Dielectric and microstructural study of the SrWO4, BaWO 4, and CaWO4 scheelite ceramics

Article abstract of DOI:10.1111/j.1551-2916.2010.04378.x

MWO₄ (M=Ca, Sr and Ba) scheelite ceramics were studied in terms of their syntheses, sintering, solubility in water, and dielectric response after storing them in dry and moist atmospheres. Of the studied scheelites, the CaWO₄ possessed the most promising dielectric properties (ε=10.9, Q × f=105 600 GHz), which were stable under the influence of humidity. BaWO₄ and SrWO₄ exhibited ε=9.0 and Q × f values of 32 200 and 62 600 GHz, respectively. The most detrimental effect of the moisture was observed for SrWO₄. A sodium impurity present in the SrCO₃ reagent (0.35 wt%), which was used for the synthesis of the SrWO₄, was found to lower the sintering temperature, enhance the grain growth, and change the other properties of the ceramics, such as humidity susceptibility and solubility in water. The evident tendency of the ceramics to attract water and the increased dissolution of tungstate were observed for all MWO₄ scheelite ceramics, which were sintered with the help of Na₂CO₃ or Li₂CO₃ (0.5 wt%) sintering aids. The results of the present study suggest that the physical and chemical properties of the ceramics should be carefully considered in the case of using of alkaline-containing sintering aids.

Full text of DOI:10.1111/j.1551-2916.2010.04378.x

J. Am. Ceram. Soc., 94 [8] 2464–2472 (2011)
DOI: 10.1111/j.1551-2916.2010.04378.x
r 2011 The American Ceramic Society
ournal
J
Dielectric and Microstructural Study of the SrWO4, BaWO4, and
CaWO4 Scheelite Ceramics
Marjeta MaWek Krmmanc,w,z Manca Logar,z Bojan BudiW,y and Danilo Suvorovz
zAdvanced Materials Department, Jozef Stefan Institute, 1000 Ljubljana, Slovenia
yAnalytical Chemistry Laboratory, National Institute of Chemistry, 1000 Ljubljana, Slovenia
MWO4 (M5 Ca, Sr and Ba) scheelite ceramics were studied in
terms of their syntheses, sintering, solubility in water, and di-
electric response after storing them in dry and moist atmo-
spheres. Of the studied scheelites, the CaWO4 possessed the
most promising dielectric properties (e 5 10.9, Q ꢀ f 5 105 600
GHz), which were stable under the influence of humidity.
BaWO4 and SrWO4 exhibited e 5 9.0 and Q ꢀ f values of
32200 and 62600 GHz, respectively. The most detrimental
effect of the moisture was observed for SrWO4. A sodium im-
purity present in the SrCO3 reagent (0.35 wt%), which was used
for the synthesis of the SrWO4, was found to lower the sintering
temperature, enhance the grain growth, and change the other
properties of the ceramics, such as humidity susceptibility and
solubility in water. The evident tendency of the ceramics to at-
tract water and the increased dissolution of tungstate were ob-
served for all MWO4 scheelite ceramics, which were sintered
with the help of Na2CO3 or Li2CO3 (0.5 wt%) sintering aids.
The results of the present study suggest that the physical and
chemical properties of the ceramics should be carefully consid-
ered in the case of using of alkaline-containing sintering aids.
and Kim3 studied the influence of LnNbO4 content on the
dielectric properties of (1ꢁx)CaWO4–xLnNbO4 ceramics (Ln 5
La, Nd, and Sm). In BaO–TiO2–WO3-based microwave dielec-
trics, BaWO4 was used to achieve a nearly zero tf.7 The doping
of (BaMg1/3Ta2/3)O3, BaZn1/3Ta2/3O3, and BaMg1/3Nb2/3O3
perovskites with BaWO4 was shown to improve the Qꢀ f values
by enhancing the B-site ordering.8 Some of these compositions
were patented.9 A systematic, comparative study of divalent
metal tungstate compounds was performed by Yoon et al.10
The authors showed that CaWO4 exhibited a higher permittivity
(e 510.4) than the Sr and Ba analogues (e58.1). The Qꢀ f val-
ues were 56 000–63 000 GHz and tf 5ꢁ53 to ꢁ78 ppm/1C. Dense
ceramics (rr495%) were obtained by sintering at 11501C.10
Although the microwave dielectric properties are usually at
the forefront of the investigations, the materials’ chemical (sta-
bility in a humid atmosphere, chemical compatibility with co-
fired materials) and mechanical (flexural strength, thermal
expansion, etc.) properties are of equal importance when the
materials are being considered for practical applications in elec-
tronics. Modern electronic devices are required to exhibit a su-
perior and reliable electronic performance. One of the
parameters that most determines the reliability of the final elec-
tronic package is the interaction of the materials with the envi-
ronment. The most common interaction of the materials with
the surrounding atmosphere is the formation of a thin layer of
adsorbed moisture. The influence of this moisture on the dielec-
tric properties was thoroughly investigated for Al2O3,11,12 where
studies revealed that the moisture alters both the permittivity
and the dielectric losses of the Al2O3. The effect of the moisture
on these characteristics strongly depends on the relative humid-
ity (RH), the porosity, the contaminations present, the surface
treatment, etc.11,12 Like with Al2O3, other oxide materials are
also expected to interact with the humid atmosphere, at least to
some degree. Therefore, the influence of humidity on their elec-
trical performance must be taken into consideration, parti-
cularly when the materials are meant for applications in
electronics. This is also true for materials that are not specifi-
cally hygroscopic. Materials that show a significant tendency to
attract or react with moisture are not appropriate for applica-
tions in electronics, because of the deterioration of the materials’
properties in the presence of a moist atmosphere. In the indus-
trial production of electronic components, a solid-state synthesis
with the homogenizing and milling occurring in water media is
still the prevailing route for the preparation of ceramic powders.
