~ACl Fen Rilin1kri EnstitCtsU Dergisi 9.CiiL 2.Say1 2005 ZnC r20~ 'in Nen1 /\ lgil an1n Davrani~inin K
2Cr<) 1 Ekknerek Gcli$tirilmcsi. N. Ka\ asoQlu ....
"'
ZnC
r
20-t 'in NEM ALGILAMA DAVRANISININ K
2Cr0-l EKLENEREK
G
ELiSTiRiLMESi
N. KAVA
S
O(
";
L
U
1 ,M. BA YHAN
2,
S. KA VASOGLl
J
11
Hacett epe Univ., Mi.ih. Fak., Fizik Mlih . 8 6llin1li, Beytepe, Ankara
..
- Mugla Univ., Fen Edeb., Fak., Fizik Boli.itn tL l(otekli, Mugla, Tel: 252 2 1 11 592, E-n1ail:n1bayhan@n1 u.edu.tr
00
OZET
Spine! 7.nCr~0_
1
seran1iginc ~e$ it l i yi.izdese l oranlarda potasyutn krornatla yap tl an katktland1rn1antn nen1e kar~ Jdavrant~t ile ilgi li etk ileri <;al t ~ tltn i$tt r. Nlin1uneler aras1nda sadece 0/o20 l(~CrO'"' i~eren ZnCr~O'"' seratnik sensorUn
0
o25 ile 90 ara~tnda dcgi$en go reli nen1 (RH ) ara1Ign1dak i d .c. dircnc i, i.i<; n1ertcbe degi$illl gosteren Ustel bir
J a v ran 1 $ se r g i I e 111 i $ t i r. Y Li k i I et i t n i n i n e sa se n proton i k o Id u
g
u v e i I et i 111 i n tan c y U z e y I er i n de sog
u r ul a n i n c e s utabakalart ara~tndan n1ctal kontaklara dogru yap tlan yUk transfer i ile kontrol edi ldigi so nucuna va r tln1 t ~t1r. u.c. enlpedan) o i~Un1 so nu~ l an t$tgJnda RC paralc l devre"ine seri bagl1 sab it faz elcnHtnlar1ndan olu$an bi r devre sen or e~dL'gcr de\ rc~ i olarak ()neri ltnek tcd ir.
;\nahta r kcli1nclcr- Ncn1 scnsori.i, ZnCr ,O~-I<:Cr0
4
, Etnpedans spektroskopisiIMPROVE
MENT OF THE HUMIDITY-SENSING BEHAVIOUR
OF ZnCr
20_. BY ADDITION OF K
2Cr0_..
ABSTRACT
The effects or the addition of various percentages of potass iun1 chron1 ate as a sintering aid on the response to air
n10 1~tu re of ZnCr20 1 spine! ceratn ic body we re studied. Only the n1aterial contai ning 20~- o I(~Cr04 in ZnCr._0 4
exhibited an exponential behaviour to hun1 id ity, which shows about three orders change in the
d
.c. resistance overthe relative hu tnidity (RH) range betwee n 25 and 90°/o. It was co ncluded that co nduct ion is due n1ain ly to protonic
n1otio n and is contro lled thro ugh the thin layers of \Vater adsorbed on the surface of the grains. with charge transfer
to the n1ctall ic electrodes. Based on
a.c.
impedance Jneasurernents, an equi valent circuit assoc iated \vith a netvvorkof RC parallel circuit in series \Vith constant phase elen1ents (CPEs) has been suggested.
Key
w
o
rds-
Hutnidity sensors, ZnCr20 4- K2Cr04, ltnpedance spectroscopy
.'.\C' Fen BiliJnlcri ~n stitCJsli o~r.._ Qisi 9.Cilt. /. Sa) I /()()5
1. INTRODUCTION
The growing dernand for the n1 easuren1ent and control of
hun1idity in an industrial or household environrnent has
led to considerable interest in the research devoted to the developn1ent of new tnaterials fo r hutnidity senso rs [
1
,2]. The re lative hutni dity (RH) , wh ich is the ratio of actualvapour pressure at a particular ten1 perature, is con1 n1on ly u ed to n1eas ure h u n1 id ity. Con1n1erc ia lly dcve loped
ensor n1aterials arc either polytncric or porous ceratnic, each of \Vhich have its O\Vn tnerits and provide a direct
electric sig... na l as a fu nction of hutn idity in the
attnosphere. Ceran1 ic type of hun1 id ity senso rs based on
intered oxides via so lid state reactions arc in ... general
superior in perforn1ance to polyn1eric types, due to their hi ~ oh thennal stab ilit ~ and n1echani cal tre n~ ~t h towards a "idc range of operating tetn pcratures, fa. t rcspon e to t h c c ha n g e s o f h u 111 id it y, res is tan c e to con t a n1 i n ants and irradiation and the ease with vvhich it rnay be produced in t h i n fi I 111 f 0 r 111
r
3 ] . H 0 we V er' they are t 0 a c e rt a i n eXtent not full y satisfac tory in tern1s of the need for periodic thern1al cyc ling to recover their hun1idity-sensitive properties . In recent years, nanostructured cerarn ic fi lrnshave been identified as a optitn al candidates fo r hun1idity
sensin... g due to the hi gh surface area exposed for
ad orption of water 1110 lecules
l
4
]
.
