This is he accep ed manusc ip p io o house edi s. ©2005 IOP Publishing L d
The Ve sion o Reco d is a J. Phys.: Condens. Ma e 17 (2005) 2645-2654 doi:10.1088/0953-8984/17/17/012
In luence o he elec ic ield on he la en hea o e oelec ic phase ansi ion in
KDP
Jose Ma ia Delgado-Sanchez, Jos´e Ma ´ıa Ma ´ın-Olalla, Ma ´ıa Ca men Galla do, and Sa u io Ramos
Depa amen o de F´ısica de la Ma e ia Condensada. Ins i u o Mix o de Ciencia de Ma e iales
CSIC-Uni e sidad de Se illa
Ap Co eos 1065, ES-41080 Se illa, SPAIN
Ma celi Ko alewski
Ins i u e o Physics, Adam Mickiewicz Uni e si y, Umul owska 85, 61-614, Poznan, Poland
Jaime del Ce o
Depa amen o de F´ısica de la Ma e ia Condensada. Ins i u o Mix o de Ciencia de Ma e iales.
CSIC-Uni e sidad de Se illa
Ap Co eos 1065, ES-41080 Se illa, SPAIN
(Submi ed: 31 Janua y 2005; Re ised: Ma ch 24 2005; Published: 15 Ap il 2005)
The speci ic hea , hea lux (DTA ace) and dielec ic cons an o KDP e oelec ic c ys al ha e
been measu ed simul aneously o a ious elec ic ields wi h a conduc ion calo ime e . The speci ic
hea p esen s a s ong anomaly bu hese simul aneous measu emen s allow us o e alua e he la en
hea accu a ely. La en hea dec eases wi h ield and he alue o c i ical elec ic ield — ha whe e
la en hea disappea s— is es ima ed o be (0.44 ±0.03) kV cm−1. Inciden ally, we ha e measu ed
simul aneously he dielec ic pe mi i i y which sugges s ha la en hea is de eloped as domains
a e g owing.
I. INTRODUCTION
The KDP amily is one o he mos ex ensi ely
s udied[1, 2] hyd ogen-bonded e oelec ic c ys als. The
po assium dihyd ogen phospha e KH2PO4c ys al ex-
hibi s a discon inuous phase ansi ion a T0= 121 K
om a e agonal pa aelec ic phase o an o ho hombic
e oelec ic phase. The speci ic hea anomaly a an-
si ion empe a u e shows[3–5] a s ong λ- ype anomaly;
pe haps, ha is why he ansi ion was ini ially consid-
e ed con inuous. Howe e Reese[4] showed ha he an-
si ion is discon inuous by measu ing i s la en hea , which
was e alua ed o be 46.1 J mol−1. I is shown ha dis-
con inui y disappea unde he in luence o p essu e o
elec ic ield.
The alue o he c i ical elec ic ield Ec o which he
discon inui y disappea s is a subjec o discussion. Reese
e al.[6] also ca ied ou a measu emen o he speci ic
hea wi h an applied ield o 294 V cm−1 inding a con o-
e sial e idence o la en hea . Abo e E= 785 V cm−1
hey ound no e idence o la en hea and sugges ed
Ec= 300 V cm−1 om he shi o he maximum o he
speci ic hea as unc ion o he elec ic ield. Measu e-
men s o speci ic hea unde elec ic ield highe han
360 V cm−1was ca ied ou by Sand old and Fossheim[7]
and he au ho s sugges ha Landau heo y o con inu-
ous ansi ions wi h 2 −4−6 po en ial was app op ia e
o desc ibe he shape o speci ic hea cu es.
