Effect of adsorbed moisture on the pore size distribution of marine-continental transitional shales: Insights from lithofacies differences and clay swelling

Highlights

• Impact of adsorbed moisture on the pore structure was experimentally investigated.

• Water in clay-rich shales can be divided into clay-bound water and adsorbed moisture.

• Thermal maturity, component wettability, and pore structure jointly control the distribution of adsorbed moisture.

• Significance of clay-swelling and pore expansion was prospected.

Abstract

The variation in pore water distribution within gas shale reservoirs has a significant effect on gas content, and thus on resource evaluation. However, the characteristics of water micro-distribution and its effects on pore parameters are still not well understood due to the mixed wettability of shale and the complexity of the pore structure. In this study, six lower Permian transitional shale samples from the southern North China Basin, humidified at four levels up to a relative humidity of 98%, were selected for moisture-equilibrated experiments and low-pressure N₂ gas adsorption measurements. The results indicate that the adsorbed moisture in transitional clay-rich shales can be divided into capillary condensation water in the micropores and monolayer–multilayer adsorbed water in the non-micropores. Moreover, thermal maturity (VR_o), total organic carbon, clay, and carbonate are positively correlated with the adsorbed moisture and micro-/mesopores, indicating that water in shales could be hosted in inorganic pores as well as in organic pores. Furthermore, the distribution of adsorbed moisture is mainly controlled by the VR_o, component wettability (i.e., organic matter, clay, pyrite, and carbonate), and pore structure (micro-/mesopore distribution). In addition, a subtle adsorbed moisture may significantly reduce both the pore volume (PV) and specific surface area (SSA) of micropores, and the effect on micropores and SSA is more pronounced than that on the respective non-micropores and PVs. Additionally, the mechanism of clay swelling and pore expansion in clayey shale can provide certain insights for water–methane competitive adsorption, identifying clay type and pore size, and the formation of organo-mineral complexes.

Introduction

Shale is a fine-grained, clay-rich sedimentary rock that is generally deposited and preserved in depositional environments including lacustrine, swamp, lagoon, and shallow/deep shelf. (Loucks and Ruppel, 2007; Passey et al., 2010; Valenza et al., 2013; Rezaee, 2015; Zhang et al., 2016). Although different sedimentary facies shales have experienced the long process of sedimentary and diagenesis, the ubiquity of connate water in oil/gas-bearing shale reservoirs and their interaction with organic matter (OM) and inorganic minerals at different maturity stages cannot be neglected in developing a systematic understanding of shale oil/gas generation, accumulation, and preservation (Powers, 1967; Gasparik et al., 2014; Merkel et al., 2015; Cheng et al., 2019). For instance, most marine shale gas reservoirs under exploitation (i.e., the Woodford, Marcellus, Fayetteville, Haynesville, and Barnett shales) in North America, with thermal maturities (as indicated by VR_o) of 1.5–2.5%, generally have a water saturation of 10–35% (Boyer et al., 2006; Bowker, 2007; King, 2010; Cipolla et al., 2010; Wu and Aguilera, 2012). The Lower Paleozoic marine shale gas reservoirs (Wufeng-Longmaxi shales) in southern China, with high VR_o values of 2.5–3.5% (Zou et al., 2016; Nie et al., 2020), still maintain water saturations of 10–60% (Liu and Wang, 2013; Wei and Wei, 2014; Fang et al., 2014). The typical potential marine shales mentioned above are generally dominated by type I and II kerogen, high total organic carbon (TOC), high brittle mineral content (especially quartz), and low clay content, while lacustrine shales and transitional shales, such as the Yanchang and Shanxi-Taiyuan shales in China, usually have different reservoir characteristics, including type III kerogen, higher clay content, and more complex lithofacies (Curtis, 2002; Zhang et al., 2009a; Passey et al., 2010; Xiao et al., 2013; Zou et al., 2019; Dang et al., 2020), which means that the latter (clay-rich shales) may have unique water-holding characteristics.

