Mol. Cryst. Liq. Cryst. 1985. Vol. 119. pp. 329-335
© 1985 Gordon and Breach. Science Publishers. Inc. and OPA Ltd.
Printed in the United States of America
Walther-Meissner-Institut fur Tieftemperaturforschung,
D-8046 Garching, FRG
Max-Planck-Institut fur Medizinische Forschung, Abteilung fur
Molekulare Physik, JahnstraBe 29, D-6900 Heidelberg, FRG
Anorganisch-Chemisches Institut der Universitat Heidelberg,
Im Neuenheimer Feld 270, D-6900 Heidelberg, FRG
The a-modification of (BEDT-TTF)213 undergoes a metal-to-insulator
transition at T
135 K. Application of hydrostatic pressure shifts
the transition temperature to lower values. Above 12 kbar semimetallic behavior is observed, but no superconductivity is detected
up to a pressure of 17 kbar and temperatures down to 100 mK. The
S-modifcation, on the other hand, exhibits ambient pressure volume
superconductivity below Tc
1.05 K. Measurements of the anisotro-
py of the diamagnetic shielding- as well as the Meissner-magnetization are presented.
The first discovery of organic superconductivi ty (~) in the family of
Bechgaard salts (TMTSF)2X /1,2/ has successfully stimulated the
search for new organic materials that exhibit this still fascinating solid state phenomenon at lew temperatures. Parkin et al. found
the sulfur based compound (BEDT-TTF)2Re04 (bis-(ethylenedithiolylenetetraehiafulvalene)-perrhenate) to be superconducting below
Tc ~ 2 K and above a pressure of ~4 kbar /3,4/. Recently, the interest has focussed strongly on compounds based on the same donor
(BEDT-TTF) with the triiodide (I3) molecule as counter ion. Very
recently, Yagubskii et al. have repQrted three of its stoichiometries to be zero pressure SC's with Tc = 1.4 - 1.5 K /5/ and
Tc = 2.5 K /6,7/. So far ambient pressure SC has been observed
only in (TMTSF)2C104 and it could be demonstrated there that SC is
a bulk property which means that all conduction electrons condense
into the SC state /8,9/.
The measurement of the electrical resisti vi ty is a common tool in
the search for SC, but a drop to a state wi th zero resisti vi ty is not
necessarily an indication of volume SC, as only one complete SC path
through the sample can yield zero resistance and thus mask the information about the state of the remaining sample volume. On the other hand,
the diamagnetic response of a SC to small magnetic fields (especially
the Meissner effect, Le. the flux expulsion from the SCI yields information about the volume that is superconducting. The compound
(TMTSF) 2FS03 may serveasai1 example of such a case. Superconductivity
has been detected by means of resistivity measurements above a pressure of'" 5 kbar and below Tc'" 2 K /10,11/;The diamagnetic measurarents
yield an upper limit of only 2 'is of the volume that is SC below a transition temperature of 1.0 K /12/.
The main purpose of this paper is to demonstrate by low field
magnetization experiments that the zero resistance state of S(BEDT-TTF) 213 (found by Yagubskii et al. below'" 1 K wi th a start of
the resistive transition at1.5~ is in fact a state of bulk volume
Crystals of the triiodide compound in the 2:1 stoichiometry
grow in two different structural modifications. The one that we
have investigated first, hereafter denoted as the ~-modification:
~- (BEDT-TTF) 213' undergoes a metal - to-insulator transition at
TMI '" 135 K /13, 14/. In this paper we report on resisti vi ty measurements of this modification under pressure in order to look for the reduction of TMI and the possible occurence of SC. Ambient pressure SC
-as mentioned before- is observed in the other modification - hereafter denoted by S-(BEDT-TTF)213' The higher Tc phases are of different stOichiometries, e.g. (BEDT-TTF) 3 (13) 2 (13)0.5 with Tc = 2.5 K
and Pc -= 0 /7/.
Crystals of the ~- and S-modification of (BEDT-TTF) 213 were prepared using standard electrochemical techniques, details of which aIE
described in /14,15/. Both modifications grew simultaneously at the
anode. In the phase diagram of (BEDT-TTF) and (13) the minima of
the thermodynamical potentials of the various stoichiometries and
modifications /7,15/ seem to be very close to each other. Actually
at the present state of the art of electrochemistry it is almost
impossible to control the parameters of the electrocrystallization
in a·way to reproducibly grow only one or the other particular
stoichiometry in a particular structural modification.
The crystals of a-(BEDT-TTF)2I3 that we used for the resistivity measurements were black thin plates of typical dimensions
(3x3xO.02 rom 3 ), where the plane of the plates is the a-b plane.