O he wo ks ha e lead o es ima ions o he c i ical
ield. Fo ins ance, S uko e al.[8] e alua ed Ec=
124 V cm−1 om elec ocalo ic expe imen s. Sidnenko
and Gladki[9] ound Ec= 370 V cm−1while Okada and
Sugie[10] ob ained 160 V cm−1 o Ec. Vallade[11] de-
duced a alue o 254 V cm−1 om bi e ingence measu e-
men s. In con as o hese alues, Kobayashi e al.[12]
ound, by X- ay measu emen s, 8000 V cm−1; Ebe ha d
and Ho n[13] de i ed a alue o 6500 V cm−1 om dielec-
ic suscep ibili y.
In his ame i would be in e es ing o e alua e he
la en hea as a unc ion o he elec ic ield o de e mine
he alue o he c i ical elec ic ield Ec o which la en
hea becomes null.
The di icul y o an accu a e de e mina ion o Ec om
calo ime ic measu emen s in s anda d equipmen s —
such as di e en ial he mal analysis DTA and di e en ial
scanning calo ime e DSC —is ha hese sys ems eally
measu e changes o en halpy which has wo con ibu ions
nea he ansi ion: one due o he la en hea and o he
due o he a ia ion o speci ic hea wi h empe a u e.
In he case o phase ansi ions nea a ic i ical poin
o a discon inuous e oelec ic phase ansi ion unde
an elec ic ield close o he c i ical ield, he speci ic
hea p esen s a s ong anomaly and la en hea becomes
e y small. This ac makes di icul o sepa a e bo h
con ibu ions and o dis inguish he empe a u e in e -
al whe e he la en hea is p esen . This may explain
he lack o s udy o he in luence o elec ic ield, smalle
han Ec, on speci ic hea and la en hea o KDP a e
Reese[6].
Ou g oup has de eloped a me hod, named squa e
modula ed di e en ial he mal analysis SMDTA[14–16],
based on conduc ion calo ime y, which is able o mea-
su e absolu e alues o speci ic hea and he hea lux ex-
changed by he sample when i s empe a u e is changed
a a a e as low as 0.1 K h−1. The compa ison o he
da a allows us o sepa a e he abo e wo con ibu ions
o he o al en halpy and o e alua e he la en hea , in
case he e we e any.
Typese by REVT
EX
2
This echnique has been success ully applied o he
s udy o he almos ic i ical phase ansi ion o
KMnF3[14], whose la en hea was i s ly measu ed wi h
his echnique. Fu he mo e, he e ec o he subs i u-
ion o Mn by Ca was also in es iga ed measu ing he
la en hea [15], which showed ha he doping makes
he ansi ion become con inuous[17]. The me hod o
SMDTA has been also applied o show ha he phase
ansi ion in CoO[18], whose cha ac e was also con o-
e sial, is con inuous.
In his pape we ha e applied his me hod o s udy
a KDP single c ys al. We ha e measu ed he speci ic
hea and he hea lux exchanged by he sample in he
neighbo hood o he e oelec ic phase ansi ion. The
measu emen s ha e been ca ied ou a ou alues o E:
0 V cm−1, 100 V cm−1, 400 V cm−1, 1000 V cm−1.
Simul aneously o hese measu emen s, he dielec ic
suscep ibili y o he sample has also been measu ed and
we ha e ela ed i s beha iou a ound ansi ion empe a-
u e wi h he empe a u e in e al whe e he la en hea
is p oduced. Dielec ic measu emen s p o ide in o ma-
ion abou he mechanism o e oelec ic phase ansi-
ion o KDP c ys al[19, 20] and he simul aneous mea-
su emen o he mal and dielec ic p ope ies would be
wo hy.
II. EXPERIMENTAL
The measu emen s we e pe o med in a high esolu ion
conduc ion calo ime e which has been desc ibed p e i-
ously in de ails[21–23]. The senso is o med by wo
iden ical hea luxme e s, each one ha ing 48 ch omel-
cons an an he mocouples connec ed elec ically in se ies
bu he mally in pa allel. The sample is p essed be ween
bo h luxme e s whose signal is measu ed by a Kei hley
182 nano ol me e . Two elec odes and wo hea e s a e
placed be ween sample and luxme e s.
