학술논문

Multicomponent diffusion in a 280 m thick argillaceous rock sequence.
Document Type
Article
Source
Applied Geochemistry. Aug2018, Vol. 95, p110-123. 14p.
Subject
*MULTIPHASE flow
*DIFFUSION
*GEOCHEMISTRY
*ROCKS
*BOREHOLES
Language
ISSN
0883-2927
Abstract
Diffusion is the main transport mechanism in many argillaceous formations. In this study, tracer and ion profiles in a 280 m thick clay-rich sequence were simulated by single component and multicomponent diffusion modelling. Drillcores from this sequence originating from a deep borehole in Schlattingen (NE Switzerland) had been previously extensively analysed in terms of porewater chemistry, mineralogy and diffusion parameters. In particular, data from high-pressure core squeezing had enabled to obtain depth profiles of major solutes and water tracers over the entire sequence. The hydrogeological conditions at the site were constrained in the model by the analogy of the nearby site at Benken. In a first step, a simple single component diffusion (SCD) model was set up to simulate the profiles of conservative tracers (δ 2 H, δ 18 O, Cl − ), to check reasonable boundary conditions for the adjacent aquifers and to estimate characteristic diffusion times. Based on these findings, a multicomponent diffusion (MCD) model considering explicitly diffusion in the electrical double layer (EDL) and the “free” water and a chemical equilibrium model was used to simulate the diffusion of major cations (Na + , Ca 2+ , Mg 2+ , K + , Sr 2+ ) and anions (Cl − , SO 4 2− , HCO 3 − ). The SCD modelling resulted in a good match of the measured water tracer and chloride profiles in spite of the uncertainty in the conditions regarding the surrounding aquifers. Diffusion times of 0.5–1 Ma were deduced which are in the same range as those postulated previously for the Benken site. Using the same type of boundary conditions, a reasonably good fit of the measured major cation and anion data could be obtained with the MCD model. The results were not sensitive to uncertainties inherent in the MCD model, such as the extent of surface charge screening by fixed cations or the thickness of the EDL. This supports the robustness of the model approach as long as key features such as anion exclusion are captured. Overall, the suitability of the MCD model for simulating cation and anion fluxes in argillaceous rocks over large distances and long timescales could be established. The results also support the validity of squeezing data from drillcores as proxy for in-situ porewater data. [ABSTRACT FROM AUTHOR]