Effect of early water contact on solubility of glass ionomer luting cements

Deniz Gemalmaz, DDS, PhD,a Binnaz Yoruc, PhD,b Mutlu Ozcan, DDS,c and
Hasan N. Alkumru, DDS, PhDd
Faculty of Dentistry, Marmara University, Istanbul, Turkey
Statement of problem. Glass ionomer cements are susceptible to attack by moisture during the initial
setting period that can result in an increased solubility.
Purpose. This study was to evaluate the solubility of glass ionomer luting cements immersed in distilled
water at early stages after mixing.
Material and methods. Four commercial glass ionomer cements were used (Ketac-Cem, Fuji Ionomer I;
AquaCem; AquaMeron). For each material, 5 resin cement holders were made with 2 circular cavities
(diameter = 5 mm; depth = 2 mm). One minute after start of mixing, the specimens were placed in a
humidifier at 37°C and 100% relative humidity. After 2, 5, and 8 minutes storage time, they were immersed
in 50 mL of distilled water in a glass weighing bottle where they were stored for 3 hours at 37°C. Next, the
specimens were removed and the water was evaporated from the weighing bottle at 130°C for 2 hours. The
difference between the final and initial weights of the bottle was taken as amount of solubility. Data were
analyzed by analysis of variance and Duncan’s test.
Results. Solubility was greatest at 3 minutes immersion time and least at 9 minutes. The lowest cement
loss was for AquaCem at 9 minutes immersion time (1.68 mg/cm2). The greatest loss was for Fuji Ionomer
I after 3 minutes immersion time (14.98 mg/cm2). Solubility of cements decreased by 39% for Ketac-Cem,
61% for Fuji Ionomer I, 38% for AquaCem, and 37% for AquaMeron when the specimens were immersed in
water 6 minutes after mixing. Comparison of all 4 cements revealed a relatively large difference between loss
of substance from AquaCem and AquaMeron, and the 2 other conventional glass ionomer cements, Ketac-
Cem and Fuji Ionomer I.
Conclusions. Increasing the time from start of mixing until immersion in water from 3 to 9 minutes
resulted in a marked decrease in loss of substance from the surface of all 4 cements. Water-hardening glass
ionomer cements (AquaCem and AquaMeron) were less sensitive to early water contamination. (J Prosthet
Dent 1998;80:474-8.)

CLINICAL IMPLICATIONS
An extended period of prevention of saliva contamination is needed to control erosion
of glass ionomer luting agents and the most sensitive period for early disintegration is
the first 6 minutes after mixing.

S olubility is one of the most important factors in

assessing the quality of materials to be used as bases,
liners, luting agents, and restorations. Among these
dental materials, luting cements require more time to
set so they are more susceptible to intraoral saliva con-
tamination. Early cement exposure to saliva can cause a
product to change during cement setting that will
change the cement properties, and prevention of mois-
ture contact is not always possible.
This study was presented as a poster presentation at the 75th annu-
al meeting of the International Association of Dental Research,
Orlando, Florida, 1997.
Supported by Marmara University Research Foundation and ESPE
Company.
aAssociate Professor, Department of Prosthodontics.
bAssociate Professor, Yildiz Technical University Faculty of Chem-
istry and Metallurgy, Istanbul.
cResearch Assistant, Department of Prosthodontics.
dProfessor, Department of Prosthodontics.
Glass ionomer cements have been formulated to
combine the best properties of silicate cements, poly-
carboxylate cements, and composites. Mechanical
properties of glass ionomer cements are better than the
zinc-phosphate and polycarboxylate cements, 1 and
most studies conclude that glass ionomer cements are
more resistant to degradation than zinc-phosphate and
polycarboxylate cements.2-4 However, it has been
reported that marginal defects around crowns appeared
sooner with glass ionomer than with zinc phosphate,
possibly because of the greater susceptibility of glass
ionomer to contamination by moisture.5-7 Contact
with water at an early stage during setting changes the
setting reaction of glass ionomer cements. The powder
component of glass ionomer cements is a calcium fluo-
roaluminasilicate glass that provides a high proportion
of multivalent ions to react with the chelating liquid
component. The liquid is either an aqueous solution of
polyacrylic acid or an acrylic acid copolymer. Chemical

