 |
| Titel |
Laboratory-scale measurements of self-potentials during CO2 releasing. |
| VerfasserIn |
Cristian Vieira, Alexis Maineult, Maria Zamora |
| Konferenz |
EGU General Assembly 2010
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 12 (2010) |
| Datensatznummer |
250032621
|
|
|
|
|
|
| Zusammenfassung |
| Carbon sequestration projects include monitoring programmes focused on tracking evolution
of injected gas. Induced changes in porous media triggered by degassing in case of leakage
comprise chemical reactions and fluid motion. These processes are likely to induce
Spontaneous Potential (SP) signals that can be detected on surface with proper arrange of
electrodes (Girard & Bitri, BRGM report 56299, 2008; Byrdina et al., JGR 114,
2009). Here we attempted to reproduce such leakage scenarios at the laboratory
scale.
Different arrays of unpolarisable electrodes were placed in a rectangular, plexiglas box
containing Fontainebleau sand saturated with 0.56 NaCl g/L water solution. Measurements
were systematically recorded during stages of i) initial conditions (i.e., without gas flow), ii)
low rate CO2 injection, iii) increased CO2 injection rate and iv) post injection. CO2 injection
was performed in one half of the tank by means of a cylindrical air diffuser. That device
allowed us to homogeneously distribute gas for better comparing differences between
signals in the perturbed zone and the non-perturbed zone. The whole set-up was
designed to be an open system since gas reaching the surface of the tank affects no
longer the signals. Acquisition was made with a High impedance voltmeter Keithley
2701. Because we wanted to analyse possible effects of the investigated volume and
signals, we registered under similar conditions SP responses with three different
unpolarisable electrodes (Cu/CuSO4, Ag/AgCl and Pb/PbCl2) having different measuring
area.
Signals obtained display a congruent behaviour. However, under identical procedure
conditions we didn’t observed similar values in terms of amplitude. The general
trend suggests that the responses are subordinated to the amount of gas reaching
the zone on the interface electrode/media. The perturbed zone is characterised by
higher voltages, especially in areas with more degassing. Beginning of injection
was always reflected as a pulse on electrodes in the perturbed zone. Increment in
injection rate was also identifiable yet not as evident as injection start. The spam of
injection is usually reflected by electrical anomalies that differ from what should be
inherent drift. When comparing signals between different electrodes, higher voltages
were obtained in smallest electrodes, probably indicating a relation with chemical
inertia.
Non-reproducibility of results was due to erratic behaviour of gas in porous media.
Degassing path during injection continuously migrated in time alternating between
accumulation and leakage. At the considered scale, flow paths of gas reaching the electrodes
varied heterogeneously, making almost impossible any computer modelling. In the same
sense, terms that compose SP signals could not be quantified. Due to the scale we are
working, we cannot state results to be obtained when applied the technique on field scale.
Further research has to be done for assessing feasibility of applying surface SP monitoring at
field scale, which is much larger than the representative elementary volume. At laboratory
scale, the relation is the opposite for being smaller than the representative elementary volume. |
|
|
|
|
|
|