The senso is placed inside a calo ime e block which is
suspended wi hin wo cylind ical adia ion shields. The
whole assembly is hen placed in a he me ic ou e case
a a high acuum. The de ice is hen su ounded by a
coiled ube and placed in an alcohol ba h. Liquid N2ci -
cula es h ough he coil and egula es he empe a u e
o he ba h wi h a good he mal s abiliza ion. As a e-
sul , i is possible o change smoo hly he empe a u e
o he sample (a a a e o abou 0.1 K h−1) wi hou ob-
se ing signi ican empe a u e luc ua ions (always less
han 10−6K) in he block empe a u e.
The speci ic hea is measu ed using he me hod p e i-
ously desc ibed[16]. The same cons an powe Wis dissi-
pa ed in bo h hea e s (dissipa ion b anch) o wel e min-
u es and a s eady s a e cha ac e ized by a cons an em-
pe a u e di e ence be ween he sample and he calo ime-
e block is eached. The powe is hen cu o un il a
new s eady s a e is eached wel e minu es la e ( elax-
a ion b anch). Then, he powe is again swi ched on and
he sequence is con inuously epea ed while he empe -
a u e o he assembly is changed a a low cons an a e.
Tha is a long-pe iodic se ial o squa e he mal pulses
is supe posed o a hea ing o cooling amp. The em-
pe a u e inc ease o he sample due o he he mal pulse
o wel e minu es (ci ca 50 ×10−3K) is highe han he
empe a u e a ia ion o he sample p oduced by he a e
o change o empe a u e in hese wel e minu es (ci ca
±20 ×10−3K). Hence, he sample is being cooled and
hea ed al e na i ely du ing a un.
The in eg a ion o he elec omo i e o ce gi en by he
luxme e be ween e e y pai o s eady s a es allows us o
de e mine sample he mal capaci y. Hence, he me hod is
able o de e mine wo da a o hea capaci y in each cycle.
The i s one is calcula ed om he dissipa ion b anch
Cd, and he second one om he elaxa ion b anch C .
Hea capaci y ob ained in ei he b anch show a egula
beha iou i he e is no phase ansi ion o i i is con inu-
ous. When a discon inuous phase ansi ion occu s, bo h
da a become di e en showing an anomalous beha iou
in he empe a u e in e al whe e he la en hea is p o-
duced as a esul o he he mal hys e esis and ansi ion
kine ics. Tha beha iou is an e idence o he discon-
inuous cha ac e o he ansi ion[16]. Inciden ally, we
mus poin ou ha speci ic hea da a a e no eliable
when his beha iou is obse ed.
On he o he hand, he DTA ace is con inuously mea-
su ed in a second un wi hou dissipa ion in he sample
and using he same empe a u e scanning a e used o
measu e he speci ic hea . Due o he high numbe o
he mocouples and hei good he mal s abili y o he
sample, he equipmen wo ks like a e y sensi i e DTA
de ice. The elec omo i e o ce gi en by he luxme e s
is p opo ional o he hea lux, φdexchanged be ween
sample and calo ime e block.
F om he speci ic hea da a ob ained in he i s un-
ning and using a me hod p e iously desc ibed[14–16], we
calcula e he hea lux φcwhich would ha e been due ex-
clusi ely o he beha iou o he he mal capaci y o he
sample a ound he ansi ion empe a u e. Compa ing
he measu ed φdand he calcula ed φcwe deduce ha
only in he empe a u e ange (T , Tp) whe e bo h da a
do no coincide he e is an e ec om he la en hea .
I s alue is de e mined by in eg a ing φd/ , whe e is
he a e o empe a u e change, be ween (T , Tp) and us-
ing he s aigh line φd/ (T )−φd/ (Tp) as baseline[15].
The sensi i i y o he me hod is es ima ed o be be e
han 5 mJ.