However, ceramics with significant solubility or an incongruent
solubility in water are not appropriate for this kind of large-scale
production due to problems connected with the loss of material,
the change of stoichiometry, pollution, etc. A knowledge of the
materials’ solubility is, therefore, indispensable for the industrial
production of ceramic powders.
I. Introduction
HE past 20 years have seen the rapid development of various
wireless communication devices, such as mobile phones,
T
wireless local area networks, bluetooth, intelligent transporta-
tion system, etc. The most important materials for these appli-
cations are microwave dielectrics with low dielectric losses (high
Q ꢀ f values), the appropriate permittivity (depending on the
application), and a temperature-stable resonant frequency. It is
for this reason that dielectric materials have been so intensively
investigated. The database on microwave dielectric resonator
materials, which is edited by M. T. Sebastian, already contains
information about the microwave dielectric properties of more
than 2200 different oxide materials.1
Owing to their low permittivities (r12) and high Q ꢀ f values
(450 000 GHz), the scheelite ceramics, with the general formula
MWO4 (M5 Ca, Sr, and Ba), are regarded as promising mate-
rials for microwave substrate applications. Kim et al. 2 suggested
CaWO4-based ceramics as a new low-temperature cofired ce-
ramic (LTCC) material. The authors decreased the sintering
temperature of the CaWO4 from 11001 to 8501C with the addi-
tion of 0.5 wt% Bi2O3 and 5 wt% H3BO3 sintering aids. Based
on an e 5 8.7, a Q ꢀ f 5 70220 GHz, and a tf 5 ꢁ15 ppm/1C,
this material already exceeds those values of commercially avail-
able LTCC substrates. There were several studies relating to the
microwave dielectric characterization of ceramics where CaWO4
or BaWO4 was combined with other ceramic materials.3–6 Kim
P. Davies—contributing editor
The fabrication of advanced electronic components using
LTCC technology is performed by the cofiring of ceramics lay-
ers with the electrodes (Ag, Cu, Au). In order to sinter the ce-
ramics below the melting point of the electrodes, various
additives in small quantities (o1 wt%) are added to reduce
Manuscript No. 28461. Received August 11, 2010; approved December 6, 2010.
wAuthor to whom correspondence should be addressed. e-mail: marjeta.macek@ijs.si
2464

August 2011
Dielectric and Microstructural Study of Scheelite Ceramics
2465
the sintering temperature. Owing to the low concentration, the
sintering aids in many cases entirely incorporate into the crystal
lattice of the matrix, and consequently their addition does not
significantly influence the properties of the ceramics. A good
example is the low-temperature sintering of BaTiO3 with Li2O,
where lithium incorporates into the titanium sites of the BaTiO3.
The lithium-containing compounds were found to be very effec-
tive sintering aids for many ceramics.13 However, the results of
the present study raise some concerns about using alkaline-con-
taining sintering aids for any ceramics system.
to 0.22 and 0.09 wt% of Na and Li, respectively, were mixed
with the milled scheelite powders before the sintering.
The sintering of the scheelite pellets was observed by means of
a heating microscope (EM 201, Hesse Instruments, Osterode,
Germany), which monitored the change of the pellet’s silhou-
ettes during heating at 101C/min.
The reaction mechanisms were followed using powder X-ray
diffraction (XRD, D4 Endeavor, Bruker AXS, Karlsruhe, Ger-
many) and PAnalytical X’Pert PRO MPD (Almelo, the Neth-
erlands) and a thermal analysis system (STA Jupiter 449,
In this paper, we try to go beyond a simple microwave dielec-
tric characterization of the MWO4 (M 5 Ca, Sr, and Ba) sche-
elites, which has already been reported on by other authors.10
We would like to shed some light on other, for practical ap-
plications, important aspects, such as the stability of the sche-
elites under the influence of a humid atmosphere and water.
The influence of alkaline-containing (Na, Li) impurities on the
sintering of scheelites and their susceptibility to humidity are
presented and discussed.
Netzsch, Selb, Germany) coupled with a 403C Aeoloss mass
¨
spectrometer (Netzsch).
The microstructural studies of the samples were conducted
with a scanning electron microscope (SEM, JSM-5800, JEOL,
Tokyo, Japan), equipped with an Oxford Link-Isis energy-
dispersive X-ray analysis system, and a transmission electron
microscope (TEM, JEM 2100, JEOL), equipped with a Gatan
ORIUS SC1000 CCD camera. SEM investigations were per-
formed for polished samples that were thermally etched at 301–
501C below the sintering temperature.
The densities of the sintered specimens were measured using
the Archimedes’ method with ethanol as the submersion liquid.
The literature data of 6.354 g/cm3 (PDF 85–0587), 6.380 g/cm3
(PDF 77–2236), and 6.383 g/cm3 (PDF 85–0588) were used for
the theoretical densities of the SrWO4, CaWO4, and BaWO4,
respectively.14 The relative densities (rr) were obtained by com-
paring the measured densities with the theoretical ones.
The Brunauer–Emmett–Teller surface areas (SBET) of the re-
agents and milled powders were analyzed by nitrogen adsorp-
tion with a Micromeritics Gemini II 2370 nitrogen-adsorption
apparatus (Norcross, GA).
The microwave characterization was performed with the help
of a network analyzer (HP 8720 C, Palo Alto, CA) using the
dielectric-resonator method developed by Krupka et al.15 The
samples, which were stored previously in a desiccator, were
measured at ambient conditions. When the influence of humid-
ity on the microwave dielectric properties was tested, the reso-
nators were measured 5 min after they were removed from the
compartment with a 100% RH, where they were stored for a
predefined time. In the case of the SW-R resonators exposed to
humidity for 14 h, this time was prolonged until the pellet was
visually dry. In this case, the dielectric characterization was also
made at ambient conditions.