'fh c principle of hutnidity n1easuren1ent is the variations in the electrical conduction and capacitance due to water
chetni. orptions and/or capi llary water co nduction \Vithin
the pores [5.6j. When adso rption starts on the clean
ox ide surface, a I ay er of hydroxy I groups is forn1ed. T'he \Vater vapour tno lec ules are chen1isorbed through a dissoc iati ve tnechani sn1 by whi~ ch tvvo surface hydroxyls
per \Vate r n1olec ul e are forn1ed. Th is layer, once fo rn1ed
is not further affected by ex posure to hun1idity. Once the first layer is forn1ed , subsequent layers of \Vater n1olccules are physica lly adsorbed. Many tn ore phy isorbed layers w ill be joined as hun1idity gets
higher. These phys isorbed layers are easily and
reversibly ren1 oved by decreasing the hun1idity.
Physisorbed water n1olecules dissociate, because of the h iCl uh electric fields in the chen1 isorbed water layer. The
charge transport occurs when hydroniutn ions release a proton to neighbouring water n1o lecules, which accepts it
while releasing another proton and so on (G rotthuss's
chain reaction). This proton n1ovcs freely along the water
layer and thus detern1ines the sensor conducti vity [7].
Finall y, at hi gh hu1nidity, liqui d water co ndenses in the
capillary pores. accord ing to Kelvin's la\v, and electrolytic cond ucti on occurs si tn u ltaneous ly \Vith proton ic transport. Materia ls based on zinc chron1ite, as an exatnp le of a ternary n1etal ox ide spinel have long been identified as a ~ cand idate for an active tnaterial in chen1ical sensing a pp I ications
f
8
,
9] . However, air rnoisture sensitive 2ZnCr20~ 'in -:\cn1 Algih.una Da\ rani~inin K ( 'r(),
Eklcncrek Cn~li~tinltnc ·i. "\ . Ka\ a::-.nglu
cerarnic sensors based on pure zinc chron1ite arc
rather poor in perforn1ance and as a rcsul t
cotn posites with T i0 2 or LiZn V0-1 have been
studied instead [ 1 0]. Add itionally, ceratn ic
con1pos ites containing K.:CrO_, or ZnO in zinc
chrotn ite have been suggested as candid ate
n1aterials for hun1idity sensi ng [ 1 I, 12]. In th is article~ the response to relative hutn idity (R// ) of
porous ceran1ics \Vith co n1position Zn Cr~O , I<,CrO~ prepared by so lid state reaction at
elevated ten1perat u res. \Vas studied.
2. EXPERJMENTAL PROCE DURE
Highest purity co n1tnercially ava ilable powders of ZnO and Cr~O : \Vere used as precursors. lntin1ate
tnixture of Zn O:Cr~O, \Vith a n1 olar ratio of I: I - '
was crushed in an agate pestl e and rn ortar for
3
0
n1in. followed by calcination at ternperatures
ranging between 800 and I l OOllC with a heating
rate of SOO')C h-1 in a n1ufne fu rnace (Carbolite
R WF 12/5 ). The con1position and the phase \V ere assessed by X -ray diffractio n patterns (Shitn adzu
XRD-
600
diffracton1eter en1ploying a Cu targetwith iron-filtered radiation of wave length
1.5418 A).
The sensor tnatcrials were fa bricated by adding in
various percentages (by we ight) of potassiun1
chrotnate ( I<~CrO~) to the Zn C r~O 1• The 111 ixture
was g' - round for 30 rn in and cornpacted into pellets (with a 13 111111 in dian1ctcr and l-3 n1tn in thick)
'
at a pressure Of J-2 tons Ctll -, \Vhich vvere then
fi red at a tern pc rate of I I 0 ()' C for about 1 2 hours .
