On the water retention behaviour of shales

Shales are among the most considered geomaterials in current energy-related geomechanical investigations; they are involved in engineering applications such as the unconventional extraction of natural gas, CO2 sequestration and nuclear waste geological storage. A deep understanding of their behaviour with regard to variations in the degree of saturation is of primary significance for such applications. This paper illustrates the advances in the experimental analysis of the water retention behaviour of shales under non-isochoric conditions which have been recently achieved by the authors. The testing methodologies involve the direct control of the water content and the subsequent measurement of the total suction at equilibrium by a psychrometer. Furthermore, the volume change response of the material upon total suction variations can be investigated: a fluid displacement technique with a non-polar liquid is used in order to assess the volume changes of the shale samples and to compute the degree of saturation. The results highlight the important features of the retention behaviour of shales, such as the existence of main wetting and drying paths, the hysteresis domain and the scanning behaviour. Selected results are presented for three shales from the northern region of Switzerland. liquid limit wL and the plastic limit wP). The results of the geotechnical identification of the cores are reported in Table 1 while Fig.1 depicts the grain size distributions of the considered shales [1]. Table 1: Geotechnical characterization of the tested shales Shale OPA-shallow BD-deep OPA-deep s (Mg/m) 2.74 2.72 2.72 (Mg/m) 2.46 2.52-2.56 2.49-2.55 w (%) 6.9 2.7 – 4.4 3.3 – 4.9 e (-) 0.21 0.09 0.12 0.09 0.15 Sr (%) 92 69 99 80 96 wL (%) 38 25 33 29 39 wP (%) 23 10-23 19-25 Fig.1. Grain size distribution of the tested shales. The pore size density (PSD) function of the tested shale cores is determined through the Mercury Intrusion Porosimetry (MIP); this technique consists in forcing the penetration of mercury into a sample and in measuring the intruded volume of mercury as a function of the applied pressure. The required pressure is inversely proportional to the size of the progressively filled pores. The obtained results are presented in Fig.2. The PSD functions are found to be unimodal for all the tested shales; they show pore modes at approximately 20 nm for the OPA-shallow and BD-deep and at 8 nm for the OPA-deep. Fig.2. Pore size density function of the tested shales. 3. TESTING METHODOLOGIES This section describes the experimental methodologies established to analyse the water retention behaviour of shales under unstressed conditions. In all the depicted testing procedures, synthetic waters are used with the aim of reproducing the in-situ pore water composition [2, 3]. The osmotic suctions of the synthetic waters used are measured by a dew-point psychrometer and result in values of 1.2 MPa for the OPA sample and 0.98 MPa for the BD-deep and OPA-deep samples. The presented testing techniques are based on the direct control of the water content and on the subsequent measurement of the total suction value [4]. The imposed water content is applied differently for the wetting and drying paths. The total suction readings are carried out by means of a dew-point psychrometer (Decagon, WP4c) [5, 6]. The apparatus gives indirect measures of the total suction of a sample by reading the relative humidity established by the sample in a closed environment. The chilled-mirror dew point technique is used to measure the relative humidity. The total suction is then obtained by the psychrometric law [7] which relates the relative humidity (RH) and the absolute temperature (T) to the total suction ():