The solubility tests for the scheelite powders were performed
with constant stirring of the powder (1.5 g) in 35 mL of water
(Milli-Q water: resistivity 18 MO ꢂ cm at 251C) for 3 days at
ambient temperature. Afterwards, the powder was separated
from the water by centrifugation and filtration. The concentra-
tions of the dissolved ions in the water were determined with a
ICP-AES (Varian, Model 715 ES) using a matrix-matched cal-
ibration and internal standardization. The dissolution of the
ions from the sintered pellets was determined in the similar way.
The scheelite pellets of 1.7 g were suspended for 3 days in the
water (30 mL), which was constantly stirred during this time.
After the solubility test, the pellet was removed from the water
solution, which was then analyzed by ICP-AES. In the event
that the pellet decomposed, the powder was separated from the
water solution by centrifugation and filtration.
II. Experimental Procedure
Scheelite CaWO4, SrWO4, and BaWO4 ceramics were prepared
from WO3 and the corresponding carbonate (MCO3, M 5 Ca,
Sr, and Ba) using the solid-state reaction technique (see the de-
notation in Table I). Compared with the published sintering be-
havior of SrWO4, the SrWO4 from this study sintered at a 5001C
lower temperature. Such different sintering behavior was ob-
served for SrWO4 prepared from SrCO3 with a purity of 99.0%
and WO3 (99.9%). In order to exclude the influence of impuri-
ties, SrWO4 was also prepared with high-purity SrCO3. Detailed
information about the reagents and the concentrations of the
impurities in SrCO3 (99.0%), which were determined by an
inductively coupled plasma atomic emission spectrometer
(ICP-AES, Varian, Model 715 ES, Mulgrave, Australia), are
presented in Table I. CaWO4 and BaWO4 were also prepared
from high-purity reagents (Table I). The synthesis of scheelites
started with the drying of the reagents (1501C), weighing and
homogenizing in a mortar. The calcinations with intermediate
grinding were performed at 7001–11001C and 7001–10001C for
the BaWO4 (CaWO4) and SrWO4, respectively. The tempera-
ture difference between each calcination step was 1001C. Before
sintering, the powders were milled in ethanol with Y-stabilized
ZrO2 balls (d 5 3 mm) for 2–4 h to a median particle size of
0.7 mm. The milled powders were then uniaxially pressed under a
pressure of 100 MPa to obtain green compacts with a diameter
of 12 mm and a thickness of around 5 mm. The green pellets
were sintered at 6251–11001C for 5 h. After the sintering, the
samples were stored in a desiccator. For studying the influence
of alkaline impurities on the sintering and other properties of the
scheelites, 0.5 wt% of Na2CO3 and Li2CO3, which corresponds
Table I. The Reagents Used for the Synthesis of MWO4
(M 5 Ca, Sr, and Ba) Scheelite Ceramics
Median
BET
(m2/g)
particle
Denotation
for MWO4
Reagents (producer, purity)
size (mm)
SrCO (Riedel de Haen,
¨
3
1.6588
10.57
SW-R,
III. Results and Discussion
99.0%) Impurities: Ba (0.66
wt%), Na (0.34 wt%), Mg
(0.0075 wt%), Ca (0.016
wt%)
SrCO3 (Alfa Aesar 99.99%)
SrCO3 (Aldrich 99.9%)
BaCO3 (Aldrich, 99.98%)
CaCO3 (Alfa Aesar,
99.95%)
(1) Synthesis, Sintering, Microstructural, and Dielectric
Studies
According to the X-ray analysis, all the SrCO3 reagents used for
the synthesis of the SrWO4 were single-phase strontianite (space
group Pmcn), while the WO3 had a monoclinic structure (space
group P21/n). The main known differences between the used
SrCO3 reagents were in the particle size distribution, the specific
surface area, the declared purity, and the origin. The high-purity
SrCO3 (Alfa Aesar, Karlsruhe, Germany and Aldrich, Steinheim,
0.9715
3.6481
0.9161
0.5176
4.46
1.29
34.71
4.67
SW-Alf
SW-Ald
BW
CW
WO3 (Fluka, 99.9%)
2.0352
5.3
—
Germany) had a metal basis, while the SrCO (Riedel de Haen,
3
¨

2466
Journal of the American Ceramic Society—Macˇek Krzˇmanc et al.
Vol. 94, No. 8
Fig. 1. TG (a), DSC (b), and CO2 evolution profile mass spectrum (c)
curves obtained upon heating (at 101C/min) of equimolar-mixed pow-
ders of SrCO3 and WO3 (SW-Alf (––), SW-Ald (—), SW-R (—)), and
TG (a) of pure SrCO3 (––) and equimolar mixed powders of SrCO3
(99.99%) and WO3 with the addition of 0.5 wt% of Na2CO3 (- - -).
Fig. 2. TG (a), DSC (b), and CO2 evolution profile mass spectrum (c)
curves obtained upon heating (at 101C/min) of pure CaCO3 and BaCO3
(only TG) and equimolar-mixed powders of CaCO3 (BaCO3) and WO3.
Hannover, Germany) did not (Table I). Ba (0.66 wt%) and Na
(0.34 wt%) were found to be the main impurities in the least-pure
The BaCO3 reagent was found to be thermally stable to a
much higher temperature than the BaCO3 in an equimolar mix-
ture with WO3 (Fig. 2(a)). The same behavior was observed by
Al-Hajji et al., 18 who proved the concurrence of the CO2 release
and the formation of BaWO4. The decomposition of the pure
CaCO3 and that in the equimolar mixture with WO3 occurred at
approximately the same temperature (Fig. 2(a)). The absence of
the intermediate CaO phase in the XRD pattern of the partially
reacted CaCO3:WO3 mixture indicates that the formation of
CaWO4 also occurs through a direct reaction between the re-
agents. The reaction between CaCO3 and WO3 occurred at con-
siderably lower temperatures than the reaction between BaCO3
and WO3 (Fig. 2(a), curves CW and BW). Both reactions were
accompanied by an endothermic heat change (Fig. 2(b)).
