Physica B 165&166 (1990) 1377-1378
North-Holland
RESISTANCE NOISE AT THE SUPERCONDUCTING TRANSITION IN THICK FILMS OF
YBa2CuS07-o'
J. M. Aponte, G. Gutierrez and M. Octavio
Centro de Ffsica. Instituto Venezolano de Investigaciones Cientlficas (I.V.I.C.). Apartado 21827.
Caracas 1020A. Venezuela.
Measurements of the power spectral density of the resistance noise were performed in thick films of
with broad superconducting tr:r,nsitions and in the frequency range from 0.1 Hz to
100rfz. The normalized power spectrum (S IV ) of the excess noise depends on frequency as 1/f
a
,
with a=1 over a wide range of temperature. We have also found that ft the transition, the magnitude of
this excess noise shows peaks at the same temperatures at which R- (dR/dT) has peaks and also very
close to the temperature at which the sample reaches its R=O state. Possible sources of noise
responsible for these effects are proposed.
Experiments on the resistance noise of
copper-oxide based superconductors show that the
excess noise, which exhibit a 1/f-type spectrum, has
a high level, in both bulk samples(1) and thin
films(2,S), as compared to the noise in classical
superconductors. These results have implications
for the practical applications of these ceramics. On
the other hand it is of interest to understand the
origin of such excess noise. All the results found in
the literature agree that the excess noise is zero in
the superconducting state and that it rises sharply
at the transition. This behavior is not yet well
undestood. In this paper we study the power
spectrum of the resistance noise near the
superconducting transition of thick films of
YBa2Cu307_o with broad resistive transitions. The
transition is approached by either changing the
temperature at a constant current or by changing
the current at a constant temperature. Our films are
about 1 thick and were prepared on various
types of substrates: Zr(Ca)02' alumina and
sapphire. The details of the fabrication of our
samples have been reported elsewhere(4). In
order to measure the resistance noise we use a
four-probe configuration. A d.c. current is applied to
the sample and the signal between the voltage
contacts is coupled through a transformer into a
PAR113 amplifier in series with a home-made
low-pass filter (cut-off frequency of 1kHz), the output
of which is measured by a spectrum analyzer
model HP3582A controlled by a microcomputer.
The background noise is measured and subtracted
in order to obtain the current-dependent excess
noise. The sample is attached to a copper block
inside a brass can which is inmersed in a liquid
nitrogen or liquid helium bath. The temperature can
be varied by pumping on the bath and/or applying
current to a heater, controlled by a temperature
controller.
The power spectral density Sv of our samples
exhibit 1/f
a
behavior, with a=1, except near
Tconset, the temperature at which the resistance
starts to drop, there we fou nd that a> 1. Figure 1
shows the magnitude of SvIV2 at f=10Hz,
normalized to its value at room temperature, as a
function of temperature for one of our samples with
room temperature resistance of 7.2 n. We observe
two peaks at the same temperatures at which
W
1
dR/dT has peaks, also shown in the same
figure. Similar behavior was also observed in other
samples. This result suggests that the temperature
fluctuation model(5) can in principle explain the
origin of this noise. However, according to this
model, the power spectrum should depend
quadratically on W
1
dR/dT and we observe that
SvIV2 oc [W
1
dRldTj'Y with y<1 as it can be seen in
0921-4526/90/$03.50 © 1990 - Elsevier Science Publishers B.V. (North-Holland)