Crystals of the S-modifications, grown simultaneously, could be
separated from the ~-phase under a microscope, since they grow as
black canted rhombohedrons with distorted hexagonal cross section
(-the a-b plane). They are somewhat smaller than a-phase-crystals.
A typical crystal that has been used in the low field magnetization
experiment is shown in the insert of fig. 3 (dimensions: 0.74 mm
lIa, 0.58 mm-l-a, 0.22 mm IIc*). For the calculation of the demagnetization coefficients /16/ the shape of the sample was assumed to
be ellipsoidal (with the sample dimensions as axes of the ellipsoid) .
The electrical resistivity of a-crystals was measured by the
usual 4 probe ac-method (30 Hz). Copper leads (0.017 mm ¢) were
attached to the sample with silver paint on previously evaporated
gold pads, yielding contact resistances of a few ohms. The electrical current was always flowing in the a-b plane. Pressure measurements were made in a BeCu-pressure clamp using a 50 : 50 mixture
of pentane and isobutane as pressure transmitting medium. The
pressure was measured at low temperatures by observing the shift
of the SC transition of a Sn-sample.
Our first observation of superconductivity in B-(BEDT-TTF)2I3
crystals came from a measurement of the ac-susceptibility. Thiswas
done in a tunnel diode (GE BD-4) oscillator circuit operated at 42 kc
in a 3He cryostat. The principal experimental set up for the low
field dc-magnetization measurements in a dilution cryostat is described in /17/ and the improved version of the apparatus that
enabled us to rotate the sample in the magnetic field while it
remains at low temperature is discussed in ref. /18/.
1) a-(BEDT-TTF)2I3
The results of the temperature dependence of the electrical resistance at various pressures are shown in fig. 1. The cooling speed
between rClomtemperature and 4.2 K was 30 - 60 K/h. The temperature
of the metal-to-insulator transition TMI, as defined by the maximum of -dlnR/dT, is plotted vs. pressure in fig. 2. At zero press~
re TMI = 135 K, as has been observed before /13,14/. Increasing
hydrostatic pressure shifts TMI to lower values (at an almost
linear rate: dTMI/dp ~-11 K/kbar). For p > 12 kbar and T > 1.3 K
a phase transition temperature can no longer be inferred from the
data. The resistivity then does not show activated behavior any
more below the temperature of the minimum (which is indicated by
the squares in fig. 2). We have cooled one sample in the pressure
bomb at 13 kbar in a dilution refrigerator down to 100 mK and did
not detect any indications of an onset of superconductivity.
The crucial question as to what causes the metal-insulator
transition in the a-phase, remains so far unanswered. We have some
indications that it is a structural transition, which however does
not take place at the sites of the (BEDT-TTF) molecules. One might
speculate that the (1 3 ) molecules distort and thereby loOse their
inversion symmetry thus opening the possibility of orientational
ordering. Detailed structural investigations to clarify the nature
of this transition are in progress.
00 kbar
02 kbar
A 75kbar
09 kbar
v12 kbar
a-b plane resistivity vs. temperature of
at different pressures.
2) 8-(BEDT-TTF)213
Samples of this compound have been cooled from roomtemperature to
80 K in 5 - 6 h, and further on to 4.2 K in 1 - 1.5 h. Measurem~s
of the ac-susceptibility clearly show evidence of an onset of diamagnetic shielding currents below 1.05 K /19/. In the dc-magnetization experiment one has to distinguish between two kind of transition curves (fig. 3). When the sample is cooled in zero field
well below Tc and the dc-field is then applied, supercurrents on
the sample surface (within a layer of the thickness of the penetration depth) are induced that screen the magnetic field from
the inside of the sample. Upon warming it up, the decay of the SC
screening currents is monitored by the magnetization change: the
diamagnetic shielding signal. Leaving the magnetic field on and
cooling back the sample leads to the formation of quantized flux
Phase diagram of a-(BEDT-TTF) 213' The dash-dotted line
corresponds to dTM1/dp = -11 K/kbar. The squares indicate the
temperature of the resistance minima in the pressure region above
12 kbar where semimetallic behavior is predomiant.
lines in the sample - that is a departure from the uniform flux
distribution in the normal state - and their subsequent expulsion
out of the sample: the Meissner effect. Pinning of flux lines at
impurity sites or crystal defects leads to a smaller Meissner magnetization than the diamagnetic shielding magnetizat~on. The largest diamagnetic shielding- and Meissner signals (62 resp 18 % of
that of a perfect se, at T = 0.115 K) are observed when the field
is oriented normal to the highly conducting a-b plane (H lic*),
fig. 3. The transition is broad and not saturated at lowest temperatures, an indication of a considerable spread of Tc over different regions in the sample. A linear extrapolation of the magnetization to zero yields Tc = 1.05 K. From the magnetization curveone
can deduce the field when the first flux ,starts penetrating into
the sample: Hc 1. This field is biggest in the orientation H J. a~b
plane: HC1c* is equal to 0.36 Oe and becomes even smaller when the
field is turned into the a-b plane: Hcla = 0.05 Oe and Hc lla=O. 09 Oe.
