Phase evolution and dielectric properties of MgTiO3-CaTiO 3-based ceramic sintered with lithium borosilicate glass for application to low temperature co-fired ceramics

Article abstract of DOI:10.1111/j.1551-2916.2005.00453.x

Effects on phase evolution caused by the addition of a new sintering agent, lithium borosilicate, Li₂O · B₂O₃ · SiO₂ (LBS) glass to 0.9MgTiO₃-0.1CaTiO ₃ ceramic and resultant dielectric properties were investigated. The added LBS glass, a liquid phase sintering agent, significantly lowered the densification temperature from 1300° to about 950°C, while yielding decomposition of MgTiO₃ into MgTi₂O₅ and Mg₂TiO₄. At the same time, the by-products of the decomposition reaction, MgO and TiO₂, were dissolved into the glass network. Such phase evolution partly compensated the influence of deleterious glass addition so that the specimen demonstrated fairly good apparent dielectric properties.

Full text of DOI:10.1111/j.1551-2916.2005.00453.x

J. Am. Ceram. Soc., 88 [9] 2461–2465 (2005)
DOI: 10.1111/j.1551-2916.2005.00453.x
r 2005 The American Ceramic Society
ournal
J
Phase Evolution and Dielectric Properties of MgTiO –CaTiO -Based
3
3
Ceramic Sintered with Lithium Borosilicate Glass for Application to
Low Temperature Co-Fired Ceramics
Ã
Hee-Kyun Shin
School of Materials Science and Engineering, Seoul National University, Seoul 151-744, Korea
Hyunho Shinw
Department of Ceramic Engineering, Kangnung National University, Kangnung 210-702, Korea
Seo-Yong Cho and Kug Sun Hong
School of Materials Science & Engineering, Seoul National University, Seoul 151-744, Korea
Effects on phase evolution caused by the addition of a new sin-
2 3
For the case when 5–10 mol% B O was added in their work,
modification of the dielectric properties was minimal, while the
.
.
tering agent, lithium borosilicate, Li O B O SiO2 (LBS)
glass to 0.9MgTiO –0.1CaTiO ceramic and resultant dielec-
2
2
3
required sintering temperature was as high as 12001C. Jantunen
3
3
5
,6
tric properties were investigated. The added LBS glass, a liquid
phase sintering agent, significantly lowered the densification
temperature from 13001 to about 9501C, while yielding decom-
position of MgTiO into MgTi O and Mg TiO . At the same
et al. have also investigated the effect of the addition of several
borosilicate glasses RO–B –SiO (R5 Zn, Ba) to the
MgTiO –CaTiO system. In their work, as high as 70 wt% glass
O
2 3
2
3
3
phase in the ceramic was required to lower the effective sintering
temperature to 9001C. Therefore, efforts to investigate appro-
priate sintering agent (glass system), which ensures suitable
densification as well as appropriate dielectric properties, are of
importance for LTCC technology. In the present work, as an
3
2
5
2
4
time, the by-products of the decomposition reaction, MgO and
TiO , were dissolved into the glass network. Such phase evolu-
2
tion partly compensated the influence of deleterious glass addi-
tion so that the specimen demonstrated fairly good apparent
dielectric properties.
5
,6
extension to the previous borosilicate glass system, the feasi-
bility of a new lithium borosilicate, Li O ꢂ B O ꢂ SiO (LBS)
2
2
3
2
glass composition has been investigated as a sintering agent for
3 3
the MgTiO –CaTiO system. This work will demonstrate that
the studied new sintering agent is capable of ensuring suitable
densification as well as appropriate dielectric properties.