The single c ys al o KH2PO4was g own a he In-
s i u e o Physics o Poznan Uni e si y (Poland). The
sample has 0.3857 g mass, wi h a hickness o 2.16 mm
along he e oelec ic axis and elec odes ci cula aces
wi h 78.5 mm2in su ace. The sample was placed ou in
he calo ime e . Gold elec odes we e e apo a ed o he
su ace o he sample; hose elec odes we e connec ed o
a capaci ance b idge ESI −SP5400 which has allowed
us o measu e he dielec ic pe mi i i y o he sample,
simul aneously o he hea lux, wi h an imposed ex e -
nal bias ield o 0 V cm−1, 100 V cm−1, 400 V cm−1and
3
1000 V cm−1.
III. RESULTS
III.1. Calo ime ic measu emen s
The empe a u e dependence o he speci ic hea cpo a
sample o KDP was measu ed on cooling, on quasies a ic
condi ions a a scanning empe a u e a e o ca. ∼
0.1 Kh−1using he me hod desc ibed in Sec. II o di -
e en applied elec ic ields. In Figu e 1, he speci ic hea
da a in a wide empe a u e in e al o di e en applied
elec ic ield (a) E= 0 V cm−1, (b) E= 100 V cm−1, (c)
E= 400 V cm−1and (d) E= 1000 V cm−1is shown.
These speci ic hea da a show a linea empe a u e de-
pendence in he pa aelec ic phase and o 0 V cm−1,
100 V cm−1and 400 V cm−1a sha p λ- ype anomaly
in a na ow empe a u e in e al is obse ed. Fo
1000 V cm−1 he maximum is mo e ounded and he
ansi ion is smea ed. The shape o speci ic hea cu e
o high elec ic ield ends o be almos symme ic, such
beha iou was also sugges ed by Reese[6]. The maxi-
mum alue o cini ially inc eases wi h inc easing ield
bu dec eases o su icien ly high alues. As expec ed,
he speci ic hea ail in he pa aelec ic phase inc eases
wi h ield as a consequence o he coupling o he o de
pa ame e o he ield.
In Figu e 2, we ha e plo ed he speci ic hea ex-
cess ob ained in he dissipa ion b anch (cd) and in he
elaxa ion b anch (c ) in a na ow empe a u e in e -
al, 1 K, o each elec ic ield. We obse ed ha o
0 V cm−1, 100 V cm−1and 400 V cm−1,cdand c da a
do no coincide a ound he ansi ion empe a u e while
o 1000 V cm−1bo h se ies o da a almos coincide in
he whole ange o empe a u e. As we ha e s a ed
abo e, he empe a u e a ia ion o he sample due o
he he mal pulses is sligh ly highe han he a ia ion
due o he empe a u e amp. This means ha in e -
e y pe iod he empe a u e o he sample inc eases and
dec eases consecu i ely. Due o he mal hys e esis, o ki-
ne ics o he phase ansi ion e c, he p ocess o hea ing
and cooling when wo phases coexis is di e en and con-
sequen ly da a ob ained in he dissipa ion b anch and e-
laxa ion b anch become di e en unde hese condi ions.
We ha e epo ed p e iously ha he di e ence is e y
no o ious e en in sys ems nea he ic i ical poin , whe e
he la en hea is e y small ( L = 0.13 J g−1 o KMnF3,
L = 0.010 J g−1 o KMn0.997Ca0.003F3)[16].
Hence, in he case o KDP, i is clea om Figu e 2 ha
o 0 V cm−1, 100 V cm−1and 400 V cm−1 he phase
ansi ion is discon inuous. On he con a y, he simi-
la beha iou o cdand c o 1000 V cm−1, Figu e 2d,
indica e ha no ace o la en hea is p esen , so we can
deduce ha o 1000 V cm−1 he phase ansi ion is con-
inuous. Hence he c i ical ield lies be ween 400 V cm−1
and 1000 V cm−1.