474 THE JOURNAL OF PROSTHETIC DENTISTRY VOLUME 80 NUMBER 4

GEMALMAZ ET AL
Table I. Glass ionomer luting cements used in the study
Product name Producer Batch no.
Ketac-Cem ESPE GmbH. D-8031, P: 6602 343
Seefeld/Oberbay, Germany L: 6602 044
Fuji Ionomer I GC Dental Inc. Corp., P: 920413C
Tokyo, Japan L: 920330A
AquaCem DeTrey/Dentsply, P: 106312/4
Surrey, England
AquaMeron VOCO, Cuxhaven, P:3239
Germany
P/L: Powder/liquid.
studies8,9 on the formation of glass ionomer cements
have described the setting mechanism as an acid-base
interaction between an aluminosilicate body and poly-
acrylic acid. Salts are generated that precipitate to form
a hard cohesive mass. The strength and insolubility of
the matrix is a result of the chains of covalent bonds in
the polymer backbone. However, there is always some
soluble material within the cement, and it has been
demonstrated that this is at a maximum in the early
phase of the cement formation.10 One of the major dis-
advantage of glass ionomer luting cements seems to be
its early moisture sensitivity.
Early moisture contamination on luting cements
during setting results not only in an increased loss of
substance but also a reduction of physical properties.
Mojon et al.11 compared the effect of early moisture
contamination on zinc-phosphate and glass ionomer
luting cements and reported that a competition
between the hardening process and the adverse effect
of water occurred in the early stages of setting that
resulted in a weaker material. They suggested that
because glass ionomer was hydrophilic, the early con-
tamination of glass ionomer resulted in the binding of
water molecules by polyacrylic acid and ion leachable
glass.11 In this way, the chemical setting was disrupted
and the decrease in hardness occurred as a result of the
absorption of water as the initial phase of degradation.
Moreover, the presence of excess water during the
growth of the hydrated silicate phase might have result-
ed also in a weaker material.
Few studies exist on the early erosion of luting
cements, except those comparing the solubility of zinc-
phosphate, polycarboxylate, and glass ionomer
cements.6,7,12 In these studies,6,7 glass ionomer
cements were more sensitive to moisture during setting
than zinc-phosphate and polycarboxylate cements;
OCTOBER 1998
THE JOURNAL OF PROSTHETIC DENTISTRY
Composition Type P/L ratio
P: calcium fluoroaluminasilicate Conventional glass 3.8/1 (g/mL)
glass ionomer
L: concentrated aqueous solution
of polymaleic acid
P: calcium fluoroaluminasilicate Conventional glass 1.8/1(g/g)
glass ionomer
L: concentrated aqueous solution
of polyacrylic acid
P: calcium aluminum silicate, Water-hardening 3.3/1(g/g)
polyacrylic acid glass ionomer
L: pure water
P: calcium aluminum fluorosilicate, Water-hardening 3.8/1(g/mL)
polyacrylic acid glass ionomer
L: pure water
however, there was little information12 on the compar-
ison of the moisture susceptibility of different glass
ionomer luting cements. Previous results on solubility
of luting glass ionomers cannot be extrapolated to the
other brands of cements because the setting speed of
cements varies markedly with the composition of the
liquid and powder. Manufacturers have modified the
chemical composition of conventional glass ionomer to
improve the handling and setting properties. Water-
hardening glass ionomers have been introduced to
improve the handling properties of the cement. In
water-hardening glass ionomers, the dried polymer is
added to the calcium fluoroaluminasilicate glass pow-
der, and the mixture is achieved by adding pure water.
In this way, shelf life of the liquid is extended by pre-
venting thickening and gelation of polyacrylic acid.
Because there is little information on the moisture sus-
ceptibility of these products, a test for comparison of
early erosion of different luting glass ionomers is of
interest.
The purpose of this study was to evaluate the solu-
bility of glass ionomer luting cements immersed in dis-
tilled water at the early stages after mixing.
MATERIAL AND METHODS
The conventional specification test for assessing the
solubility and disintegration of dental cements was
developed by Paffenbarger et al.13,14 It consists of
immersing 2 thin flat disks of cement in water 1 hour
after preparation. Twenty-four hours later, the disks are
removed from the water and the water evaporated to
dryness. The amount of material lost is determined
from the weight of the residue and is usually expressed
as a percentage of the original weight of the cement.
This method consists of the basis of solubility evalua-
tions because the organic or polymeric materials that