I. Introduction
OW-TEMPERATURE co-fired ceramic (LTCC) technology has
received much attention because of the benefits of package
L
miniature, reduced cost, and suitable high radio frequency (RF)
applications to wireless local area networks (LAN) and mobile
microwave communication module parts such as bluetooth, res-
II. Experimental Procedure
1
onator, phase shifter, and duplexer, etc. MgTiO –CaTiO is a
3
The starting materials to individually prepare MgTiO3 and
3
well-known ceramic material system for LTCC technology be-
cause of its appropriate dielectric constant (k) for microwave
dielectric applications, high quality factor (Q ꢀ f), and high tem-
CaTiO
Tokyo, Japan), TiO
3
compounds were MgO (High Purity Chemical Lab.,
(rutile, High Purity Chemical Lab.), and
CaCO (High Purity Chemical Lab.), all of which were 99.9%
2
3
perature-stability (t ). Above all, the dielectric properties of
pure powders. An appropriate amount of starting powders were
f
MgTiO
of each constituent, i.e., illmenite-structured MgTiO
Q ꢀ f 5 160000 GHz, t 5 ꢁ45 ppm/1C) and perovskite-struc-
3
–CaTiO
3
can be tailored by controlling the molar ratio
mixed and ball-milled in a polyethylene bottle with ZrO
2
balls
) as
3
(k 5 17,
for 24 h using ethanol (MgTiO ) and distilled water (CaTiO
3
3
medium. Each mixture was rapidly dried and calcined at 11001C
for 2 h.
f
2
,3
tured CaTiO
For LTCC applications, the addition of a glass to the
MgTiO –CaTiO ceramic composition is required to lower the
3
(k 5 170, Q ꢀ f 5 3600 GHz, t
f
5 800 ppm/1C).
The glass was prepared by mixing LBS with a weight percent
ratio of 3:6:1 and melted at 9001C for 1 h in a platinum crucible.
Samples were quenched in air and pulverized to pass 200 mesh
3
3
sintering temperature; however, this addition inevitably sacrific-
es dielectric properties. Thus, in studying the influence of a
sintering agent, the issues regarding the lowering of the
z
sieve. The calcined MgTiO , CaTiO
3
3
, and glass frits were mixed
–0.1CaTiO (MCT)
stoichiometrically to prepare 0.9MgTiO
3
3
densification temperature and the resultant dielectric properties
with varying weight percentages of LBS glass. Then, the mix-
ture was ball-milled for 48 h and dried, granulated, and pressed
at 1000 kg/cm to form pellets 8 mm in diameter and 3 mm in
4
are important. In a previous work by Yamamoto et al.,
mol% addition of V to 0.95MgTiO –0.05CaTiO
a
lowered
3
7
O
2 5
3
3
the sintering temperature from about 13001 to about 10001C
while displaying significant deterioration of dielectric properties.
thickness. The pellets were sintered from 9001 to 10501C for 2 h
at a heating rate of 51C/min.
Shrinkage of the specimens during heating was measured us-
ing a horizontal-loading dilatomter with alumina rams and
boats (model DIL 402C, Netzsch Instruments, Shelb/Byern,
Germany). The crystal structure of sintered sample was inves-
tigated using X-ray powder diffraction (model M18XHF, Mac-
T. Vanderah—contributing editor
Manuscript No. 20308. Received July 28, 2004; approved March 16, 2005.
Current address: Samsung Corning Precision Glass, Asan, Chung-Nam, 336-840,
Ã
z
Through a separate dilatometric expansion analysis of the compacted LBS glass frits,
glass transition temperature was about 2301C. The compacted LBS glass frits crystallized to
Korea.
wAuthor to whom correspondence should be addressed. e-mail: hshin@kangnung.ac.kr
2 4 7
Li B O at about 5001C followed by melting at about 7201C.
2
461

2
462
Journal of the American Ceramic Society—Shin et al.
Vol. 88, No. 9
science Instruments, Yokohama, Japan) in the 2y-range from
01 to 601. Polished and thermally etched surfaces of sintered
*
s
1
* * s
000°C
2
*
*
s
s
m b c
*
*
b
specimens were examined using field emission scanning electron
microscopy (FESEM: model JSM-6330F, JEOL, Tokyo,
Japan). Microwave dielectric properties of sintered samples
were measured using a network analyzer (model HP8720C,
Hewlett Packard, Palo Alto, CA) in a frequency range from 8
to 10 GHz. The Q ꢀ f was measured through the transmission
*
s
*
* *
*
*
*
s
*
950°C
m b c
b
*
7
s
s
cavity method using a Cu cavity and Teflon supporter. Permit-
8
tivity (k) was measured using the post resonator method and
s
s
m
m
b
c
900°C
800°C
*
T
T
b
b
*
f
the temperature coefficient of the resonant frequency (t ) was
measured using an Invar cavity in a temperature range from 101
9
to 801C.
