156
K. Adrjanowicz and R. Richert
two parallel invar-steel plates of 18 mm diameter and fixed distance of 79 μm was
used [27].
High static electric fields were applied using a Trek PZD-700 high-voltage amplifier which boosts the generator voltage of the Solartron SI-1260 by a factor of 200. The
current was detected via the voltage drop across an RC-shunt, which is connected to
the analyzer via a buffer amplifier and a dc-blocking capacitor with C 0 = 220 nF. The
buffer amplifier is meant to protect the analyzer from sample failure, as it tolerates
voltages up to 500 V p without time limit. With this system, the dielectric response
of the sample can be measured for frequencies between 1 Hz and 100 kHz. The
capacitor cell used for high-field experiments consists of two spring-loaded polished
stainless-steel disks (17 and 20 mm diameter), separated by a Teflon ring of 25 μm
thickness. The Teflon ring leaves an inner area of 14 mm diameter which is filled
with the sample.
The nominal electrode separation of the high-field cell is defined by the 25 μm
thickness of the Teflon spacer, but the actual electrode distance could be different.
Uncertainties in the electrode separation d will translate into uncertainties in the
magnitude of the dc bias field E = V /d. The actual electrode separation d can be
determined from the ratio of ε
(at ν = 10 kHz) measured in the high-field cell and
invar-steel cell, the latter serving as reference value because its geometry is well
defined. This approach provides electric fields with an accuracy of about ±5%. The
kinetics of crystallization was monitored on the basis of changes in the dielectric
response of the sample at a fixed frequency of ν = 10 kHz, with or without the
presence of a dc bias field, in time steps of 5 seconds. The amplitude (RMS) of the
ac measuring field was set to 20% of the applied dc bias field in each high-field
experiment. Therefore, all ac field amplitudes result in responses within the linear
regime. Further details can be found elsewhere[16].
A variety of protocols have been employed regarding the change of temperature
versus time and the range in which a high dc electric field has been applied. The
different T (t) and E(t) schemes are identified schematically in Fig. 5, and are labeled
A through D for reference used below.
All temperature profiles begin and end above the melting point T m , and after the
final melting, all field-induced changes turned out fully reversible. In these graphs,
T c refers to “crystallization temperature”, i.e., the temperature at which the progress
of crystallization is monitored as a function of time under isothermal conditions.
2.1 Results and Discussion
The crystallization behavior of VEC at zero electric field is illustrated in Fig. 6.
After melting, the permittivity was recorded during cooling at a rate of −2 K min
−1 ,
and then during subsequent heating at +0.3 K min
−1 . The cooling scan reveals no
crystallization, as the level of ε
(ν = 10 kHz) follows the expected ε s ∝ 1/T behavior
for T > 195 K, while the decrease at lower temperature is the result of the relaxation
time, τ α , exceeding the reciprocal test frequency of 10 kHz.
K. Adrjanowicz and R. Richert
two parallel invar-steel plates of 18 mm diameter and fixed distance of 79 μm was
used [27].
High static electric fields were applied using a Trek PZD-700 high-voltage amplifier which boosts the generator voltage of the Solartron SI-1260 by a factor of 200. The
current was detected via the voltage drop across an RC-shunt, which is connected to
the analyzer via a buffer amplifier and a dc-blocking capacitor with C 0 = 220 nF. The
buffer amplifier is meant to protect the analyzer from sample failure, as it tolerates
voltages up to 500 V p without time limit. With this system, the dielectric response
of the sample can be measured for frequencies between 1 Hz and 100 kHz. The
capacitor cell used for high-field experiments consists of two spring-loaded polished
stainless-steel disks (17 and 20 mm diameter), separated by a Teflon ring of 25 μm
thickness. The Teflon ring leaves an inner area of 14 mm diameter which is filled
with the sample.
The nominal electrode separation of the high-field cell is defined by the 25 μm
thickness of the Teflon spacer, but the actual electrode distance could be different.
Uncertainties in the electrode separation d will translate into uncertainties in the
magnitude of the dc bias field E = V /d. The actual electrode separation d can be
determined from the ratio of ε
(at ν = 10 kHz) measured in the high-field cell and
invar-steel cell, the latter serving as reference value because its geometry is well
defined. This approach provides electric fields with an accuracy of about ±5%. The
kinetics of crystallization was monitored on the basis of changes in the dielectric
response of the sample at a fixed frequency of ν = 10 kHz, with or without the
presence of a dc bias field, in time steps of 5 seconds. The amplitude (RMS) of the
ac measuring field was set to 20% of the applied dc bias field in each high-field
experiment. Therefore, all ac field amplitudes result in responses within the linear
regime. Further details can be found elsewhere[16].
A variety of protocols have been employed regarding the change of temperature
versus time and the range in which a high dc electric field has been applied. The
different T (t) and E(t) schemes are identified schematically in Fig. 5, and are labeled
A through D for reference used below.
All temperature profiles begin and end above the melting point T m , and after the
final melting, all field-induced changes turned out fully reversible. In these graphs,
T c refers to “crystallization temperature”, i.e., the temperature at which the progress
of crystallization is monitored as a function of time under isothermal conditions.
2.1 Results and Discussion
The crystallization behavior of VEC at zero electric field is illustrated in Fig. 6.
After melting, the permittivity was recorded during cooling at a rate of −2 K min
−1 ,
and then during subsequent heating at +0.3 K min
−1 . The cooling scan reveals no
crystallization, as the level of ε
(ν = 10 kHz) follows the expected ε s ∝ 1/T behavior
for T > 195 K, while the decrease at lower temperature is the result of the relaxation
time, τ α , exceeding the reciprocal test frequency of 10 kHz.