475
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 1. Time schedule for experimental procedure.
might be collected would not have been degraded or
volatilized during the drying steps. The conventional
method was originally intended for the quality control
of commercial products. However, the method has
often been altered by research workers to evaluate new
materials and to compare different types of dental
cements.6,7,15 Um and Øilo6 altered the conventional
method to evaluate the solubility of luting cements at
the early stages after mixing. Instead of thin flat disks,
they used resin molds as cement holders and immersed
cements in distilled water at the early stages of mixing
before the final set of the cement. The same method,
except immersion times, was used in this study to
record the early solubility of glass ionomer cements.
To quantify loss of substance from 4 different glass
ionomer luting cements (Table I), 2 circular cavities
(diameter of 5 mm, depth of 2 mm) were made in a
resin block. The cements were all mixed on a glass slab
at room temperature by the same person using a stain-
less steel spatula with a stiff blade. The powder/liquid
ratios used were in accordance with the manufacturers’
recommendations. If the powder/liquid ratios were
not given in the instructions, they were calculated from
the proportioning system given by the manufacturer by
weighing scoops of powder or drops of liquid. The
powder/liquid ratios of each cement are given in
Table I. In each case, the powder was weighed on a
digital balance that had a measuring accuracy of 0.1 mg
and the liquid was dispensed from a 1 mm tuberculin
syringe calibrated to the nearest 0.1 mL. All cements
were mixed within 30 seconds and placed in small resin
cups in 30 seconds. The surface was flattened and the
excess cement was removed with a spatula and a dry
cotton pellet. One minute after the start of mixing, the
specimens were placed in a humidifier at 37°C and
100% relative humidity. After 2, 5, or 8 minutes of stor-
age time, specimens were removed from the humidifier
and immersed in 50 mL of distilled water (pH=6) in a
glass weighing bottle (Shoot, Mainz, Germany) with
the exposed surface area pointed upward.
The mass of the empty bottle had been previously
established by using a digital balance(Chyo JL-180,
YMC Co., Ltd., Tokyo, Japan) that had a measuring
476
GEMALMAZ ET AL
accuracy of 0.1 mg. Before weighing, the bottle was
carefully rinsed, dried at 130°C for 2 hours, and cooled
in a desiccation for 20 minutes. Bottles were stored for
3 hours at 37°C and, thereafter, the cement holders
were removed from the water and the bottles were
stored at 130°C for 2 hours to allow evaporation of
water. Bottles were then cooled in a desiccator and
weighed as previously described. The amount of sub-
stance dissolved was determined by subtracting the first
established weight of the bottle from the second. The
time schedule for the total procedure is illustrated in
Figure 1.
Five specimens were used of each cement at each
immersion storage. The mean weight loss per square
centimeter was calculated. The exposed surface area of
each specimen was calculated (r = 2.5 mm) and the
exposed surface area was doubled as each cement hold-
er had 2 cavities. Data were analyzed by 2-way analysis
of variance (P<.01). Duncan’s multiple range test
analysis was also used to distinguish statistically signifi-
cant groups.
RESULTS
The recorded mean values and standard deviations
for each luting glass ionomer cement at each immersion
time is presented in Table II. The lowest cement loss
was recorded for AquaCem cement after 9 minutes
immersion time (1.68 mg/cm2). The greatest loss was
for Fuji Ionomer I after 3 minutes immersion time
(14.98 mg/cm2). For all of the observed luting
cements, AquaCem showed the lowest mean loss of
substance at all immersion times. However, the loss of
AquaCem cement was not significantly different from
Fuji Ionomer I cement at 6 minutes immersion time.
Duncan’s multiple range test also revealed that the sol-
ubility values of water-hardening cements were signifi-
cantly less (P<.01) than the 2 conventional cements at
9 minutes immersion time.
Increasing the time from the start of mixing until
immersion in water from 3 to 9 minutes resulted in a
marked decrease in loss of substance from the surface of
all 4 cements (Fig. 2). The solubility of the cements
decreased by 39% for Ketac-Cem, 61% for Fuji
VOLUME 80 NUMBER 4
GEMALMAZ ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 2. Graphic representation of solubility of glass ionomer luting cements in relation to