On he o he hand, he baseline espec which he
luxme e em is in eg a ed in o de o calcula e he spe-
ci ic hea is he unde line signal due o he empe a u e
amp imposed on he calo ime e . This signal changes
e y slowly and i does no a ec o he speci ic hea
measu emen excep when a la en hea e ec happens.
I p oduces a non linea a ia ion o he baseline and
consequen ly e oneous da a o speci ic hea may appea .
Keeping in mind ha he measu emen s ha e been ca -
ied ou in a cooling amp, in Figu e 2a he i s c a he
beginning o he ansi ion which does no coincide wi h
he co esponding cdda a is lowe han he egula con-
ibu ion o he speci ic hea . In Figu e 2b we can also
obse e such a simila poin bu his e ec is smea ed.
Finally, in Figu e 2c we canno obse e any dec ease in
c . Ne e heless, c de ia es om cdin a small ange
o empe a u e as in he wo p e ious igu es. Hence,
we can deduce ha la en hea dec eases wi h ield and
should be e y small o 400 V cm−1.
To con i m his sugges ion and o calcula e he la en
hea , we measu ed he DTA ace in a second un, chang-
ing he empe a u e o he sample a he same cons an
a e used in he speci ic hea measu emen s o make bo h
se s o da a compa able. We mus poin ou ha he
a e o change o empe a u e is abou wo o de s o
magni ude lowe han he minimum alue achie ed in
con en ional DTA equipmen s. In Figu e 3, we ep e-
sen he hea lux gi en by he luxme e s φd/ (DTA
ace) and he hea lux φc/ calcula ed om he speci ic
hea da a, using he me hod p e iously desc ibed, o he
ields 0 V cm−1(a), 100 V cm−1(b), 400 V cm−1(c).
Fo 0 V cm−1and 100 V cm−1φd/ is highe han
φc/ in a e y small empe a u e in e al o abou 0.05 K,
showing he e ec o la en hea . Fo 400 V cm−1φd/ is
also highe han φc/ , bu he di e ence be ween hem
is smalle , indica ing a e y small la en hea .
The choosing o he baseline o de e mine he la en
hea alue o 0 V cm−1in g aph Figu e 3a(i) is di i-
cul due o he e y small empe a u e in e al whe e
he la en hea is p esen and, consequen ly, he ew da a
eco ded inside ha in e al. Anyway, by conside ing as
baseline he s aigh line be ween he ex eme empe a-
u es whe e φd/ and φc/ coincide, we ob ain a alue o
43 J mol−1. This line is also ep esen ed in Figu e 3a).
The alue ob ained o KDP deu e a ed a 80% using he
same se -up and p ocedu e was 317 J mol−1[24].
To con i m he alidi y o ha baseline elec ion we
again measu ed he hea lux exchanged by he sample
in iden ical condi ions. Those hea lux da a a e also
plo ed in he Figu e 3a(ii). The igu e shows ha bo h
hea lux da a coincide wi h φc(s a s) in he same ange
o empe a u e hus suppo ing he baseline used o he
de e mina ion o he la en hea . In ac , he la en hea
o he second un is 45 J mol−1in good ag eemen wi h
ha o he i s one. On he o he hand, hese alues
o he la en hea a e in ag eemen wi h Reese[6]. I is
no ewo hy ha peak a ea —i.e. la en hea — is qui e
ep oducible despi e he kine ic e ec s ha showed each
expe imen . On he con a y, he di e ence obse ed i
4
119 120 121 122 123 124 125
0
200
400
600
800
(b)
c/J K−1mol−1
T/K
119 120 121 122 123 124 125
0
200
400
600
800
(d)
c/J K−1mol−1
T/K
0
200
400
600
800
119 120 121 122 123 124 125
(a)
c/J K−1mol−1
T/K
0
200
400
600
800
119 120 121 122 123 124 125
(c)
c/J K−1mol−1
T/K
FIG. 1. The speci ic hea o KDP o di e en applied elec ic ields in a wide empe a u e in e al. F om le o igh and op
o bo om, (a) 0 V cm−1and (b) 100 V cm−1, (c) 400 V cm−1and (d) 1000 V cm−1.