L
L
T
b c
c
T
T
*
*
III. Results and Discussion
7
50°C
T
T
s
T
T
T
s
s
T
T
m b
b
b
(1) Densification, Phase Evolution, and Microstructure
Figure 1 shows the change in density of MCT ceramic with
varying amounts of LBS glass as a function of sintering tem-
perature. Although the theoretical density for full densification
is hard to determine because of the presence of LBS glass and
complicated material reactions in the specimen (will be discussed
later), relative density has been determined Fig. 1(b) by assum-
ing no material reactions in order to roughly estimate the degree
of densification. From Fig. 1, 9501C seems to be sufficient for
the appropriate densification of 10 and 20 wt% glass-added
700°C
L
s T
c
c
M
M
b
6
50°C
T
s
2
0
30
40
θ (CuKα)
Fig. 2. X-ray diffraction patterns of 0.9MgTiO
50
60
2
3
–0.1CaTiO
3
ceramic
(
a)
sintered with 20 wt% lithium borosilicate glass at varying temperatures.
Ã
Unmarked peaks are from MgTiO
Mg TiO ; c, CaTiO ; T, TiO (rutile); m, MgO; L, Li
(Mg,Ti) (BO )O.
3
.
2 5
, MgTi O ; M, MgTiO
3
; s,
; and b,
2
4
3
2
2 4 7
B O
3
3
2
2
.5
.0
.5
.0
2
3
specimens. For the 5 wt% glass-added specimen, acceptable
densification is achieved at above 10001C. Considering that
glass-free MCT specimen requires sintering temperatures above
about 13001C, the addition of LBS glass in this work signifi-
cantly lowers the sintering temperature. In the cases of speci-
mens with a relatively large amount of glass, e.g., 20 wt%, firing
at overly high temperatures, seems to slightly lower the density,
probably because the volatile species in the glass, such as lithium
and boron oxides, increase porosity.
5
1
2
wt% LBS in MCT
0 wt% LBS in MCT
0 wt% LBS in MCT
The phase evolution during the sintering of 20 wt% glass-
added specimens has been investigated as a function of rising
temperature using XRD and the result is shown in Fig. 2. In
order to efficiently study the change in maximum peak intensi-
ties with rising temperature, the samples were pulled out from
9
00
950
1000
1050
Temperature (°C)
y
the furnace when the intended temperature was reached. As
seen in the figure, the decrease in peak intensities of MgTiO
3
1
b)
00
(
(
unmarked in Fig. 2) is apparent as temperature rises unlike the
3
case of CaTiO (denoted as c). In addition, various secondary
Ã
phases such as MgTi O (marked as ), Mg TiO (marked as s),
2
9
8
7
6
5
0
0
0
0
0
5
2
4
2
MgO (marked as m), and TiO (marked as T) are observed. The
increase in the peak intensity of secondary phases such as
MgTi O and Mg TiO with the temperature is so apparent
2
5
2
4
that, at 10001C, MgTiO (unmarked) is no longer the major
3
Ã
phase: MgTi
the major phases.
From the decrease in XRD peak intensity of MgTiO and
2 5 2 4
O (marked as ) and Mg TiO (marked as s) are
3
increase in Mg
2 4 2 4
TiO with temperature, Mg TiO is interpreted
to form by the decomposition of MgTiO
3
,
5
1
2
wt% LBS in MCT
0 wt% LBS in MCT
0 wt% LBS in MCT
2
MgTiO ¼ Mg TiO4 þ TiO2 ðdissolved into LBS glassÞ
(1)
3
2
In Fig. 2, the peak intensity of Mg TiO
2
tonically increases with temperature while that of TiO (marked
2
as T), which increases up to 7001C, diminishes above 7001C. The
peak intensity eventually disappears above 9501C. From this
4
(marked as s) mono-
9
00
950
1000
1050
Temperature (°C)
Fig. 1. Change in apparent density of 0.9MgTiO –0.1CaTiO
3
3
(MCT)
ceramic as a function of sintering temperature at various lithium boro-
silicate (LBS) glass addition levels.
yIdentifying the specimens held for 2 h at the firing temperatures showed almost
fully grown peaks so that the change in peak intensities with rising temperature was less
significant.