The surface and fractured n1orpho logies of the pellets vvere observed by scann ing electron n1 i c rose o p y ( J eo 1 J S M I C 8 4 8). E 1 en1 e nta I spec ies present in the pellets \\'ere also analysed using energy dispersive analysis by X-rays (EDX). The d .c. resistance as a function of relative hun1idity n1easuren1ents were carried out using two point probe techniq ue. To this purpose, a
f(eithley-236 source and n1 easure unit vvas used.
The san1ples were ini tiall y tnounted in a custom built chatnbcr, where the hutn idity was controll ed by pass ing a carr ier gas through water at
predetern1ined rates. The re lative hutnidity in the
charnber was n1onitored using a Testa 625
refe rence h un1 id ity n1eter. Metallic contacts were
n1ade tl·o nl ga lliu n1 which was fo und to give
Oh n1 ic behaviour. The a.c. in1pedance
n1easuren1ents \Vere perforn1ed using a Hevvlett Packard HP 4 192/\ in1 pedance ana lyser over the
SAC! Fen Bilin1leri EnstitUsCI Dergisi 9.CiiL / .Say1 2005
freq uency range bet\veen 5 Hz- 13 M J-lz,
c\citatio n signal of <SO n1V an1plitude.
.
us1n g an
'-'
3. RESULTS AND DISCUSSION
'"I though the diffraction peaks
a
t
20
11 ( 08 : 8
r
a
gg
I f- ""'0 c:'' ""'<; 9 ' ""'7 .c'' 4""' c \' 54 0° <i7 _,,
d r 1 g ! e ) a n g e s o _) . _, . .) _ . . .) . _, , .) . _, , _ . , ~ . ) ,
an.i 63 .2' obtained fron1 the powder of ?ZnO: I Cr (), ..l
Iter
fir ingat
ten1pcratures bet\veen 800- 1 l OO''C for~:, ·~!1 12 hours are assoc iated with corresponding planes
~nCr~O-l , the peaks at 20
11 angles of 34. 5°, 47.7' .. ,
·: , 68. 1'', 69 .2~~ arc assoc iated with ZnO and the peak
1 angle of 3 1.9(' with Cr,O~. These results indicated
1csp ite the stoic h ion1ctric ratio chosen to produce a
con1pound of ZnCr,O , the forn1ation of ZnCr~O ,
- -l - .,
ot gone to con1p let ion the res idua Is of both ZnO
_
0
5 which are evident in the X-ray patterns ..3 \ie ly, the Bragg peaks in the X-ray diffracto tneter
0 111 the po\vder
or
1 ZnO: I Cr20 , fired at 800 -at '~0
11
') ""' 9 I '' ' _ _) , , I f' 18 4""'" ") 0 ""'""' '' ""'5 74° ""'7 ..., 8° ang es o .1 .) ~ .) . .J.) .) . , .) . .J , .._ 57.47'', 63.121' , 7 1.63'' and 74.70° all pond to p IClnes 0r
ZnCr .0 I i lllp lying the fonnat ionnonorhase cubic body (see Figure 1 ). The Mi ller
and the Bragg peak rat ios estin1ated are in detail
cd elsc\.vhcrc
1131.
The lattice co nstant, 00 ,
d
fro
n1
each difTraction peak was plotted againstJ / sin() (not shown here) and the extrapolation of
ta to
0
= rrI
2 yie lded the lattice constant as'A
[14].
: 2 shows typica l secondary electron micrograph
1e
as- fired surface of ZnCr2
0-~-1(2
CrO-I (20%)fired at I 000(\C fo r about 1 2 hours. The cerarn ic
\Vas porous in nature with a grain size about
1-nd a great nun1ber of pores, indicative of
JLation of one or tnore of the cotnponents
nly, potassiun1) at high te1nperatures. Moreover,
scans over the regions of grain and grain
Jaries revealed that potass iun1 content of grain
· rlaries was significantly higher as COillpa red to that
c grains as is shown in the Figure 3.
