The capillary sealing efficiency of fine‐grained sedimentary rocks has been investigated by gas breakthrough experiments on fully water saturated claystones and siltstones (Boom Clay from Belgium, Opalinus Clay from Switzerland and Tertiary mudstone from offshore Norway) of different lithological compositions. Sand contents of the samples were consistently below 12%, major clay minerals were illite and smectite. Porosities determined by mercury injection lay between 10 and 30% while specific surface areas determined by nitrogen adsorption (BET method) ranged from 20 to 48 m2 g − 1. Total organic carbon contents were below 2%. Prior to the gas breakthrough experiments the absolute (single phase) permeability (kabs) of the samples was determined by steady state flow tests with water or NaCl brine. The kabs values ranged between 3 and 550 nDarcy (3 × 10−21 and 5.5 × 10−19 m2). The maximum effective permeability to the gas‐phase (keff) measured after gas breakthrough on initially water‐saturated samples extended from 0.01 nDarcy (1 × 10−23 m2) up to 1100 nDarcy (1.1 × 10−18 m2). The residual differential pressures after re‐imbibition of the water phase, referred to as the ‘minimum capillary displacement pressures’ (Pd), ranged from 0.06 to 6.7 MPa. During the re‐imbibition process the effective permeability to the gas phase decreases with decreasing differential pressure. The recorded permeability/pressure data were used to derive the pore size distribution (mostly between 8 and 60 nm) and the transport porosity of the conducting pore system (10‐5–10‐2%). Correlations could be established between (i) absolute permeability coefficients and the maximum effective permeability coefficients and (ii) effective or absolute permeability coefficients and capillary sealing efficiency. No correlation was found between the capillary displacement pressures determined from gas breakthrough experiments and those derived theoretically by mercury injection.

The capillary-sealing ef®ciency of intermediate-to low-permeable sedimentary rocks has been investigated by N 2 , CO 2 and CH 4 breakthrough experiments on initially fully water-saturated rocks of different lithological compositions. Differential gas pressures up to 20 MPa were imposed across samples of 10±20 mm thickness, and the decline of the differential pressures was monitored over time. Absolute (single-phase) permeability coef®cients (k abs ), determined by steady-state¯uid¯ow tests, ranged between 10 À22 and 10 À15 m 2 . Maximum effective permeabilities to the gas phase k eff (max), measured after gas breakthrough at maximum gas saturation, extended from 10 À26 to 10 À18 m 2 . Because of re-imbibition of water into the interconnected gas-conducting pore system, the effective permeability to the gas phase decreases with decreasing differential (capillary) pressure. At the end of the breakthrough experiments, a residual pressure difference persists, indicating the shut-off of the gas-conducting pore system. These pressures, referred to as the`minimum capillary displacement pressures' (P d ), ranged from 0.1 up to 6.7 MPa. Correlations were established between (i) absolute and effective permeability coef®cients and (ii) effective or absolute permeability and capillary displacement pressure. Results indicate systematic differences in gas breakthrough behaviour of N 2 , CO 2 and CH 4 , re¯ecting differences in wettability and interfacial tension. Additionally, a simple dynamic model for gas leakage through a capillary seal is presented, taking into account the variation of effective permeability as a function of buoyancy pressure exerted by a gas column underneath the seal.

No abstract