September 2005
Phase Evolution and Dielectric Properties of MgTiO –CaTiO3
3
2463
MCT + X wt% LBS (1000°C/2h)
*
s
*
2
0 wt%
s
s
m
b
*
*
b
*
*
s
*
**
*
*
*
1
0 wt%
5
wt%
2
0
30
40
θ (CuKα)
Fig. 3. X-ray diffraction patterns from 0.9MgTiO
50
60
2
3
–0.1CaTiO
3
(MCT)
ceramic sintered with 5, 10, and 20 wt% lithium borosilicate (LBS) glass
at 10001C for 2 h. Unmarked peaks are from MgTiO ; s,
. Ã, MgTi
Mg TiO ; m, MgO; and b, (Mg,Ti) (BO )O.
3
2 5
O
2
4
2
3
2
result, TiO is interpreted as dissolving into the LBS glass as the
temperature rises.
Similarly, MgTi O forms via the reaction,
2
5
2
MgTiO ¼ MgTi O5þMgO ðpartially dissolved into LBS glassÞ
3
2
(
2)
Ã
2 5
Although MgTi O (marked as ) and MgO (marked as m)
form simultaneously, in Fig. 2 the increase in the peak intensity
of MgTi O with temperature is drastic whereas that of MgO is
2
5
moderate. This difference implies that MgO is also partially dis-
solved into the LBS glass as temperature increases.
Having shown the effect of temperature on the phase evolu-
tion in Fig. 2, the effect of glass addition on the phase evolution
at a constant temperature, i.e., 10001C, is shown in Fig. 3. In
order to allow enough reaction time between the glass and the
MCT ceramic, samples were held for 2 h at 10001C. In Fig. 3,
the peak intensities of MgTiO decrease with glass addition. At
3
2 5 2 4
the same time, the peak intensities of MgTi O and Mg TiO
increase with the addition of LBS glass: eventually they are more
predominant over the MgTiO phase in 20 wt% glass-added
specimen. This result indicates that the decomposition of
3
Fig. 4. Scanning electron micrographs of sintered 0.9MgTiO
TiO
3
–0.1Ca-
MgTiO
Fig. 2) but also by the increase in glass addition (Fig. 3).
Figure 4 shows a SEM micrograph of polished specimens
3
is fostered not only by the increase in temperature
3
(MCT) ceramic at 9501C for 2 h with varying amounts of lithium
borosilicate (LBS) glass: (a) 5, (b) 10, and (c) 20 wt%.
(
sintered at 9501C for 2 h with varying amounts of LBS glass. As
shown in this figure, the 5 wt% specimen yields apparent for-
mation of pores as consistent with the trend in density (Fig. 1).
The increased addition of glass to 10 wt% results in an enhanced
densification with the growth of grains, which is attributed to
liquid phase sintering by the low-temperature-softening LBS
glass. In a separate work, the LBS glass showed a drastic sof-
tening at about 7001C in a thermal shrinkage measurement. The
glass addition as high as 20 wt% yields further growth of grains.
Based on energy dispersive spectroscopy (EDS) (not shown),
grains in 20 wt% specimen (Fig. 3(c)) mostly consist of MgTi O
vation. In summary, the addition of LBS glass into MCT
ceramic results in an improved densification (Fig. 1), decompo-
3
sition of MgTiO into the secondary phases (Figs. 2 and 3), and
grain growth (Fig. 4).
(
2) Microwave Dielectric Properties
Microwave dielectric properties of the studied specimens with
varying amounts of LBS glass are shown in Fig. 5 as a function
of sintering temperature. For the interpretation of the results,
2
5
(
marked as ‘‘a’’) and Mg
minor volume of CaTiO (marked as ‘‘c’’), LBS glass (marked as
‘d’’), and undecomposed MgTiO (marked as ‘‘e’’). These mi-
2
TiO
4
(marked as ‘‘b’’) grains with a
the reference material MCT and the secondary phases MgTi
and Mg TiO have been prepared separately and their dielectric
properties are shown in Table I.
