HR: 0830h
AN: GP11A-0253 [PDF]
TI: Study of dielectric dispersion in moist rocks at low frequency as a possible clue for sounding of
H$_{2}$0 in the Earth's crust
AU: * Matsumoto, H
EM: matsumo@tono.jnc.go.jp
AF: Tono Geoscience Center, Japan Nuclear Cycle Development Institute, 1-63 Yamanouchi, Akeyocyo,
Mizunami, Gif 509-6132
Japan
AU: Shigeta, N
EM: taka@tono.jnc.go.jp
AF: Tono Geoscience Center, Japan Nuclear Cycle Development Institute, 1-63 Yamanouchi, Akeyocyo,
Mizunami, Gif 509-6132
Japan
AU: Kumazawa, M
EM: mkz@tono.jnc.go.jp
AF: Tono Geoscience Center, Japan Nuclear Cycle Development Institute, 1-63 Yamanouchi, Akeyocyo,
Mizunami, Gif 509-6132
Japan
AU: Nakajima, T
EM: takahiro@tono.jnc.go.jp
AF: Tono Geoscience Center, Japan Nuclear Cycle Development Institute, 1-63 Yamanouchi, Akeyocyo,
Mizunami, Gif 509-6132
Japan
AB:
This work is motivated to examine the possibility of remote observation of H$_{2}$O and its tectonic roles by studying the
nature of dielectric dispersion. Slight wetting of rocks appears to show a drastic increase in dielectric permittivity
$\varepsilon$ in reciprocal proportion to frequency below several hundreds kilohertz, whereas electric conductivity $\sigma$
is almost independent of frequency (Lockner \& Byerlee 1985, Jonscher 1976, etc.). This peculiar dielectric dispersion may be
useful to characterize H$_{2}$O within the Earth's crust remotely by low frequency electromagnetic sounding from the
surface. Field observation as well as property measurement in laboratory provide complex conductivity $\sigma^{*} = \sigma +
j\omega\varepsilon = \sigma(1+j\gamma)$ (which is physically equivalent to complex dielectric permittivity, $\varepsilon^{*}
= \varepsilon + \sigma/j\omega$ as a function of frequency. Usually the quantity defined by
$\gamma=\omega\varepsilon/\sigma=1/\tan\delta$) has been supposed to be too small (less than ~0.1) in comparison with the
observation error both in situ and in laboratory. Further, laboratory studies on the dielectric dispersion have been quite
controversial so far. However, there is a hope in future availability of high-precision EM sounding method (Nakajima {\it et
al}, 2000), and (1) we developed a reliable and precise four electrode method of measuring complex dielectric permittivity at
low frequencies to a demanded precision of 10$^{-3}$, and (2) laboratory measurements of dielectric dispersion are made to
clarify the basic characteristics in order to correlate with the observation in situ. We noted that the ordinary measurement
methods usually show the large (1) measurement error originated from noise, and also (2) significant bias in $\gamma$ due to
the disturbance of electric field in the sample caused by potential electrodes even in 4 electrode method. Therefore, we have
used very accurately controlled sinusoidal signals to obtain the high S/N ratio by data stacking, and also a 4 electrode
method to enable us to measure the bulk property and interface property simultaneously by using a new type of potential
electrodes with least polarization. As a result, we came to determine both real part of complex conductivity to a precision
to 10$^{-5}$ and imaginary parts at least to a precision of 10$^{-3}$ at the frequency down to 10 mHz. This means that
$\gamma=\omega\varepsilon/\sigma$ can be determine to a precision of 10$^{-3}$. Our results of measurements are as follows
(a) Heterogeneous materials such as mixture of glass beads soaked in water show that $\gamma$ is nearly constant 0.01 $\sim$
0.003. This suggests that $\varepsilon$ is in reciprocal proportion to frequency and there should be some electrochemical
reactions involved in this dielectric dispersion. The demanded precision in field observation to obtain the meaningful data
on dielectric dispersion in situ is $\sim$0.003 or better. (b) uniform materials such as water and glass do not show any
increase of $\varepsilon$ in reciprocal proportion to frequency, whereas some previous paper reported the frequency
dependence of $\varepsilon$ very probably due to the bias originated from the distortion of electric field by potential
electrodes in the measurement.
DE: 0694 Instrumentation and techniques
DE: 3914 Electrical properties
DE: 3947 Surfaces and interfaces
DE: 5194 Instruments and techniques
DE: 8124 Earth's interior--composition and state (old 8105)
SC: Geomagnetism and Paleomagnetism [GP]
MN: 2003 Fall Meeting