-
Cl) ... c: :::s Q) >·
... -ro-
Q) '-c: · -Cl) c Q) _. cZnCr20 4 'in Nen1 Al gilan1a Duvrani $inin K~CrO.~
E klc ne re k Ge I i ~t i ri I n1es i, N. Kav asogl u 1Zn0:1Cr 0 2 3 •
-
1100°C'
,
1 . l • l • l ... , l I ... .. '"' ,... 6 ... • 1 A. ... • ... '" 1 .I A • , -• 800°C L I - J-y -- 1 :=r L ~-~-:L r- _, - - - l -( 111 ) (220)(311 l (400) (422) (511 M440}(620)(533) • 1 • ' ' ' •'
• ' ' ' • 0 10 20 30 40 50 60 70 80 90 28 ( 0 ) BFigure I. X-ray diffraction patterns
o
r
1Zn0 - 1Cr20 1 powder
0
obtained after firin~ g in the ran~ rre 800- 1 I 00 C for about 12
hours in air.
Figure 2. The as-fired surface secondary electron micrograph
of ZnCr
2
0~-K2
C r0-1 (20'Yo) sample sintered at l000°C for about 12 hours.-
U) ...·-
c: :::l Q) >·-
... n:s-
Q) L.. c: -~·-
U) c: Q) _. c: ~ . . 0 50 Grain Pores Cr Zn 100 150 200 Energy, ( eV)Figure 3. Typical EDX spectra recorded from the pellet
containing 20o/o K2Cr0-1. In the inset the potassiutn peak is re
SAl ! Fen Bilin1leri Enstitlisli Dcrgisi 9.Cilt. 2.Sa\'t ?005 ~
Figure 4 shows d .c. resistance as a fun ction of relative
hun1idity for ZnCr::O.~-IZ::CrO-l ceran1ic systen1 at various
percentages of potassiun1 chron1ate. "rhe d .c. res istance
\Vas a strong functi on of the h un1 id ity for all
con1pos itions, decreasing by several ord ers of tnagnitude
upon increasi ng hutnidity, aJthough onl y n1aterial
contatntng 20o/o K::CrO-l displayed an exponential
behaviour as function of
RH
.
This is broadly consistent\Vith \Vhat is norn1ally reported and is generally
understood in tern1s of through an adso rbed \Vater phase
[ 1 5 , 1 6] . The con d u c t ion tn e c ha n i s 111 con c l u de d fro 111 the
evidence given above is ionic and that protons are probably the dotninant charge carriers [ l 7]. A
least-squares best fit to the values of d.c. resistance versus
re lative hurn id ity for the satn p le contai ning 20°/o K.::CrO-~
y i e Id e d log ( R) == 1 0. 2 2 - 0. 0 5 (
RH
)
,
(for R in oh n1 s and RH in o/0). 1010 ' ... .... ... T ' . ' ' • ... ... ZnCr 2 0 :K CrO • • 4 2 4 109"' • • ... ...-
-
... c:-
..
... Q:: • 1 08., • • ... .,-
# Q) . . .. - . ... . . .. ... (J ••••• ... t: ....
. . ro . .....
107"' ••• .. . ... . .. (/) .A. • I ' (/) • K,CrO, (10%) • • ... . Q) •.
.
...,.
.
'- • K,cre. (20%).
.
..
. 106, K,cre. (30%) • u ....
... ., ti • · • Precursor • • 105 • ' • I • .... . ' 20 40 60 80 100Relative humidity, RH (0/o)
Figure -t d .c. resistance versu~ re la tiYc hum id ity characteristic of
ZnCr::(\- K::Cr0
4 ceramic system at various percentages of potassium
chn)matl:.
In an atte tnpt to increase conductivity, CuO was added to so 111 e sa 111 p I e s con t a i n i n g 2 0 o/o pot ass i u 111 c h r o 111 ate. The
corresponding d .c. resistance versus relative hu 111 id ity
characteristics are g.._ iven in figure ... 5 and reveal that the
addition of CuO had a dran1atic effect on the hun1id ity
dependence. The addition of only I (Yo CuO reduced the
resistance at low hun1idity (RH <40o/o) by approxin1ately
an order of n1agn itude, while at higher hun1 id ity it
appeared to have no effect. Increasing the level of CuO
doping not on ly reduced the low hun1idity resistance
even n1orc, it reversed the slope of the characteristic, so
that for RH <40o/o resistance increased with increasing
hun1idity. The reasons for this reversal of slope are not at
present clear.