φ/ du ing he phase ansi ions sugges s he exis ence
o non-equilib ium kine ic p ocess as hose expec ed o
a discon inuous phase ansi ion — o ins ance phase
on gene a ion— which, as a gene al ac , a e non e-
p oducible.
Fo E= 100 V cm−1and E= 400 V cm−1we de-
c eased he scanning a e o empe a u e when measu ing
speci ic hea so as o ge a highe numbe o da a poin s
in he neighbo hood o he phase ansi ion. Hence, he
de e mina ion o he baseline becomes easie as shown in
Figu e 3. The in eg a ion o he peaks gi es 35 J mol−1
and 4.2 J mol−1 espec i ely.
The ob ained esul con i m ha la en hea dimin-
ishes wi h elec ic ield (see Figu e 4). A 400 V cm−1
la en hea is educed by and o de o magni ude wi h
espec o ha o ze o ield and i is close o he c i i-
cal ield. Assuming a linea beha iou o he la en hea
wi h he ield we deduce ha he c i ical ield lies on
5
121 121.2 121.4 121.6 121.8 0
200
400
600
800
(b)
c/J K−1mol−1
T/K
120.8 121 121.2 121.4 121.6 121.8
0
200
400
600
800
(d)
c/J K−1mol−1
T/K
0
200
400
600
800
120.6 120.8 121 121.2 121.4 121.6
(a)
c/J K−1mol−1
T/K
0
200
400
600
800
120.8 121 121.2 121.4 121.6 121.8
(c)
c/J K−1mol−1
T/K
FIG. 2. The same as igu e 1 bu in a na owe empe a u e in e al a ound he ansi ion poin . Bold poin s s and o he
dissipa ion b anch (cd), open poin s s and o he elaxa ion b anch (c ).
(0.44 ±0.3) kV cm−1.
III.2. Dielec ic measu emen s
In o de o ela e calo ime ic and dielec ic beha iou
he dielec ic pe mi i i y along he e oelec ic axis has
been measu ed simul aneously o he hea lux.
In Figu e 5 we ep esen ε(T) o he di e en elec-
ic ields. The dielec ic pe mi i i y inc eases in all
cases du ing he phase ansi ion ollowing he Cu ie law.
In he e oelec ic phase he dielec ic pe mi i i y e-
mains in a pla eau, indica ing also he la ge domain wall
con ibu ion o he dielec ic pe mi i i y, bu he max-
imum o pe mi i i y dec eases wi h he applied elec ic
ield. I has been p e iously epo ed ha he maxi-
mum alue o dielec ic pe mi i i y is due o he con-
ibu ion o domain wall [25]; unde elec ic ield he

6
8
4
0
121.5121.4121.3
(a)
(i) (ii)
φ −1/J K−1
T/K
121.5121.4121.3
(b)
T/K
121.6121.5121.4121.3
8
4
0
(c)
φ −1/J K−1
T/K
FIG. 3. Hea lux di ided by a e o change o empe a u e in he neighbo hood o he KDP e oelec ic phase ansi ion.