The phase compositions of the reaction products for the
different SrCO3 and WO3 mixtures after the calcinations were
in accordance with their different reaction kinetics observed by
the thermal analysis. The XRD pattern of the SW-R composi-
tion revealed single-phase SrWO4 after the calcinations at 7001C
(Fig. 3), while the SW-Ald and SW-Alf samples required at least
three calcinations at 7001–9001C to obtain an XRD pattern free
of the WO3 and SrCO3 diffraction lines. According to the XRD,
single-phase BaWO4 and CaWO4 were obtained after the pre-
reactions at 9001 and 8001C, respectively (Fig. 3). The presence
of semiconducting WO3 in the concentrations below the detec-
tion limit of the XRD was found to cause a significant deteri-
oration of the Q ꢀ f values. For the complete elimination of the
reactants, the prereactions were performed up to 10001 and
11001C for the SrWO4 and CaWO4 (BaWO4), respectively.
The sintering temperature required for the preparation of
dense ceramics (rr497%) varied in accordance with the differ-
ent sintering behaviors of the studied scheelites (Figs. 4, 5(a),
and 6(a)). The SW-R sample, which started to shrink at
B5001C, attained a high relative density of 98.4% at just
SrCO (Riedel de Haen) reagent. The sodium content of 0.34 wt%
3
¨
in SrCO3 corresponds to 0.15 wt% of Na in the SrWO4 ceramics.
The TG/DSC analysis in combination with the mass spectrometry
revealed rather different reaction kinetics for stoichiometric mix-
tures of the various SrCO3 and WO3 reagents. The reaction be-
tween the SrCO (Riedel de Haen) and WO started below 6001C
¨
3
3
and reached its maximum rate at 7801C. The accelerated decom-
position process at 7801C resulted in a rapid loss of mass
(Fig. 1(a), curve SW-R), a narrow endothermic DSC peak
(Fig. 1(b), curve SW-R), and a sharp peak in the mass spectrum
(MS) of CO2 (Fig. 1(c), curve SW-R). In the case of the other two
SrCO3 (Alfa, Aldrich) and WO3 mixtures, the onset and comple-
tion of the reaction were shifted to higher temperatures (Fig. 1(a)).
The decomposition of the SrCO3 in the mixture with WO3 in all
cases occurred at a lower temperature than the decomposition of
pure SrCO3 (Fig. 1(a)). This could be an indication of the direct
reaction between the SrCO3 and the WO3 without the formation
of an intermediate SrO phase. The XRD pattern of mixtures after
the isothermal calcinations at 7001C consisted of diffraction lines
of the SrWO4 and the reactants. This additionally supports the
single-step mechanism for SrWO4 formation. The endothermic
peaks in the DSC curves (Fig. 1(b)) agree with the reported pos-
itive enthalpy change for the SrWO4 formation from the SrCO3
and WO3.16,17 The accelerated reaction between SrCO3 (Riedel de
Haen) and WO was assumed to be the result of the Na impurity
3
¨
rather than the Ba or other properties of the reagent. In order to
verify this hypothesis, a TG/DSC-MS analysis of the WO3 and
SrCO3 (99.99%, Alfa Aesar) mixture with the addition of 0.5
wt% of Na2CO3 was performed. The similar decomposition path-
way of this mixture compared with that of SrCO3 (Riedel de
Haen) and WO (Fig. 1(a)) indirectly proves that the Na impurity
¨
promotes the reaction.
3

August 2011
Dielectric and Microstructural Study of Scheelite Ceramics
2467
Fig. 3. X-ray diffraction pattern of CaWO4 (CW), SrWO4 (SW-R), and
BaWO4 (BW) scheelites after the prereactions at 8001, 7001, and 9001C,
respectively.
6251C (Figs. 4 and 5(a)). With a further increase in the sintering
temperature, the density decreased. In contrast to the SW-R, the
SW-Alf and SW-Ald sintered at a much higher temperature of
9001–10001C. Also for these samples, the density started to de-
crease when the temperature increased to higher values
(Fig. 5(a)). It is commonly observed in ceramics that the den-
sity, after reaching the maximum, starts to decrease with a fur-
ther increase in the sintering temperature. This phenomenon is
ascribed to the increase of the pore size and a decrease of the
pore pressure.19 Because of faster grain growth, the SW-R ce-
ramics attained their maximum density at a lower temperature
than the other two SrWO4 ceramics. In order to verify whether
Fig. 5. Density (a), permittivity (b), and Q ꢀ f (c) of sintered SrWO4
ceramics as a function of sintering temperature; SW-R (–& –), SW-Alf
(–J–), SW-Ald (–n–). The Q ꢀ f values were determined at B11 GHz.
the deviated sintering behavior of the SW-R ceramics resulted
from the Na impurity, the sintering studies were performed
for SW-Alf ceramics with the addition of 0.5 wt% Na2CO3
(0.21 wt% of Na). The heating microscope measurements
revealed a significant lowering of the sintering temperature.