2 5
O
3
2
4
z
‘
3
crostructural features are in accordance with the XRD obser-
In Fig. 5(a), the permittivity of the glass-added specimens
such as 10 and 20 wt% samples show decreased values from the
MCT reference, in proportional to the amount of the added
glass. This decrease seems to be natural as the permittivity of the
zThe elongated ones were rich in silicon in EDS unlike any other grain shapes, indicating
the LBS glass region: it is possibly localized during thermal etching for sample character-
ization.

2
464
Journal of the American Ceramic Society—Shin et al.
Vol. 88, No. 9
(
a)
The change in the Q ꢀ f of the specimens with varying
amounts of LBS glass is shown in Fig. 5(b) as a function of
sintering temperature. The decrease of the Q ꢀ f of the 5 wt%
specimen at temperatures below 10001C is again correlated with
the poor densification of the specimen. For the case of well-sin-
tered specimens at above 10001C, a decrease in the Q ꢀ f from
the MCT reference is apparent with the glass addition, although
the additions more than 10 wt% up to 20 wt% did not yield
further decrease. The reason why no further decrease resulted
can be explained by the fact that the formation of the secondary
2
2
1
1
4
1
8
5
2 5 2 4
phases such as MgTi O and Mg TiO is enhanced by the fur-
(
b)
ther addition of the LBS glass (Fig. 3). As seen in Table I, the
Q ꢀ f of such secondary phases are fairly high, e.g., 47 000–
4
3
2
1
0000
0000
0000
0000
55000 GHz, even higher than the MCT (42 800 GHz). Thus,
although the 10–20 wt% addition of glass itself would tend to
decrease the Q ꢀ f, its enhancement of the decomposition of
2 5 2 4
MCT to MgTi O and Mg TiO (Fig. 2) is interpreted to in-
crease the Q ꢀ f; thereby, the overall Q ꢀ f is compensated.
Figure 5(c) shows the temperature coefficient of the resonance
frequency for specimens with varying LBS glass. In the figure,
all the glass-added samples show fairly diminished t
of MCT ceramic (t 5 40 ppm/1C). This decrease is associated
with the fact that the t of appearing (forming) compounds
f
from that
f
(
c)
MCT
wt% LBS in MCT
0 wt% LBS in MCT
f
4
3
2
1
0
0
0
0
0
MgTi
are more negative than the disappearing (decomposing) com-
pound MgTiO (t 5 40 ppm/1C) by the glass addition. Howev-
er, the contribution of the MgO- and TiO -dissolution into the
O
2 5
(t
f
5 ꢁ66 ppm/1C) and Mg
2
TiO
4
(t
f
5 ꢁ54 ppm/1C)
5
1
3
f
20 wt% LBS in MCT
2
LBS glass to the t
Although the decomposition of MgTiO
dition decreases t in general, the detailed trends with temper-
ature and glass addition, as shown in Fig. 5(c), require
interpretation. For the case when 5 wt% glass is added, the t
f
of the specimen is not clear at this moment.
3
because of glass ad-
f
−
10
f
9
50
1000
1050
1300
decreases at temperatures below 10001C, which occurs simulta-
neously with the poor densification of the samples in this tem-
perature range. For the well-sintered specimens at temperatures
above about 10001C, increase in glass addition from 5 to 10
Temperature (°C)
Fig. 5. Microwave dielectric properties of 0.9MgTiO
3
–0.1CaTiO
3
(
MCT) ceramic as a function of sintering temperature for varying lith-
ium borosilicate (LBS) glass additions. (a) relative dielectric constant, k,
b) quality factor, Q ꢀ f, and (c) temperature coefficient of resonant fre-
quency, t . The reference sample MCT (denoted as star) was sintered at
3001C for 2 h.
wt% decreases the t
which, as mentioned above, is believed to be associated with the
more enhanced formation of the secondary phases, MgTi
and Mg TiO having highly negative values.
f
from about 13 to roughly about 0 ppm/1C,
(
f
2 5
O
1
2
4
The specimens with 10 and 20 wt% glass additions are well-
densified throughout the temperature range shown in Fig. 5(c).