F om le o igh , op o bo om, (a) 0 V cm−1, (b) 100 V cm−1, (c) 400 V cm−1. Poin s ep esen hea lux da a gi en by he
luxme e s —in (a) wo simila unning labelled (i) and (ii) we e ca ied ou —. S a s ep esen he con ibu ion due o speci ic
hea o Figu e 2. S aigh lines show he baseline used o de e mina ion o he la en hea which happens o be he peak a ea
sub ended by expe imen al poin s and he baseline.
sample becomes close o monodomain s a e, so num-
be o domain walls dec ease and also does he dielec ic
pe mi i i y. Ne e heless he beha iou a ansi ion
empe a u e is di e en o 0 V cm−1and 100 V cm−1
—discon inuous ansi ion— han o 400 V cm−1—
close- o-c i ical-poin ansi ion— and 1000 V cm−1—
con inuous ansi ion—. We will ela e in Figu e 6
he shape o dielec ic pe mi i i y o 0 V cm−1and
400 V cm−1.
The hea lux and he in e se o he dielec ic pe mi -
i i y e sus empe a u e o E= 0 V cm−1a e ep e-
sen ed in Figu e 6(a). Th ee egions may be dis inguish
in his igu e: (i) he pa aelec ic phase whe e he pe mi -
i i y ollows he Cu ie law, (ii) he phase ans o ma ion
in e al whe e he pe mi i i y sligh ly de ia es om he
p e ious beha iou and (iii) he e oelec ic phase whe e
he pe mi i i y shows a pla eau. We mus poin ou ha
he maximum o he pe mi i i y ma ches wi h he end
o phase ans o ma ion.
Fo 400 V cm−1,ε(T) shows a di e en beha iou —
see Figu e 6(b)—. A he ansi ion empe a u e i
is p esen a minimum. This has been obse ed by
Bo na el[19] in c ys als o KDP o highe alues o elec-
ic ield (abou 1 kV cm−1). In ha wo k Bo na el ex-
plains his beha iou in e ms o some domains a ange-
men s du ing he phase ansi ion; ε(T) is desc ibed as
he sum o he beha iou o ε1(T) con ibu ion ha co -
espond o he beha iou o monodomain sample, and he
co esponding ε2(T) due o he con ibu ions o domains.
Mo eo e ε1(T) inc eases om pa aelec ic phase and a -
e he maximum a ansi ion empe a u e dec eases o
ze o in e oelec ic phase; ε2(T) inc eases om ze o a
ansi ion empe a u e and emains in a pla eau in e o-
elec ic phase. The sum o bo h con ibu ions gi es he
appea ance o ε(T) in igu e 6b.
I may be seen also om he esul s p esen ed on Fig-
u e 6 ha he peak o he hea lux a ansi ion em-
pe a u e appea a empe a u e whe e he p oduc ion o
domains become dominan .
IV. CONCLUSIONS
The he mal and dielec ic beha iou o KDP c ys-
al nea he empe a u e o i s e o-pa aelec ic phase
ansi ion has been simul aneously s udied unde he in-
luence o elec ic ield. Fo 0 V cm−1 he la en hea
has been measu ed wice. Al hough bo h measu emen s
shows a di e en kine ic, hei esul s a e simila and in
good ag eemen wi h hose ob ained by Reese[6]. Despi e
he high inc ease o he speci ic hea a ound he an-
si ion empe a u e, he e y small alues o he la en
hea and he na ow empe a u e ange —ca. 0.1 K—
whe e ansi ion is de eloped, we ha e been able o dis-
c imina e he con ibu ion o he la en hea o he o al
change o en halpy. Finally, simul aneous measu emen
o hea lux and dielec ic suscep ibili y sugges ha he
e ec o la en hea appea s a empe a u es ange whe e
7
0
20
40
60
0 200 400
E/V cm−1
∆h/J mol−1
FIG. 4. Plo o he la en hea ∆ha a ious elec ic ields. C i ical ield is ound o be (0.44 ±0.03 kV cm−1
0
5000
10000
15000
120 121 122 123 124
T/K
ε
0 V cm−1
100 V cm−1
400 V cm−1
1000 V cm−1
FIG. 5. Plo o he dielec ic pe mi i i y a a ious elec ic
ields. Pe mi i i y da a we e measu ed simul aneously o
hea lux da a o igu e 3.
domains a e g owing. Maximum dielec ic pe mi i i y
was no iced a empe a u e whe e phase ansi ion ends.