The addition of Na2CO3 also had a similar effect on the sinte-
ring of BaWO4 and CaWO4 (Fig. 4(b)). Compared with Li2CO3,
the sintering with Na2CO3 has not been well studied. We found
that Li2CO3 (0.5 wt%) also effectively lowers the sintering tem-
perature of scheelites. The influence of Li2CO3 on the sintering
of CaWO4 is presented in Fig. 4(b). It is not surprising that ad-
ditives in a concentration of r1% enhance the sintering. There
are numerous reports, especially in the field of LTCC technol-
ogy, about lowering the sintering temperature with the help of a
small quantity of additives.13,20,21 In particular, lithium-contain-
ing compounds (Li2O, Li2CO3) were found to be extremely
effective in lowering the sintering temperature at very low con-
centrations (0.3 wt% Li2O).13
The microstructural development in SW-R ceramics also
differed from that in the SW-Alf and SW-Ald ceramics
(Fig. 7). SW-R ceramics sintered at 6301C consisted of a ma-
trix of predominantly micrometer-sized grains with some sepa-
rate, exaggeratedly grown grains, which implied that the
sintering was promoted by the presence of a liquid phase
(Fig. 7(a)). The sintering at 9001C led to coarse-grained SW-R
ceramics with an obviously deteriorated mechanical strength
(Fig. 7(b)). In contrast, the SW-Alf and SW-Ald ceramics sin-
tered at this temperature contained more uniform and smaller
grains. At 10001C, the microstructures of the SW-Alf and SW-
Ald ceramics became very similar, with a grain size of 1–10 mm
(Figs. 7(d) and (f)).
The CaWO4 ceramics sintered to a high density of 98% at
10001C, while the BaWO4 showed a high and rather constant
density (98%) in a broad sintering range from 8501 to 11001C
Fig. 4. Sintering profiles of the milled MWO4 powders (a) and the
MWO4 powders with the addition of 0.5 wt% of Na2CO3 (Li2CO3) (b).

2468
Journal of the American Ceramic Society—Macˇek Krzˇmanc et al.
Vol. 94, No. 8
Na2CO3) was a small amount of Na detected within the grains,
which indicates the incorporation of Na into the BaWO4 matrix.
The strong tendency of SW-R and MWO4 (10.5 wt% Na2CO3)
to attract water could be the consequence of the unincorporated
Na2O due to its hygroscopity. The presence of eventual carbon-
ate remains in the sintered ceramics was checked by the TG/
DSC-MS analysis of the crushed powders of sintered ceramics.
Taking into account the fact that no CO2 evolution was ob-
served up to 12001C, we believe that there were no undecom-
posed carbonates present in the sintered scheelite ceramics.
The permittivity of the scheelites changed with the sintering
temperature in a similar manner to the density, i.e., a drop in the
density is reflected in a decrease in the permittivity (Figs. 5(b)
and 6(b)). The permittivity of dense CaWO4 was slightly higher
(e 5 10.9) than that determined by Yoon et al.10 (e 5 10.4). Sim-
ilarly, the permittivities of the BaWO4 (e 5 9.0) and SrWO4
(e 5 9.0) ceramics were also higher in comparison with the re-
ported values of 8.1.10 In contrast to the SrWO4 and CaWO4,
the BaWO4 showed a constant permittivity over a broad tem-
perature range (8501–11001C) (Fig. 6(b)). Regarding the permit-
tivity of the densely sintered ceramics, there was no significant
difference between the three SrWO4 ceramics (Fig. 5(b)). In ac-
cordance with the different sintering behavior of the SW-R, this
ceramic attained the highest density and permittivity at a con-
siderably lower temperature than the SW-Ald and SW-Alf
ceramics. While the density and permittivity of the SW-R
ceramics decreased already at T ꢃ 7001C, the maximum Q ꢀ f
value was obtained at 9001C (Fig. 5(c)). The Q ꢀ f values of the
CaWO4 and SrWO4 from this study were greater than those re-
ported by Yoon and colleagues. However, for BaWO4, we mea-
sured nearly two times lower Q ꢀ f values compared with the
literature value of 57500 GHz.10 Even when the BaWO4 was
prepared from other initial reagents, its Q ꢀ f values never ex-
ceeded 40 000 GHz.
Fig. 6. Density (a), permittivity (b), and Q ꢀ f (c) of sintered CaWO4
and BaWO4 ceramics as a function of sintering temperature. The Q ꢀ f
values of CaWO4 and BaWO4 were determined at 10.5 and 11.2 GHz,
respectively.
(2) The Influence of Water on Scheelites
Our preliminary studies of SW-R revealed that the Q ꢀ f values
decreased from 70 000 to 20 000 GHz when the samples were
stored under ambient conditions for a long time. We assumed
that this deterioration might be caused by the air humidity. In
order to verify the hypothesis, the SrWO4 powders and densely
sintered pellets were put in a compartment with 100% RH. Vi-
sual observations showed very different behaviors for the SW-R
in comparison with other sintered scheelite ceramics. The SW-R
pellet was completely surrounded by water after several hours,
whereas there were no visual signs of water on the other scheelite
pellets, even after several weeks. CaWO4, BaWO4, and SrWO4
(high-purity reagents) sintered with the addition of a Na2CO3
and Li2CO3 sintering aid showed the same behavior with regard
to the attraction of water-like sintered SW-R ceramics; the pel-
lets were covered with water when stored in a high-humidity
atmosphere for a few hours. The SW-R powder did not collect
the water as visibly as the SW-R pellets. We believe that the
reason for such different behaviors between the powder and the
dense pellet lies in the fact that the powders exhibit a consider-
ably higher surface area than the sintered ceramics. The TG
measurement of the SW-R powder and pellet stored previously
in 100 RH revealed that the water is very weakly bonded to the
SrWO4. Namely, the majority of the water already disappeared
during the stabilization of the balance and purging with dry
gasses, i.e., before the start of the TG experiment. According to
the mass loss, the remaining water (0.5 wt%) was removed up to
1201C.