Comparing these specimens, the increase in glass addition from
Table I. Microwave Dielectric Properties of MgTi O and
2
5
2
Mg TiO3
f
10 to 20 wt% does not necessarily diminish t correspondingly at
temperatures between 9501 and 10001C. This phenomenon is
hardly interpretable because the enhanced decomposition of
Sintering condition
k
Q ꢀ f
t
f
MgTi O
2
15001C, 2 h
13001C, 2 h
17.4
14.4
47 000
55 000
ꢁ66
ꢁ54
MgTiO
3
and formation of MgTi
2
O
5
and Mg
2 4
TiO with in-
5
Mg TiO
2
creased glass addition should have lowered t
at 9501–10001C. Such behavior can only be interpreted by as-
suming that the change in the composition of the LBS glass be-
f
correspondingly
3
cause of the dissolution of MgO and TiO
value, inhibiting the corresponding decrease in t . Such a role of
2
tend to increase t
f
f
1
0
J
LBS glass (k 5 7.97 ) is lower than that of MCT (k 5 23). For
the case of 5 wt% specimen, especially when sintered at tem-
peratures lower than 10001C, the permittivity is even lower than
the 10 and 20 wt% specimens although the glass addition was
minimal. The presence of pores (k 5 1) in this specimen because
of the poor densification at low temperatures (Figs. 1 and 4(a)) is
interpreted to yield such behavior.
MgO- and TiO -dissolution into LBS glass is further supported
by the fact that the 20 wt% glass added specimen shows greatly
2
f
increased t at 10501C. Since the increased temperature enhances
the decomposition of MgTiO (the source for lowering t ), the
3
f
plausible source for such a drastic increase in t at 10501C can
only be assigned to the change in glass composition because of
f
Regarding the formation of the secondary phases during the
densification, the presence of the phases such as Mg TiO and
MgTi O does not seem to much alter the permittivity of the
2 5
2
the dissolution of MgO and TiO . A separate work is under way
to further confirm this proposition. In summary, at high tem-
perature and high glass addition conditions, there is a tendency
2
3
overall specimen since their permittivity values (Table I) are
similar to the range of the overall values, i.e., 13–17 (Fig. 5(a)).
The dissolved byproduct TiO and MgO by Eqs. (1) and (2) into
that the contribution from the formation of MgTi
Mg TiO phases (which decreases t ) is compensated or over-
whelmed by the change in glass composition because of the dis-
solution of MgO and TiO (which is interpreted to increase t ).
From above, the studied LBS glass system is shown to ensure
the densification and decomposition of MCT ceramic at a fairly
2 5
O and
2
4
f
2
the LBS glass is anticipated to have increased the permittivity of
the glass itself. This increase is the result of these oxides behav-
ing as if their permittivities were 25.5 and 14.0, respectively,
2
f
1
1
once they are dissolved in the glass. These values are much
higher than the LBS glass itself (7.97). After all, the eventual
permittivity values of the studied specimens are 13–17, which are
in the intermediate range between the values of MCT ceramic
J
2
Although the physical origin is not clear, dissolution of crystalline TiO , having the
most high permittivity (k 5 100) among common oxides for glass network, results in higher
effective permittivity (k 5 25) than any other commonly dissolved oxides in glass: TiO dis-
in glass were 25 . Similarly, the very
(400 ppm/1C) may also increase the t of the glass system when it
2
1
1
solution increases the k of the glass as if the k of TiO
high t of crystalline TiO
is dissolved.
2
f
2
f
(23) and LBS glass itself (7.97).

September 2005
Phase Evolution and Dielectric Properties of MgTiO –CaTiO3
3
2465
low temperature. The LBS glass-added specimen yields fairly
good resultant dielectric properties for the application of the
studied material to LTCC technology. The 10 wt% LBS glass-
added specimen fired at 9501C demonstrated a permittivity of
even be increased. The investigated new sintering agent, lithium
borosilicate glass, was capable of ensuring suitable densification
as well as fairly good apparent dielectric properties: the 10 wt%
LBS glass-added specimen fired at 9501C for 2 h demonstrated
1
9.1, a Q ꢀ f of 13000 GHz, and a t of ꢁ1.3 ppm/1C.
a permittivity of 19.1, a Q ꢀ f of 13 000 GHz, and a t of
f
f
ꢁ
1.3 ppm/1C.