immersion time.

Table II. Means and standard deviations of loss of substance tration and reducing the outward transportation of
for 4 glass ionomer cements cations.9,10,16
Loss of substance (mg/cm2) In all 4 glass ionomer cements, an extension of time
Water immersion time between the start of mixing and the immersion of the
3 min 6 min 9 min
cement in water reduced the loss of substance. Signifi-
Cement Mean SD Mean SD Mean SD cant disintegration was observed during the period
Ketac-Cem 11.8 1.3 7.2 1 4.7 2 from 3 to 6 minutes after the commencement of mix-
Fuji I 15 1.5 5.8 1.5 4.9 1 ing, and the loss of substance was greatly reduced 9
AquaMeron 8.8 1.3 5.5 1.3 2.9 1 minutes after the commencement of mixing, but could
AquaCem 6.7 0.5 4.1 0.5 1.7 0.5 still be observed for all materials. These observations
Means connected by vertical lines are not significantly different. support the results of Øilo7 who concluded a signifi-
cant disintegration in glass ionomer cements during the
period of 4 to 8 minutes after mixing. Thus, an extend-
Ionomer I, 38% for AquaCem, and 37% for ed period of prevention of saliva contamination is need-
AquaMeron when the specimens were immersed in ed to control erosion of glass ionomer luting agents,
water 6 minutes after mixing. The difference in solubil- and this period was defined as the first 15 minutes in
ity between 3 and 6 minutes immersion times was most the mouth by Mojon et al.11 However, it should be
marked for Fuji Ionomer I (61%). For the 9-minute emphasized that the most sensitive period for early dis-
immersion time, the solubility decreased 61%, 67%, integration is the first 6 minutes after mixing.
75%, and 67% for Ketac-Cem, Fuji Ionomer I, Of the 4 glass ionomer cements, Fuji Ionomer I
AquaCem, and AquaMeron cements, respectively. The showed the greatest loss of substance for the 3-minute
decrease in solubility was greater between the 3- and 6- immersion time. The solubility of Fuji Ionomer I
minute immersion times than that recorded between cement reduced by 61% after the 6-minute immersion
the 6- and 9-minute immersion times (P<.01). time, whereas the reduction in solubility of Ketac-Cem,
AquaCem, and AquaMeron cements were 39%, 38%,
DISCUSSION
and 37%, respectively. However, when reduction in sol-
The data in this study demonstrated that all glass ubility at the 9-minute immersion time was considered,
ionomer cements lost weight under the test conditions. no significant variance was observed among all 4
Early dissolution of glass ionomer cements in water can cements. These results imply that Fuji Ionomer I
be attributed to 2 causes: (1) They contain sodium that cement is more sensitive to water contact in the early
forms water-soluble salts with the matrix forming period of 6 minutes after mixing than other 3 glass
anions, and (2) free calcium and aluminum ions that ionomer cements.
are present in the fresh cement can be removed by Comparison of the powder liquid ratios of the luting
chemical reactions. In addition, aluminum ions react cements used in this study revealed that Fuji Ionomer I
rather slowly with the matrix-forming anions and, cement has the lowest powder/liquid ratio (Table I).
before they are bound, are vulnerable to early water Crisp et al.17 investigated the chemistry of erosion in
leaching. As the setting progresses, the cement glass ionomer luting cements and concluded that the
becomes more and more solid, preventing water pene- main chemical species that elutes from glass ionomer