I was es ablished ha he la en hea dec eases wi h
he ield and he c i ical elec ic ield is es ima ed o be
(0.44 ±0.03) kV cm−1.
4
2
0
121.5121.4121.3
0.0002
0.0001
0
(a)
φ −1/J K−1
T/K
1/ε
0
2
4
121.4 121.6 121.8
4000
6000
8000
(b)
φ −1/J K−1
T/K
ε
FIG. 6. (a) Plo o he in e se o he dielec ic pe mi i i y
( igh axis) and φ −1(le axis) o E= 0 V cm−1. (b) Plo
o he dielec ic pe mi i i y ( igh axis) and φ −1(le axis)
E= 400 V cm−1.
ACKNOWLEDGMENTS
We wish o hank P o . Bo na el o ui ul discus-
sions. This wo k was suppo ed by Spanish Minis e io de
Ciencia y Tecnolog´ıa con ac numbe BFM2002-02237.
[1] Y. Xu, Fe oelec ics ma e ials and hei applica ions
(No h-Holland, 1991), ISBN 0-444-88354-1.
[2] M. Lines and A. Glass, P inciples and applica ions o e -
oelec ics and ela ed ma e ials (Cla endon P ess, Ox-
8
o d, 1977).
[3] C. S ephenson, J. Phys. Chem. 66, 1397 (1944).
[4] W. Reese, Physical Re iew 162, 510 (1967).
[5] B. S uko , M. Amin, and V. Kopchik, Physics S a us
Solidi 27, 741 (1969).
[6] W. Reese, Physical Re iew 181, 905 (1969).
[7] E. Sand old and K. Fossheim, J. Phys. C: Solid S a e
Phys. 19, 1481 (1986).
[8] B. S uko , M. Ko zhue , A. Baddu , and V. Kop sik,
So . Phys. Solid S a e 13, 1569 (1972).
[9] E. Sidnenko and V. Gladkii, So . Phys. C ys allg . 18,
83 (1973).
[10] K. Okada and H. Sugie, Fe oelec ics 17, 325 (1977).
[11] M. Vallade, Physical Re iew 12, 3755 (1975).
[12] J. Kobayashi, Y. Uesu, and Y. Enomo o, Phys. S a us
Solidi (b) 45, 293 (1971).
[13] J. Ebe ha and P. Ho n, Solid S a e Communica ions 16,
1343 (1975).
[14] J. del Ce o, F. Rome o, M. C. Galla do, S. Haywa d,
and J. Jim´enez, The mochimica Ac a 343 (2000).
[15] F. Rome o, M. C. Galla do, J. Jim´enez, and J. del Ce o,
The mochimica Ac a 372, 25 (2001).
[16] J. del Ce o, J. M. Ma ´ın-Olalla, and F. Rome o, The -
mochimica Ac a 401, 149 (2003).
[17] M. C. Galla do, F. Rome o, S. Haywa d, E. Salje, and
J. del Ce o, Mine alogical Magazine 64, 971 (2000).
[18] F. Rome o, J. Jim´enez, and J. del Ce o, Jou nal o Mag-
ne ic Ma e ials (in p ess).
[19] J. Bo na el, Fe oelec ics 54, 245 (1984).
[20] E. Nakamu a, Fe oelec ics 135, 237 (1992).
[21] J. del Ce o, Jou nal o The mal Analysis 34, 335 (1988).
[22] F. Jimenez, S. Ramos, and J. del Ce o, Phase T ansi-
ions 12, 275 (1988).
[23] J.-M. Ma ´ın, J. del Ce o, and S. Ramos, Phase T ansi-
ions 64, 45 (1997).
[24] M. C. Galla do, J. Jim´enez, M. Ko alewski, and J. del
Ce o, J. App. Phys. 81, 2584 (1997).
[25] J. Bo na el and R. Cach, Phys. Re . B 60, 3806 (1999).