The variation of the 1nagnitude of itnpedance ( Z )
versus freq uency (
.l)
for the san1ple containing 20%K2Cr0.~ (see Figure 6) under a dry hun1 id arnbient shows
a high frequency fa ll-off of
Z
as.
f
-
o_g, vvhich is due4
ZnCr20-t 'in Nen1 Algilan1a Da,·rani$inin K2Cr() 1
Eklenerck Gcli$tiriln1csi. . I< a\ asuglu
tnainly to capac 1ttve effects \Vithin the systc n1
fro 111 cab I e s. the e q u i p 111 e n t i n put cap a c it an c e an d
ph y s i c a I g_e o 111 et ry of the con tact s . At fr e q u e n c i e s
"
below I 0 Hz, n1agn itude of i n1pedance varies
very slowl y \vith _/ , indicating that res i ·ti ve-type
behav ior is n1ore don1 i nant. Thjs type of be hav ior
is reported wide I y i n s i 111 i I a r cera 1 n i c s y s t e rn s
[ 18,19]. The Co le- Co le plot of con1p lex
" 1/
i 111 p e dance ( i.e., Z versus Z where Z and
I
Z correspond to co n1plex and real part of the
i1npedance) is also displayed in the inset of Figure
6. It consists of a depressed sern ic ire le
corresponding to a paral lel con1bination of a d.c.
conductance and a slightly dispersive capacitance
together \Vith a so-called pseudo- inductance loop
[20]. 109 ' ' ' ' ' ' ' • ' .. ' .... T ! : • 1% Cu0 • • • 2%Cu0 • -
c:
108 ~ • • ... 5% Cu0 ~ • • -Q:: • • •-
"j :.J I 0 Q) 0 (J 107 "! c u e G "l t: tl c i ro IJ •....
... ... ... ~ (/) • l'J•
... ... ... ... ... 0 • ... ... ~· (/) Q) ... '- Ll ...•
106 "'J ... ... 0 • "'J u • • : ... "0 ... • • • • 105 I • T • ' • '.. •
.... ' ' ' ' ' 0 10 20 30 40 50 60 70 80 90 100 RH (%)Figure S. d .c. re~istancc versus relative humidit\·
characteristic
or
the 7.nCr~04-K~Cr04 (20%) ceramic systen1at various percentages
orcuo
.
-
c:
-104 I • • • • • • • I • • • • • • • • • • • • • I.
'· I ~ ., .=;. ,,,o· N ~ • Ill h 10' c • Cl Ill E -0 0 • • • • • • • • • • • • • V 2x10' 4x 1 D~ 6x1o' 8>10' Real z (0) Frequency, f (Hz).
I
•'-\ \•
I I . I \ ' \ . \•
. I .Figure (). A plot
or
Z versus frequency for the samplecontaining 201
Yo K::Cr0-1 under 22<X> RH. In the inset Cole
-Cole plot
or
the impedance measured for this sample is also\ .. \ ( · 1 en B i I i 111 k r i C n s t i t Cl s Cl 0 c r g is i tJ. C i I t. 2 . S u) 1 2 0 0 5
:\ hon1e-n1ade ZnCr:O:-I<_CrO_ (70° o) ccran1 ic hun1id it)
~cnsor \\aS also satisfactori ly calibrat ed u ing a reference
lnttnidit) cnsor SHT75, obtained fron1 ·cnsirion. To
that ain1. a nove l c ircuitry consisting
o
r
a 12 bit 1\ D.~I I 1'75 interface circuit and I to V bridge an1plificr \vas d (: \' e I 0 pc d . T h c t
y
p i c a I sa 111 p I i n g rat~ 0r
d c V c I 0 p e d 1\ IDcircuit is about I 0 kl-lz and it \Vorks 0-5 V range \\'ith an
a pp r o \ i 111 at c I; 1 . 2 5 t n V res o lu t i o n . The c i r c u it i ·
..:~>ntro lled b:r
a
Qbas ic routine on the PC' \\·hich allo\VS to ;~uuire one hu ndred data per second . l'hc SIrr
7.5
i aI
--in~l~ chip null ti sensor n1 odulc inc luding a capacitive
-.icd\ 1ncr hutnidity sensing cletncnt and a band gap
#
_ '11p~raturc ~ensor.
4. CONC L USION
~11oisturc '-.en~i ti vc ZnCr_()-1 ceran1 ic body produced
., the so ltd ~tale reaction bct\veen 1.inc oxide and
) n 11 L\ ox 1 de at c I c \ at c d t e 111 pc rat u r c s d is p I aye d a
·r poor .;;cn~ing per r()rnla nce to hutnidi t) . ll o\vever.
~~istancc a~ a fu nction of re lati ve hu n1id ity ShO\Ved
· I.g depende nce, dccrca ing over several orders
o
r
tu<.k \\ ith incrca~i n... ~ hu n1idit\' ~ bc in ... ~ a ·n1ooth
~ n c n t i a I i n t h c cas
c
o f the sa 111 p I c prep arc d us i n g· ., ). ('~0°o) in /n('r .C)
1• The addi ti on of C'u() resulted
,1 incrl'a',c in the co nductivitY bu t had a deleterious
~
et 011 the hu tni dit\ . These n~~u lt~ indi cate that
~
·· () I" l rl)
.