The anisotropy in the a-b plane is not very pronounced,corroborating the two dimensional character of this compound.
Our main finding, in conclusion, is the existence of rather
complete dc-shielding supercurrents that demonstrate the existence
of closed supercurrent paths on the sample surface which are able
Diamag. Shielding
~ Signal
Happl. =0.12 Oe
T (K)
Diamagnetic shielding and Meissner susceptibility vs. temp~
rature of S-(BEDT-TTF)2I3 in a field applied perpendicular to the
a-b plane. The insert relates the morphology of the investigated
crystal to its axes.
to keep the bulk of the sample field free. The proof for bulk SC
is the ability of the crystal to expell the magnetic flux when
cooled in a field. Taking into account flux pinning effects - that
are usually observed in bulk type II SC, like (TMTSF)2C104 /20/ the existence of a considerable Meissner effect in 8-(BEDT-TTF)2I3
shows that SC in this compound, though anisotropic, is a bulk
This work was supported by the Stiftung Volkswagenwerk in Hannover.
/1/ For a review see, Proc. Intern. Conf. on Low-Dimensional
Conductors (Boulder, Colorado, August 1981),
Mol. Cryst. Liq. Cryst. 72., (1982)
/2/ A further review, Proc. of the Intern. CNRS Colloquium on the
Physics and Chemistry of Synthetic and Organic Metals (Les
Arcs, December 1982) J. Physique 44, Colloque C3 (1983'
/3/ S.S.P. Parkin, E.M. Engler, R.R. Schumaker, R. Lagier,
V.Y. Lee, J.C. Scott and R.L. Greene, Phys. Rev. Lett. 50,
270 (1983)
/4/ S.S.P. Parkin, E.M. Engler, R.R. Schumaker, R. Lagier,
V.Y. Lee, J. Voiron, K. Carneiro, J. C. Scott and R.L. Greene,
J. Physique ~, Colloque C3, 791 (1983)
/5/ E.B. Yagubskii, I.F. Shchegolev, V.N. Laukhin, P.A. Kononovich
M.V. Karatsovnik, A.V. Zvarykina and L.J. Buravov, JETP Lett.
~, 12 (1984)
/6/ E.B. Yagubskii, I.F. Shchegolev, S.I. Pesotski, V.N. Laukhin,
P.A. Kononovich. M.B. Kartsovnik and A.V. Zvarykin~, JETP LetL
~, 275 (1984)
/7/ R.P. Shibaeva, this proceedings
/8/ H. Schwenk, K. Neumaier, K. Andres, F. Wudl and E. AharonShalom, Mol. Cryst. Liq. Cryst. 72., 277 (1982)
/9/ P. Garoche, R. Brusetti, D. Jerome and K. Bechgaard,
J. Physique Lett. ~, L-147 (1982)
/10/ R.C. Lacoe. S.A. Wolf, P.M. Chaikin, F. Wudl and E. AharonShalom, Phys. Rev. B27, 1947 (1983)
/11/ R.C. Lacoe, P.M. Chaikin, F. Wudl and E. Aharon-Shalom,
J. Physique 44, Colloque C3, 767 (1983)
/12/ F. Gross, H. Schwenk, K. Andres, F. Wudl, S.D. Cox and
J. Brennan, Phys. Rev. B in press, and this proceedings
/13/ K. Bender, K. Dietz, H. Endres, H.W. Helberg, I. Henning,
H.J. Keller, H.W. Schafer and D. Schweitzer, Mol. Cryst. Liq.
Cryst. 107, 45 (1984)
/14/ K. Bender, I. Hennig, D. Schweitzer, K. Dietz, H. Endres and
H.J. Keller, Mol. Cryst. Liq. Cryst., in press
/15/ I. Hennig, K. Bender, D. Schweitzer, K. Dietz, H. Endres,
H.J. Keller, H.W. Schafer and H.W. Helberg, this proceedings
/16/ R.A. Klemm, J. Low Temp. Phys. ~, 589 (1980)
/17/ K. Andres, F. Wudl, D.B. McWhan, G.A. Thomas, D. Nalewajek and
A.L. Stevens, Phys. Rev. Lett. 45, 1449 (1980)
/18/ K. Andres, H. Schwenk and F. Wudl, Helv. Phys. Acta ~, 675
/19/ see also,
J.M. Williams, this proceedings
/20/ H. Schwenk, K. Andres and F. Wudl, Solid State Commun. 49,
723 (1984)