reduce the
¯exural strength after 2 days of storage. However, it is
possible that the higher solubility shown by the materials
containing ZnSO4 could decrease ¯exural strength over
time. Considering that GIC and RMGIC are not indicated
as permanent materials for high stress bearing restorations,
the reduction in ¯exural strength values may not be clini-
cally relevant.
Specimens with higher ZnSO4 content showed higher Zn
release on the ®rst day. No increase in Zn release was
detected after specimen recharges. Vitremer released more
Zn than Ketac-Fil. There was no Zn release from the speci-
mens prepared with the original materials (K0 and V0),
although some authors had reported Zn presence in glass-
ionomer cements [5,6,8].
The amount and duration of F release depends upon the
test method used, composition of the materials, difference in
surface energy, and porosity [17]. Other critical factors that
could in¯uence the results would be the frequency in which
the storage solution is renewed, the interval between
changes, and the storage solutions [17,18].
 P.W.R. Osinaga et al. / Dental Materials 19 (2003) 212±217
The F release was higher during the ®rst day, followed by
a decrease to a nearly constant level, which agrees with
many other studies. The similar values of F release obtained
for both materials are in agreement with some reported
results [19], but differ from others [18]. Another fact that
could be observed is an overall parallel F release pro®le for
GIC and RMGIC, also previously reported [19]. Our results
showed that the higher the ZnSO4 addition, the higher the F
released up to the third day (Fig. 3).
The F released from the recharged specimens did not
produced F release values as high as that observed on
the ®rst day after mixing, which is in agreement with
other results [17,19]. After recharging, the K0 specimens
released higher amounts of F, compared to the other
materials. Other authors showed a high F release during
the ®rst week from specimens of GIC and RMGIC
recharged with APF gel [1].
Zn presence increased the growth inhibition halos for
both materials. The results showed (Fig. 4) that only
the freshly mixed materials could lead to signi®cant
growth inhibition of S. mutans. Other authors also
observed this fact [6]. There was a marked antibacterial
activity, but the inhibitory effect decreased abruptly
during the 15-day period. Even on the ®rst day after
recharging, the specimens showed no signi®cant anti-
bacterial effect.
The results of this ®rst study led us to conclude that
ZnSO4 addition to ionomer products showed signi®cant
inhibition of S. mutans growth, and increased F release.
Though an increase in solubility was observed, it was main-
tained below the requirements of the ISO 7489 and, more-
over, the ¯exural strength of the materials was not affected.
The Zn release occurred only during the ®rst 15-day period.
Further investigations must be conducted in order to evalu-
ate the clinical performance of the materials used in this
study.
Acknowledgements
This investigation was part of a thesis submitted to the
graduate faculty, University of SaÄo Paulo, as partial ful®ll-
ment of the requirements for the PhD degree. The supply of
Vitremer by 3M Dental Products, and the technical assis-
tance or advice provided by Dr Jose Nicolau, Dr Elizabeth
de Oliveira, Dr Roberto Braga and Douglas N. Souza was
appreciated. This study was partially supported by CAPES.
217
References
[1] Gao W, Smales RJ, Gale MS. Fluoride release/uptake from newer
glass-ionomer cements used with the ART approach. Am J Dent
2000;13:201±4.
[2] White JM, Eakle WS. Rationale and treatment approach in minimally
invasive dentistry. J Am Dent Assoc 2000;131:13S±9S.
[3] Forsten L. Fluoride release and uptake by glass-ionomers and related
materials and its clinical effect. Biomaterials 1998;19:503±8.
[4] Kim Y, Hirano S, Hirasawa T. Physical properties of resin-modi®ed
glass-ionomers. Dent Mater J 1998;17:68±76.
[5] Prati C, Fava F, Di Gioia D, Selighini M, Pashley DH. Antibacterial
effectiveness of dentin bonding systems. Dent Mater 1993;9:338±43.
[6] Loyola-Rodriguez JP, Garcia-Godoy F, Lindquist R. Growth inhibi-
tion of glass ionomer cements on mutans streptococci. Pediatr Dent
1994;16:346±9.
[7] Meiers JC, Miller GA. Antibacterial activity of dentin bonding
systems, resin-modi®ed glass ionomers, and polyacid-modi®ed com-
posite resins. Operat Dent 1996;21:257±64.
[8] Izaguirre-FernaÂndez EJ, Eisenberg AD, Curzon MEJ. Interactions of
zinc with ¯uoride on growth, glycolysis and survival of Streptococcus
mutans GS-5. Caries Res 1989;23:18±25.
[9] Ferracane JL. Status of research on new ®llers and new resins for
dental composites. In: Tagami J, Toledano M, Prati C, editors.
Advanced adhesive dentistryÐGranada international symposium.
Cirimido: Gra®che Erredue, 2000. p. 3±29.
[10] Jedrychowski JR, Caputo AA, Kerper S. Antibacterial and mechan-
ical properties of restorative materials combined with chlorhexidines.
J Oral Rehab 1983;10:373±81.
[11] Bates DG, Navia JM. Chemotherapeutic effect of zinc on Strepto-
coccus mutans and rat dental caries. Arch Oral Biol 1979;24:799±805.
[12] Ingram GS, Horay CP, Stead WJ. Interaction of zinc with dental
mineral. Caries Res 1992;26:248±53.
[13] ISOÐInternational Organization for Standardization. Dental glass
polyalkenoate cements. Switzerland: International Organization for
Standardization, 1986 p. 13 (International Standard, 7489).
[14] Momoi Y, Hirosaki K, Kohno A, McCabe JF. Flexural properties of
resin-modi®ed hybrid glass-ionomers in comparison with conven-
tional acid±base glass-ionomers. Dent Mater J 1995;14:109±19.
[15] ISOÐInternational Organization for Standardization. DentistryÐ
resinbased ®lling materials. Switzerland: International Organization
for Standardization, 1988 p. 11 (International Standard, 4049).
[16] Li J, Liu Y, Liu Y, SoÈremark R, SundstroÈm F. Flexure strength of
resin-modi®ed glass ionomer cements and their bond strength to
dental composites. Acta Odontol Scand 1996;54:55±8.
[17] Hatibovic-Kofman S, Koch G, Ekstrand J. Glass ionomer materials as
a rechargeable ¯uoride-release system. Int J Pediatr Dent 1997;7:65±
73.
[18] Carvalho AS, Cury JA. Fluoride release from some dental materials in
different solutions. Operat Dent 1999;24:14±9.
[19] SeppaÈ L, Korhonen A, Nuutinen A. Inhibitory effect on S. mutans by
¯uoride-treated conventional and resin-reinforced glass-ionomer
cements. Eur J Oral Sci 1995;103:182±5.