IV. Conclusions
References
The effects of LBS glass addition to MCT ceramic on densifi-
cation and phase evolution, as well as the resultant dielectric
properties, were investigated. The glass addition apparently im-
proved the densification of the specimen via liquid phase sinte-
1
C. Q. Scrantom and J. C. Lawson, ‘‘LTCC Technology: Where We Are and
Where We’re Going-II,’’ IEEE Symposium on Technologies for Wireless Applica-
tions; pp. 193–200. IEEE, New York, 1999.
K. Wakino, ‘‘Recent Development of Dielectric Resonator Materials and
2
Filters in Japan,’’ Ferroelectrics, 91, 69–86 (1989).
R. C. Kell, A. C. Greenham, and G. C. E. Olds, ‘‘High-Permittivity Temper-
ring. During the firing process, decomposition of MgTiO
3
to
3
MgTi O and Mg TiO was apparent, yielding dissolvable MgO
2
5
2
4
ature-Stable Ceramic Dielectrics with Low Microwave Loss,’’ J. Am. Ceram. Soc,
56 [11] 352–4 (1973).
H. Yamamoto, A. Koga, S. Shibagaki, and N. Ichinose, ‘‘Low Temperature
2
and TiO , respectively, into the LBS glass. Such phenomenon
4
was further enhanced as either the temperature or glass addition
increased. Permittivity of the specimen decreased with the ad-
dition of the low-permittivity glass, as qualitatively consistent
with the rule of mixture. The glass-assisted decomposed prod-
Firing of MgTiO
3 3 2 3
–CaTiO Microwave Dielectric Ceramics Modified with B O or
,’’ J. Ceram. Soc. Jpn., 106 [3] 339–43 (1998).
2 5
V O
5
H. Jantnunen, R. Rautioaho, A. Uusima
¨ ¨
ki, and S. Leppavuori, ‘‘Composition
of MgTiO –CaTiO Ceramics with Two Borosilicate Glasses for LTCC Technol-
3
3
ucts, either crystalline phases such as MgTi
the dissolved products MgO and TiO , did not much alter the
2
overall permittivity value. The Q ꢀ f of the specimen also de-
creased because of the addition of low-Q ꢀ f glass. However,
an overly high addition, e.g., 20 wt%, did not decrease the
factor correspondingly, because of the enhanced formation of
2
O
5
and Mg
2
TiO , or
4
ogy,’’ J. Eur. Ceram. Soc, 20 [14–15] 2331–6 (2000).
6
¨ ¨
H. Jantnunen, A. Uusimaki, R. Rautioaho, and S. Leppavuori, ‘‘Temperature
Coefficient of Microwave Resonance Frequency of a Low-Temperature Cofired
Ceramic (LTCC) System,’’ J. Am. Ceram. Soc., 85 [3] 697–9 (2002).
7
D. Kaifez and P. Gullion, Dielectric Resonators; pp. 327–76. Artech House,
Norwood, MA, 1986.
B. W. Hakki and P. D. Coleman, ‘‘A Dielectric Resonator Method of Meas-
8
uring Inductive Capacities in the Millimeter Range,’’ IRE Trans. Microwave
Theory Technol., 8, 402–10 (1960).
T. Nishikawa, K. Wakino, H. Tamura, H. Tanaka, and Y. Ishikawa, ‘‘Precise
the MgTi
of the temperature coefficient of resonance frequency, t
decomposition products MgTi O and Mg TiO4 decreased
O
2 5
and Mg
2
TiO
4
having very high Q ꢀ f. For the case
9
f
, the
Measurement Method for Temperature Coefficient of Microwave Dielectric Res-
onator Material,’’ IEEE Microwave Theory Technol. Symp. Dig., 3, 277–80 (1987).
2
5
2
the t , while the change in glass composition because of the dis-
f
10
D. K. Lee, Low Firing and Microwave Dielectric Properties of BaTi
Glass Addition; pp. 47, M.S. Thesis, Seoul National University, 2000.
4 9
O by
solution of MgO and TiO
to increase the t . Thus, an overly high addition of glass, e.g.,
0 wt%, does not necessarily decrease the t ; in fact, the t can
2
was interpreted to be a source
11
f
H. Scholze, Glass, Nature, Structure, and Properties, 3rd edition; p. 189, 318.
Springer, Berlin, 1991.
2
&
f
f