Therefore, it is essential to consider the role of water in transitional shales with complex lithofacies and strong heterogeneity. Shale is mainly composed of OM and inorganic minerals (quartz, clay, feldspar, pyrite, carbonate, etc.) and usually has ubiquitous micro–nano pores (Curtis et al., 2010; Loucks et al., 2012; Rezaee, 2015), which cause a significant water wettability difference between organic and inorganic pores. The traditionally held view of OM is that it is hydrophobic (Lewan, 1997; Boyer et al., 2006; Borysenko et al., 2009; Passey et al., 2010; Odusina et al., 2011; Sulucarnain et al., 2012), whereas inorganic minerals are hydrophilic, particularly clay (Wang and Chen, 2006; Zhang et al., 2010; Jin and Firoozabadi, 2014; Korb et al., 2014); that is, OM-hosted pores are hydrophobic or hardly contain water (Odusina et al., 2011; Sulucarnain et al., 2012; Yassin et al., 2017), and mineral pores are hydrophilic or easily adsorb water (Korb et al., 2014; Zolfaghari et al., 2017a; Yuan and Rezaee, 2019; Feng et al., 2018a, Feng et al., 2018b). Mineralogically and texturally distinct shales show different affinities for water, ranging from strongly hydrophilic to strongly hydrophobic, and thus, have heterogeneous mixed wettability (Dehghanpour et al., 2012; Hu et al., 2014; Guo et al., 2020). Furthermore, whether organic pores are accessible to water depends on the OM maturity, kerogen type, and especially oxygen-containing functional groups (Do and Do, 2000; Bekyarova et al., 2002; Do et al., 2009; Chalmers and Bustin, 2010; Ruppert et al., 2013; Hu et al., 2014; Bahadur et al., 2017). For example, Larsen and Aida (2004) observed that four different types of kerogen could adsorb water, and suspected that the polar functionalities in these kerogens were involved in both diffusion and water absorption. Zou et al. (2020) also reported that kerogen in Bakken shales can adsorb water molecules and change pore size distribution (PSD). Furthermore, some organic materials without oxygen-containing functional groups retain their hydrophilic ability. For instance, Liu and Monson (2005) found that pore size and bulk vapor pressure determine water ad-/desorption isotherms in porous carbons using Grand Canonical Monte Carlo (GCMC) simulations. Kozbial et al. (2014) and Wei and Jia (2015) explained that graphite is intrinsically more hydrophilic than previously believed, and that surface-adsorbed airborne hydrocarbons or factors such as surface contamination and roughness are the source of hydrophobicity. In addition to OM in shales, inorganic minerals are not always hydrophilic. Fassi-Fihri et al., 1995 believed that expandable clays were more likely to be oil-wet than non-expandable clays. Smectite typically undergoes diagenetic transformation during burial (as defined by the illite/smectite ratio), and diagenetic reactions in the reservoir may lead to evolution in wettability (Powers, 1967). Khan et al. (2014) believed that bitumen and water are roughly equally strongly adsorbed to plagioclase and calcite, whereas water displaces bitumen from quartz, gypsum, potassium feldspar, and mica surfaces. Tabrizy et al. (2011) found that the hydrophilicity of quartz, kaolinite, and calcite decreases sequentially by using microcalorimetry; however, Barclay and Worden (2000) reported that the pH of formation water and the source of silica (internal or external to the reservoir) can influence quartz wettability. Moreover, iron-rich silicates and carbonates (such as smectite, chlorite, siderite, or ankerite) are also often oil-wet (Treiber et al., 1972; Chilingar and Yen, 1983; Anderson, 1986; Cuiec, 1987), while feldspars can display mixed wettability (Barclay and Worden, 2000). Most interestingly, some hydrophilic minerals may turn to lipophilicity resulting from surface coats of hydrophobic substances such as pyrobitumen, Fe-oxides, minor smectite, chlorite clay, or a few percent carbonate minerals (Barclay and Worden, 2000; Lan et al., 2014). In summary, the wettability of OM and inorganic minerals is not absolute or immutable, but is usually restricted by its own physical and chemical structure and external conditions (such as water chemistry conditions, temperature and pressure, and redox conditions).

Compared with conventional reservoirs, pore-type shales are complex and diverse, with not only OM pores but also many inorganic pores; the former are mostly less than 50 nm, while the latter (including clay pores, rigid mineral intergranular pores, and carbonate dissolution pores) have a wide PSD ranging from a few nanometers to thousands of nanometers (Loucks et al., 2009, Loucks et al., 2012; Chen et al., 2016; Dang et al., 2019). The pore structure and surface properties of shale inevitably change due to the influence of moisture content. Most studies have reported that OM pores are hydrophobic and hardly condense water, while inorganic pores are usually hydrophilic (Boyer et al., 2006; Borysenko et al., 2009; Odusina et al., 2011; Sulucarnain et al., 2012; Firouzi et al., 2014; Korb et al., 2014; Gasparik et al., 2014). As mentioned above, however, some OM can be hydrophilic so that it may actually contain water molecules. Some scholars believe that water can be condensed in small pores (< 5–6 nm) and adsorbed on non-microporous surfaces (> 5–6 nm) in shale clay (Lewan, 1997; Li et al., 2016, Li et al., 2017). Cheng et al. (2017) studied over-mature marine shale and found that adsorbed water mainly exists in mesopores (2–50 nm) and macropores (> 50 nm) instead of micropores (< 2 nm). Similarly, Sang et al. (2005) and Zhang et al., 2009a believed that liquid water can only wet the outer surface of coal and some large pores (> 1000 nm, seepage pores), but cannot enter small pores (10–100 nm) and micropores (< 10 nm) because of interfacial tension. In addition, the water in a low-rank coal can be distributed by the wet-migration of liquid water and diffusion of gaseous water, and stored in the forms of liquid free water, capillary water, and bound water (Li et al., 2018). Nevertheless, Schmid (1964) showed that fine pores in preserved cores were water-wet, whereas the larger pores were much less water-wet. Recently, Hu et al. (2019) also suggested that adsorbed water mainly exists in micropores (< 2 nm) and mesopores (2–50 nm), rather than macropores (> 50 nm), by studying over-mature marine shales. Li et al. (2019a) concluded that adsorbed water mainly occurs in micro–/mesopores (< 10 nm) by using an imbibition test and NMR experiments. Interestingly, Ruppert et al. (2013) found that most pores associated with OM in Barnett shales were accessible to both water and methane over a wide size range of 10–10000 nm, and pores smaller than 30 nm preferred water.

Therefore, although there are many studies on the influence of water on shale, the evaluation of water micro-distribution characteristics and its effect on pore parameters is still not well understood due to the mixed wettability of shale and the complexity of the pore structure. In this work, the effect of adsorbed moisture on the microscopic pore structure of transitional shales from the southern North China Basin (SNCB) was evaluated quantitatively by considering the wettability of material compositions and swelling characteristics of clay. The objectives of the present study are three-fold: (a) to investigate the water distribution characteristics of transitional shales; (b) to analyze the effect of adsorbed moisture on the microscopic pore structure, and (c) to assess the geological control mechanism for moisture distribution in transitional shales.