~\ .. ten t once calc ined, co n1pactcd and_tld unde r proper conditions ha· potent ial use as an
l nuttcria l for the hutnidi tv se.. nsors.
AC KN()WLEDGE M ENTS
· authu rs like to ackncnvledge financial support by the
1 \'Cr'51l\' o l' Mugla, Research Project Foundati on (AFP
· t ' . I h c \ ~ arc .... ~rat c ru l to A . W . B r i n k 111 an for he I p fu I
· ..!C .... til1ns and disc uss ions. Thanks are also due to $.
l t;k ror hi', conti nuous support of this \Vork.
5
ZnCr20 4 'in Nc rn /\lgilama Dav rani~ inin K:Cr04
Eklcncrek Gc li~tirilmcs i. N. Kc\\ asoglu
REFERENCES
[ I ] . N . Y a 111 az o e and Y . S h i 111 i z u , Se n s . 1\ c tu at or s,
10 ( 1986) 379.
f2l . 11. Arai and T. Seiyan1a, ''1-ftnnidity sensors"
in W. Gc1 pcl. J. 1-fesse and J. N. Ze1ncl ( Eel s.),
''Sensors. ('heJnicul und !3 iocln}lnicul
Sensors". VC"' I-1. \Veinhein1. 199 1 .
[3]. S. Aga\\·al. G.L. Shanna. Sens. Actuato r~ B.
85 (200') 205.
f 4
1
.
W . P·
r
a i and J . H 0 h . T h in o I id F i I n1 s. 4; 2 (2002) 22 0. [5]. B. M. I<ul\vick i, J. An1. Cerarn. Soc., 74 (4) ( 199 1) 697.[6]
J. Hole, .J. Slunecko,M
.
Ilrovat. Sens. Actuators 8 , 26-27 ( 1995) 3 12.[
7
].
J.H
.
i\nderson andG
.I\ . Park, J. Ph ysical Chen1i ·try, 72 ( 1 0) ( 1968) 3662 .[8]. Wu Ming-Tang, Sun Hong-Tao and Li Ping.
Sens. Actuators B, 17 ( 1994) 9. [9]. X. Sun . N. W. Jones, J. C. Gcsic k~ L. Xu and G. W. Roberts. Appl ied Catalys is A, 231 (7002) 269. [ 1 0] . Y . Y o k o n1 i z o. S . U no. M . I-I a nlt 8 and H . ll iraki, Scns. Actuator . 4 ( 198 '"\ ) 599. 111 1. M. Rayhan. T. Hashcrni and /\ . W. Brin k nHln , J. Mat cr. · c i., 3 2 ( I 9 9 7) 6 6 I 9. I_ I ' ]. S . P o k h r c I. B . .I e y a raj . J( . .' . N agar aj a. Mater. I.ctt .. 1-6 (2003 ) 4378.
[1 3J. N. Kavasoglu. Functionct!
I hun idirr . Sen.,·ors, M .Se. The is.
of Mu~ Qia, 200 I. C'erLlln ic Un iversitv "" [ 14]. JCPD ' (International Centre of Diffraction Data) Card No: 22- 1 107, 1996 . [ I
5
1
.
G .
'To s h k o, P. N en o v, D.S
tef
c
h o and P.Yorclanov. "( 'er cflnic Sensors Technology and
;( p
p
li
c u 1 ions", 'T e c h n o 111 i c P u b I ish i n g C o n1 pan y, Pen n s y lava n i a, I 9 9 6 . [ 16]. E. McCafferty and A. C. Zettletnoyer . Discus. Faraday Soc., 52 (197 1) 239. [ 171. P. Knauth, J. Electroceran1ics, 5 (2) (2000) l I I . 11 8] . M . J . I I o g an, A . W . 8 rink n1 an and T. Hashcn1i, 1\ppl. Phys. Lett. 72 ( 1998) 3077.r
l 9] . K . V ij a)' a In 0 ha n an' G . path a k' M . p . V i n 0 d.S. S. Pin..., ~a l c and S. F. PatiL Mater. Chen1 .
Phys., 46 ( 1996) 72.
[20 ]. A. I<. .J on se her. D iele ctric Relaxation in
Solids (Chelsea Dielectric Press, London