(Fig. 6(a)). The scheelite ceramics in this study sintered to a high
density at lower temperatures, compared with the literature re-
port of 11501C.10 Accordingly, the grain growth of the scheelites
from this study differed from that reported in the literature. In
10
general, the scheelite ceramics prepared by Yoon et al.
at-
tained a certain grain size at a considerably higher temperature
than our scheelites. One of the reasons for this could be the 3-h
longer sintering time in our study. From among the studied
scheelites, the CaWO4 ceramics consisted of the smallest grains
(Fig. 8). This is the opposite of the observation of Yoon et al., 10
where CaWO4 was the ceramic with the largest grains. Among
the important factors affecting the grain growth are impurities.
The impurities that cause the formation of a liquid phase pro-
mote the grain growth. In contrast, the impurity particles seg-
regated at the grain boundaries often retard the grain growth.
Therefore, it is to be expected that the ceramics prepared from
the reagents with different purities lead to ceramics with a differ-
ent grain size. Although there are no data about the purity of the
reagents in the study of Yoon et al, 10 we believe that this is one
of the reasons for the difference in the grain growth behavior
observed in their study and ours. We have to note that the sche-
elites from this study obviously lose their mechanical strength
with an increase of the sintering temperature. For example, the
fragility of the BaWO4 ceramics sintered at 11001C became so
high that the sample broke into pieces during the polishing. For
this reason, the microstructure in Fig. 8(d) is represented by
grains from the surface of the pellet. In accordance with the
Hall–Petch relation, the CaWO4 ceramics are expected to ex-
hibit the highest mechanical strengths.
Taking into account the strong tendency of the SW-R ceramic
to attract water, it was expected that the humidity would detri-
mentally influence its microwave dielectric properties. The SrWO4,
BaWO4, and CaWO4 ceramics prepared from high-purity reagents
were not so evidently hygroscopous. However, this was not a
proof that humidity had no effect on their dielectric properties.
The dielectric measurements of the resonators that were stored
previously in 100 RH were performed in order to find out how
The EDS analysis of the grains in the SW-R ceramics did not
show the presence of Na within the grains. The same was also
observed for the SrWO4 and CaWO4 ceramics sintered with the
addition of Na2CO3. Only in case of BaWO4 (10.5 wt%

August 2011
Dielectric and Microstructural Study of Scheelite Ceramics
2469
Fig. 7. Scanning electron microscope micrographs of thermally etched SrWO4 ceramics.
the humidity influences the dielectric properties of MWO4
ceramics. Regardless of the purity, the SrWO4 ceramics were
found to be the most moisture-sensitive material. The influence
of humidity was noticeable for the Q ꢀ f values, but not for
the permittivity. In spite of the visually different susceptibility
for water, the SW-R, SW-Ald, and SW-Alf samples displayed a
very similar deterioration of the Q ꢀ f values under the influence
of humidity. The Q ꢀ f values dropped below 10 000 GHz after
17 h of exposure to 100% RH for all three SrWO4 samples
(Fig. 9). Nevertheless, in all cases, the Q ꢀ f values could be re-
stored by drying. In contrast to the SrWO4, the dielectric prop-
erties of the CaWO4 and BaWO4 were not significantly affected
by the moisture. In particular, CaWO4 exhibited high and very
stable Q ꢀ f values above 100000 GHz (Fig. 9).
disintegrated in water more than powder. In the sintered SW-R
ceramics, the width of a newly formed amorphous phase was
B50 nm, while only a few nanometers thick amorphous layer
was formed around the SW-R particle (Fig. 11). Detailed solu-
bility studies were performed in order to obtain a better insight
into the differences between the studied scheelites in terms of
their reactivity with water. In the case of SrWO4, the investiga-
tions revealed a higher solubility of the tungstate compared with
that of strontium. A comparison between the solubility of the
powders and pellets (Tables II and III) revealed the deviation of
SrWO4 from the other two scheelites. The concentrations of the
dissolved tungstate from the sintered ceramics (Table III) are, in
all the SrWO4 ceramics, higher than those dissolved from pow-
ders (Table II). These results are consistent with the higher de-
gradation effect of the water on the sintered SrWO4 ceramics,
which was also observed with the TEM. The solubility experi-
ments for the different SW-R samples gave very similar results
for the concentration of the dissolved tungstate, but varied for
the strontium (Tables II and III). The reason for the large scatter
in the strontium concentration is not known at this time. How-
ever, we believe that it might be connected with the different
decomposition level of the newly formed strontium-rich amor-
The remarkable susceptibility of SW-R to water was the mo-
tivation for the study of the reaction between SrWO4 ceramics
and water. SEM investigations showed that water collected on
the sintered SW-R pellet in 100% RH caused a degradation of
the surface grains and the formation of a smooth but cracked
layer (Fig. 10). The amorphous nature of the reaction product
between the SrWO4 and water was confirmed with the TEM.
According to the TEM observations, sintered SW-R ceramics

2470
Journal of the American Ceramic Society—Macˇek Krzˇmanc et al.
Vol. 94, No. 8
Ã
Ã
Fig. 8. Scanning electron microscope micrographs of thermally etched CaWO4 ceramics (a and b) and BaWO4 (c and d ). Surface of the pellet
(without polishing and thermal etching).
phous phase. Like with the SrWO4, the solubility of the CaWO4
was also found to be incongruent with the higher solubility of
tungstate compared with that of calcium. The same phenome-
non was also observed by other researchers.22 The incongruent
solubility of scheelites in water is most probably the conse-
quence of several competitive equilibrium reactions, which
were studied in detail for CaWO4 by Marinakis and Kelsall.22
In contrast to the SrWO4, the CaWO4 showed a higher disso-
lution of tungstate from powders than from sintered ceramics.