OCTOBER 1998 477

THE JOURNAL OF PROSTHETIC DENTISTRY
cements were silica, sodium, and fluoride ions. Minor
amounts of calcium and aluminum salts were eluted in
the early stages of cement formation. The observation
confirmed that calcium and aluminum ions are rapidly
bound into the cement matrix, the rate being greater
for calcium ions. Ion balance calculations indicated that
some polyacrylate anions are lost in the early stages of
cement formation. The authors17 also showed that the
elusion of ions increased as the powder/liquid ratio
decreased. Thus, it might be presumed that the
amounts of material eluted was more for Fuji Ionomer
I cement during the early stages of setting because of
its lower powder/liquid ratio.
Comparison of all 4 cements revealed a relatively
large difference between loss of substance from
AquaCem and AquaMeron cements and the other
2 conventional glass ionomer cements, Ketac-Cem and
Fuji Ionomer I. This result was in agreement with the
results of previous studies.6,7,12 Um and Øilo6 investi-
gated the early solubility of zinc-phosphate, polycar-
boxylate, and glass ionomer cements at the immersion
times of 3, 5, 7, and 10 minutes and concluded that the
early solubility of glass ionomer luting cements was
higher than zinc-phosphate and polycarboxylate
cements. In comparison of different glass ionomer lut-
ings, the authors showed that the early solubility of
AquaCem was lower than those of Fuji Ionomer I and
Ketac-Cem cements.
Setchell et al.12 evaluated the solubilities of 4 glass
ionomer cements with different compositions. The
authors concluded that the polyacrylic acid-based
materials (Chemfil and AquaCem) were less soluble
than their polymaleic acid analogous (Ketac-Cem and
Chelon). They attributed this to the lower dislocation
constant of the polyacrylic acid salt as opposed to the
polymaleic acid salt.
Øilo7 compared the early disintegration of glass
ionomer, polycarboxylate, and zinc-phosphate cements
and concluded with higher disintegration for glass
ionomer cements. It was assumed that the tartaric acid,
added to glass ionomer to facilitate the solution of cations
from the powder and to improve the setting characteris-
tics, prevents the matrix-forming reaction for a certain
period.7 The lack of a matrix would result in a material
with low cohesion during the working time. Thus, anoth-
er explanation is that early disintegration may be depen-
dent on the amount of tartaric acid and the initial reaction
products. In water-hardening glass ionomers, the dried
polymer is added to the fluoroaluminasilicate glass powder
and the mixture is achieved by adding pure water. This
suggests that the lack of tartaric acid decreases the amount
of elusion because tartaric acid was shown to extract ions
from the glass more effectively than it can from complex
polyvalent metal ions. In conclusion, all these results may
imply that water-hardening glass ionomer cements are less
sensitive to early water contamination. However, further
478
GEMALMAZ ET AL
research is needed with an extended group of luting mate-
rials to clarify this hypothesis.
CONCLUSIONS
Within the limits of this study, the following conclu-
sions were drawn.
1. Higher levels of solubility were associated with
earlier exposures of mixed cement to water.
2. Glass ionomer luting cements were highly sensi-
tive to water contact during the first 6 minutes after
mixing.
3. Among the 4 glass ionomer cements tested, early
solubility of water-hardening glass ionomer cements
were less than those of conventional ones and the rank
order of the cement solubilities in decreasing order was
as follows: Fuji Ionomer I > Ketac-Cem > AquaMeron
> AquaCem.
REFERENCES
1. Øilo G. Luting cements: a review and comparison. Int Dent J 1991;41:81-8.
2. Pluim LJ, Arends J, Havinga P, Jongebloed WL, Stokroos L. Quantitative
cement solubility experiments in vivo. J Oral Rehabil 1984;11:171-9.
3. Sidler P, Strub JR. In vivo study of the solubility and sealing capacity of 3
fixation cements. [in German] Dtsch Zahnartzl Z 1983;38:564-71.
4. Mitchem JC, Gronas DG. Clinical evaluation of cement solubility. J Pros-
thet Dent 1978;40:453-6.
5. Knibbs PJ, Walls AW. A laboratory and clinical evaluation of three dental
luting cements. J Oral Rehabil 1989;16:467-73.
6. Um CM, Øilo G. The effect of early water contact on glass-ionomer
cements. Quintessence Int 1992;23:209-14.
7. Øilo G. Early erosion of dental cements. Scand J Dent Res 1984;92:539-43.
8. Crisp S, Wilson AD. Reactions in glass ionomer cements: I. Decomposi-
tion of the powder. J Dent Res 1974;53:1408-13.
9. Crisp S, Wilson AD. Reactions in glass ionomer cements: III. The precipi-
tation reaction. J Dent Res 1974;53:1420-4.
10. Crisp S, Pringuer MA, Wardleworth D, Wilson AD. Reactions in glass
ionomer cements: an infrared spectroscopic study. J Dent Res 1974;53:
1414-9.
11. Mojon P, Kaltio R, Feduik D, Hawbolt EB, Mac Entee ML. Short-term con-
tamination of luting cements by water and saliva. Dent Mater 1996;
12:83-7.
12. Setchell DJ, Teo CK, Khun AT. The relative solubilities of four modern glass
ionomer cements. Br Dent J 1985;158:220-2.
13. Paffenbarger GC, Swenny WT, Issacs A. Zinc phosphate cements: physi-
cal properties and specification. J Dent Res 1934;21:1097-11.
14. Paffenbarger GC, Schoonova IC, Souder W. Dental silicate cements: phys-
ical properties and a specification. J Dent Res 1938;25:32-87.
15. Beech DR, Bandyopadhyay S. A new laboratory method for evaluating the
relative solubility and erosion of dental cements. J Oral Rehabil 1983;10:
57-63.
16. Crisp S, Lewis BG, Wilson AD. Characterization of glass-ionomer
cements. 6. A study of erosion and water absorption in both neutral and
acidic media. J Dent 1980;8:68-74.
17. Crisp S, Lewis BG, Wilson AD. Glass ionomer cements: chemistry of ero-
sion. J Dent Res 1976;55:1032-41.
Reprint requests to:
DR DENIZ GEMALMAZ
MARMARA UNIVERSITESI
DIS HEKIMLIGI FAKULTESI
80200 NISANTASI
ISTANBUL
TURKEY
Copyright © 1998 by The Editorial Council of The Journal of Prosthetic
Dentistry.
0022-3913/98/$5.00 + 0. 10/1/92493
VOLUME 80 NUMBER 4