Geoscience Reference
In-Depth Information
hydraulic fracturing events and localizing these events in
heterogeneous materials. However, with AE, there is
insufficient knowledge of where the fluids are actually
moving within the subsurface formations during hydrau-
lic fracturing as well as the actual extent of the resulting
fracture network (e.g., Warpinski, 1991). Also, there is
a recognized need for new and improved methods to
detect and localize fluid leakages around the walls of a
borehole during the life of a well. This is leakage detection
and localization need is particularly critical for boreholes
that traverse through freshwater aquifers where a risk of
contamination is a concern (Cihan et al., 2011).
The self-potential method corresponds to the passive
measurement of electrical signals associated with a
variety of source current mechanisms in the conductive
subsurface of the Earth, including a redox-based
contribution (Sato & Mooney, 1960; Castermant et al.,
2008; Mendonça, 2008; Revil et al., 2010) and a stream-
ing current contribution related to groundwater flow
(Revil & Linde, 2006; Revil et al., 2011; Ikard et al.,
2012). The self-potential inverse problem is similar, in
essence, to EEG in medical imaging. In the last decade,
the recording and inversion of EEG signals have been
instrumental in our understanding of how the brain
works and in the mapping of
pore water (see also Wurmstich & Morgan, 1994,
and Ushijima et al., 1999). We believe that these voltages
carry information about the fracture network that is
expected to be complementary to the information deter-
mined from microseismic, tiltmeter, wellhead pressure,
and wellhead flow measurements (Keck & Withers,
1979). Mahardika et al. (2012) recently provided a com-
prehensive framework to perform full waveform joint
inversion of passive seismic and associated electrical data
to invert for the position and moment tensor of hydro-
mechanical events.
In this section of this chapter, we are concerned with
the time-lapse monitoring of a laboratory-based hydrau-
lic fracturing/leak-off experiment using the time-lapse
record of self-potential signals. The goal here is to show
that the inversion of the electrical potential measure-
ments can be used to detect and localize the streaming
current disturbances caused by a borehole seal failure
and the resulting undesirable fluid migration along
the borehole during hydraulic fracturing operations.
A leak-off resulted from an attempt to hydraulically frac-
ture a porous block while our initial goal was to study the
electrical signals associated with such fracturing events.
5.3.2 Material and method
The porous material used for the laboratory fracturing
tests was a cement mixture (FastSet Grout Mix) and
was cured for about 10months before the tests proceeded.
The porous sample had a roughly cubical shape (
x
=30.5
cm×
y
=30.5cm×
z
= 27.5 cm; Figure 5.15). After curing,
several 10mm diameter holes (named #1 to #10 below)
were drilled into the block to varying depths such that
various tube sealing methods could be tested (see
Figure 5.15). Stainless steel tubing with 9.5mm outside
diameter was placed into a few holes using Loctite Instant
Mix 5 Minute epoxy as the tube sealing agent. The voltage
measurement electrodes were attached to the top and one
side of the block (16 electrodes on each face) using a plastic
template for precise positioning. AE sensors were also
mounted to three faces of the block for fracture event
localization. The electrodes were solid sintered Ag grains
with a solid AgCl coating and an active diameter of about
1mm. Each electrode had a voltage amplifier built into the
electrode casing, and each of these electrodes was electri-
cally connected to the block surface through a drop of con-
ductive gel that was normally used for EEG.
The electrical response during the experiment was
measured using a very sensitive multichannel voltmeter
its various functions
(Grech et al., 2008).
The flow of pore water associated with hydraulic
fracturing, waste and other injection wells, multiphase
oil and gas production, and borehole and other leakages
into the subsurface results in the generation of
measurable voltages. This occurs during field operations
in reservoir environments (Chen et al., 2005;Entov
et al., 2010) and in shallow aquifers (Wishart et al.,
2008) and has been associated with artificial seismic
sources (Kuznetsov et al., 2001). Similar conclusions
have been reached in volcanic environments (e.g.,
Byrdina et al., 2003), and there is a relatively broad base
of literature on laboratory observations of electromagnetic
fields associated with hydromechanical disturbances
(Moore & Glaser, 2007; Nie et al., 2009; Jia et al., 2009;
Chen & Wang, 2011; Wang et al., 2011; He et al., 2011,
2012; Onuma et al., 2011).
For instance, Moore and Glaser (2007) investigated
unconfined and confined samples of granite subjected
to hydraulic fracturing in the laboratory. Their results
indicate that the principal mechanism for the self-
potential response is due to the generation of a streaming
source current density associated with the flow of the