This is in accordance with the larger specific surface area of
the powders compared with that of the sintered pellet. Both
the BaWO4 powder and ceramics exhibited a surprisingly larger
dissolution of barium than tungstate. The dissolution of BaWO4
in water is expected to occur via similar mechanisms as CaWO4,
but with different equilibria constants. One of the reactions
involved in the dissolution of MWO4 is also the equilibrium
M(OH)2(s)2M2112OHꢁ.22 The solubility of Ba(OH)2 is
higher than that of Sr(OH)2 and Ca(OH)2. Perhaps, this could
be one of the reasons for the higher concentrations of barium
than tungstate in the BaWO4-saturated solutions. The pH was
monitored for the dissolution of MWO4 from the pellets. A pH
in the range from 6 to 7 was determined for all the samples,
except for the SW-R sample, where the pH increased to 8.7
Table III). No significant change of pH was observed in the
timeframe from a few hours to 3 days after the start of the dis-
solution experiment. The solubility studies of the MWO4 ceram-
ics, which were sintered with the addition of Na2CO3, showed
the significant increase of the tungstate dissolution from CaWO4
and SrWO4 but less from BaWO4 (Table IV). The dissolution of
SrWO4 (SW-Alf) in alkaline water solution with pH 10–11 did
not result in an increase of the tungstate concentration. Based on
this, we believe that the increase in the pH is not the main reason
for the high dissolution of the tungstate.
Fig. 9. Q ꢀ f values of MWO4 (M5 Ca, Sr, Ba) scheelite ceramics after
Fig. 10. Scanning electron microscope micrograph of the reaction
product between water and sintered SW-R ceramics.
storing in 100% relative humidity.

August 2011
Dielectric and Microstructural Study of Scheelite Ceramics
2471
Table IV. The Water Concentrations of Cations (Sr, Ba, Ca)
and Tungstate Dissolved from the MWO4 (M 5 Ca, Sr, and
Ba) Scheelite Pellets (1.7 g) Sintered with the Help of 0.5 wt%
Na2CO3 after Exposure to Water (30 mL) for 3 Days
CM (M5 Ca, Sr, Ba)
Sintered pellet (9001C/5 h)
(mg/L)
CW (mg/L)
pH
CW10.5 wt% Na2CO3
SW10.5 wt% Na2CO3
BW10.5 wt% Na2CO3
0.113 (Ca)
1.022 (Sr)
1.754 (Ba)
414.493
248.17
29.36
6.5
11.0
10.4
in a small concentration (0.15 wt%) could enhance the SrWO4
formation kinetics, the sintering, the grain growth, and the sol-
ubility in water. SrWO4 ceramics prepared from high-purity re-
agents were not visibly hygroscopic and exhibited a lower
solubility in water. However, a deterioration of the Q ꢀ f val-
ues in the moist atmosphere was observed for all the investigated
SrWO4 ceramics, irrespective of the reagents’ purity. Under the
influence of humidity, the Q ꢀ f values of the SrWO4 decreased
from 62600 GHz to values o10 000 GHz, while e remained
constant at 9.0. In contrast, the CaWO4 retained its high Q ꢀ f
values of 105 600 GHz and e of 10.9 after storage in a moist
atmosphere.
The sintering temperature of MWO4 scheelites could be effec-
tively lowered by the addition of Na2CO3 or Li2CO3 (0.5 wt%).
However, even low concentrations of alkalines (Li: 0.09 wt%,
Na: 0.15 wt%) made the MWO4 ceramics evidently susceptible
to humidity. The presence of sodium in the scheelite ceramics
was also recognized as the reason for their degradation in water
with a high dissolution of tungstate. Alkaline-containing (Na,
Li) sintering aids were found to be inappropriate for lowering
the sintering temperature of the scheelites.
Fig. 11. Transmission electron microscope observations of the reaction
(a) between water and sintered SW-R ceramics and (b) between water
and SW-R powder.
References
1M. T. Sebastian, Dielectric Materials for Wireless Communications, 1st Edition.
Elsevier, Amsterdam, 2008, pp. 541–617.
Table II. The Water Concentrations of Cations (Sr, Ba, Ca)
and Tungstate after 3-Day Mixing of the MWO4 (M 5 Ca, Sr,
and Ba) Scheelite Powder (1.5 g) in Water (35 mL)
2E. S. Kim, S. H. Kim, and B. I. Lee, ‘‘Low-Temperature Sintering and Mi-
crowave Dielectric Properties of CaWO4 Ceramics for LTCC Applications,’’
J. Eur. Ceram. Soc., 26, 2101–4 (2006).
3E. S. Kim and S. H. Kim, ‘‘Influence of Bond Valence on Microwave Dielectric
Properties of (1ꢁx)CaWO4–xLnNbO4 (Ln 5 Nd, Sm) Solid Solutions,’’ J. Elect-
roceram., 21, 2–7 (2008).
Powder
SBET (m2/g)
CM (M5 Ca, Sr, Ba) (mg/L)
CW (mg/L)
SW-R
SW-Alf
SW-Ald
BW
6.6
5.61
8.3
7.3
8.1
0.87 (Sr)
2.13 (Sr)
1.57 (Sr)
25.07 (Ba)
5.22 (Ca)
279.7
8.19
7.39
1.73
58.1
4E. S. Kim and S. H. Kim, ‘‘Effects of Structural Characteristics on Microwave
Dielectric Properties of (1ꢁx)CaWO4–xLaNbO4 Ceramics,’’ J. Electroceram., 17,
471–7 (2006).
5S. H. Yoon, G.-K. Choi, D.-W. Kim, S.-Y. Cho, and K. S. Hong, ‘‘Mixture
Behavior and Microwave Dielectric Properties of (1ꢁx)CaWO4–xTiO2,’’ J. Eur.
Ceram. Soc., 27, 3087–91 (2007).
CW
6I.-H. Park, B.-S. Kim, K.-Y. Kim, and B.-H. Kim, ‘‘Microwave Dielectric
Properties and Mixture Behavior of CaWO4–Mg2SiO4 Ceramics,’’ Jpn. J. Appl.
Phys., 40, 4956–60 (2001).
Table III. The Water Concentrations of Cations (Sr, Ba, Ca)
and Tungstate Dissolved from the Sintered MWO4 (M5 Ca,
Sr, and Ba) Scheelite Pellets (1.7 g) that were in Water (30 mL)
for 3 Days
7S. Nishigaki, S. Yano, H. Kato, T. Hirai, and T. Nonomura, ‘‘BaO–TiO2–WO3
Microwave Ceramics and Crystalline BaWO4,’’ J. Am. Ceram. Soc., 71 [1] C-11–7
(1988).
8Z.-Q. Tian, H.-X. Liu, H.-T. Yu, and S.-X. Ouyang, ‘‘Effect of BaWO4 on
Microstructure Microwave Dielectric Properties of Ba(Mg1/3Nb2/3)O3,’’ Mater.
Chem. Phys., 86, 228–32 (2004).
Relative
CM (M5 Ca, Sr, Ba)
CW
9H. J. Jung, K. H. Yoon, D. P. Kim, and E. S. Kim, ‘‘Compositions of High
Frequency Dielectric Ceramics’’; U. S. Patent Issued on August 22, 1995.
10S. H. Yoon, D.-W. Kim, S.-Y. Cho, and K. S. Hong, ‘‘Investigation of the
Relations Between Structure and Microwave Dielectric Properties of Divalent
Metal Tungstate Compounds,’’ J. Eur. Ceram. Soc., 26, 2051–4 (2006).
11I. Chatterjee, M. Misra, G. W. Warren, and C. Neve, ‘‘Effect of Moisture and
Contaminants on the Dielectric Constant of Thin Film Alumina Substrates’’;
pp. 335–441 in Materials Reliability Issues in Microelectronics. Materials Research
Society Symposium Proceedings, Pittsburgh, PA, Vol. 5, edited by J. R. Lloyd,
P. S. Ho, C. T. Sah, and F. Yost, Materials Research Society, Warrendale,
PA, 1991.
Sintered pellet Ts (1C) density (%)
(mg/L)
(mg/L)
pH
Ã
SW-R
SW-R
SW-Alf
SW-Ald
BW
625
900
900
1000
900
1000
98.4
90.1
98.3
97.4
97.8
98.0
5.33 (Sr)
0.34 (Sr)
0.21 (Sr)
0.39 (Sr)
9.59 (Ba)
0.17 (Ca)
364
345
16.3 7.0
11.7
2.79 6.2
0.39 6.6
8.7
—
—
CW
Ã
The pellet decomposed during the solubility experiment.
12J. Molla, M. Gonzales, R. Vila, and A. Ibarra, ‘‘Effect of Humidity on Mi-
crowave Dielectric Losses of Porous Alumina,’’ J. Appl. Phys., 85 [3] 1727–9
(1999).
IV. Conclusions
13M. Valant, D. Suvorov, R. C. Pullar, K. Sarma, and N. M. Alford, ‘‘A Mech-
anism for Low-Temperature Sintering,’’ J. Eur. Ceram. Soc., 26, 2777–83 (2007).
14JCPDS-International Center for Diffraction Data. PCPDFWIN Version 2.3, ,
June 2001.
Based on the XRD and TG/DSC-MS results, we can conclude
that the formation of MWO4 (M 5 Ca, Sr, Ba) scheelites goes
through a direct reaction between the MCO3 (M5 Ca, Sr, Ba)
and WO3 reagents. The investigations of the SrWO4 prepared
from reagents with different purities revealed that Na impurity
15J. Krupka, K. Derzakowski, B. Riddle, and J. B. Jarvis, ‘‘A Dielectric
Resonator for Measurements of Complex Permittivity of Low Loss Dielectric
Materials as a Function of Temperature,’’ Meas. Sci. Technol., 9, 1751–6 (1998).

2472
Journal of the American Ceramic Society—Macˇek Krzˇmanc et al.
Vol. 94, No. 8
16Q. Guo and O. J. Kleppa, ‘‘Enthalpies of Formation From the Component
19R. M. German, ‘‘Final Stage Processes: Microstructural Coarsening’’; pp.
127–51 in Liquid Phase Sintering. Plenum Press, New York, 1985.
20M. T. Sebastian and H. Jantunen, ‘‘Low Loss Dielectric Materials for LTCC
Applications: A. Review,’’ Int. Mater. Rev., 53 [2] 57–90 (2008).
Oxides of MgWO4, CaWO4 (Scheelite), SrWO4, and BaWO4, Determined by
High-Temperature Direct Synthesis Calorimetry,’’ Thermochim. Acta, 288, 53–61
(1996).
17Q. Guo and O. J. Kleppa, ‘‘Note on the Enthalpies of Formation of
SrWO4 and BaWO4 Determined by High-Temperature Direct Synthesis Calo-
rimetry From SrCO31WO3 and BaCO31WO3,’’ Thermochim. Acta, 303, 183–6
(1997).
21N. Qin, M. MaWek Krmmanc, and D. Suvorov, ‘‘Glass-Free KxBa1ꢁx
Ga2ꢁxGe21xO8 Ceramics for Low-Temperature Confired Ceramics Technology:
Synthesis, Phase Transition, Sintering, and Microwave Dielectric Properties,’’
J. Am. Ceram. Soc., 27, 1–8 (2008).
18L. A. Al-Hajji, M. A. Hassan, and M. I. Zaki, ‘‘Kinetics of Formation of
Barium Tungstate in Equimola Powder Mixture of BaCO3 and WO3,’’ J. Therm.
Anal. Calorim., 100, 43–9 (2010).
22K. I. Marinakis and G. H. Kelsall, ‘‘The Surface Chemical Properties of
Scheelite (CaWO4) I. The Scheelite/Water Interface and CaWO4 Solubility,’’ Col-
loids Surf., 25, 369–85 (1987).
&