The TOC contents and mineral compositions of the shale samples were presented in Table 1. The TOC content of four shale samples was between 1.21 wt.% and 3.24 wt.% with an average value of 2.24 wt.%. These shale samples mainly consisted of clay minerals, quartz and carbonates. More specifically, the clay content was 20.6 wt.%–64.3 wt.% with an average value of 39.2 wt.%, the quartz content was 12.6 wt.%–46.4 wt.% with an average value of 32.2 wt.% and the carbonates were 5.2 wt.%–63.4 wt.% with an average value of 22.8 wt.%. And the lithofacies of these shale samples classified according to Tang et al. (2016) based on TOC content and mineralogy was also presented in Table 1.
Sample Well No. Depth (m) TOC
Clay minerals (wt.%) Pyrite (wt.%) Lithofacies B1 X1 3 393.2 1.21 28.5 1.4 5.2 64.3 0 Organic-moderate clayey shale B5 X4 4 064.0 1.98 41.1 5.5 10.5 39.3 3.6 Organic-moderate Siliceous shale B12 X3 2 323.5 2.54 12.6 1.2 63.4 20.6 2.2 Organic-rich calcareous shale B14 X2 3 607.8 3.24 46.4 4 12 32.5 5.1 Organic-rich siliceous shale
Table 1. TOC content and mineralogy of Longmaxi Formation shale samples used in this study
OM pores were the dominant pore type in B5, B12 and B14 samples and the OM-clay complexes were commonly observed in these three shale samples. There are clay minerals embedded in OM, and some clay minerals are located outside OM and directly connected with external space (Figs. 4a, 4c, 4e). The OM pores of these shale samples were well preserved due to the high contents of rigid minerals (e.g., quartz and pyrite) which had a strong ability to resist compaction (Figs. 4b, 4f). However, the shape, quantity and sizes of OM pores in these shale samples were quite different. As shown in Fig. 4b, the OM pores in sample B5 were mainly bubble pores probably related to the thermal evolution process of OMs while the primary OM pores with amorphous shapes formed by stacking of sedimentary OMs were eliminated a lot and less developed due to the strong compaction at this high thermal maturity stage. And about 53.0% of the total OM pores in sample B5 was less than 10 nm while OM pores larger than 50 nm only accounted for 1.1% and OM pores larger than 100 nm were almost not developed (Fig. 5). The area porosity of OMs in sample B5 was generally less than 5%. As shown in Fig. 5, few OM pores less than 5 nm were developed in samples B12 and B14, which accounted for 10.3% and 2.1% in samples B12 and B14 respectively. However, more OM pores larger than 50 nm were developed in samples B12 (7.6%) and B14 (12.2%) compared with sample B5 (1.1%) (Fig. 5). The amorphous primary OM pores and bubble pores with a large diameter were the main OM pore type developed in samples B12 (Fig. 4d) and B14 respectively (Fig. 4e). Furthermore, the area porosity of OMs in samples B12 and B14 was usually greater than 5%.
Figure 4. Pore development characteristics of organic matter in shale samples. (a) Clay minerals in OM (organic matter) and clay minerals outside OM in sample B5; (b) less distribution of amorphous primary OM pores, abundant existence of bubble pores within OM and existence of pyrite in sample B5; (c) clay minerals in OM and clay minerals outside OM in sample B12; (d) abundant existence of amorphous primary OM pores with a large diameter in sample B12; (e) clay minerals in OM, clay minerals outside OM and abundant existence of bubble pores with a large diameter in sample B14; (f) existence of pyrite and quartz in sample B14.
All the N2 adsorption hysteresis loops of four shale samples shown in Fig. 6 belonged to H4 type according to the IUPAC classification standard (Sing, 1985), which indicated that the slit pores were mainly developed in these samples (Sing, 1985). The specific surface area distribution curves of the mesopores in these shale samples were shown in Fig. 7 and mesopores with diameter 2–6 nm made a great contribution to the total specific surface area of these shale samples. As shown in Table 2, the BET surface area of four shale samples ranged from 15.83 to 28.80 m2/g, which was comparable to the reported data in the literature (Wang et al., 2016). The HPMIP curves of shale samples were presented in Fig. 8 and sample B14 showed the highest mercury retraction efficiency (~60%) indicating its wellconnected pore system. The median pore-throat diameter corresponding to 50% mercury saturation of these shale samples was also presented in Table 2. Sample B5 had much larger median pore-throat diameter than sample B14. As OM pores in sample B5 had smaller pore sizes compared with OM pores in samples B14, it is appropriate to speculate that the inorganic pores in B5 had larger pore sizes than inorganic pores in B14.
Figure 6. N2 adsorption/desorption isotherms for Longmaxi Formation shale samples. TOC stands for total organic carbon; P stands for the equilibrium pressure; P0 stands for the saturated vapor pressure.
Sample BET surface area (m2/g) Volume of pores < 50 nm (V1) from N2 adsorption (mL/g) Volume of pores between50 nm and 1 μm (V2) from HPMIP (mL/g) Total pore volume combining HPMIP and N2adsorption (V1+V2)a (mL/g) N2 adsorption-HPMIP Porosity (ϕN2–HPMIP))b(%) Median pore-throat diameter fromHPMIP (nm) B1 19.09 0.016 2 0.002 70 0.018 9 4.97 13.7 B5 17.57 0.010 3 0.001 32 0.011 6 3.12 43.8 B12 15.83 0.010 8 0.001 65 0.012 4 3.24 46.2 B14 28.80 0.017 2 0.001 45 0.018 6 4.81 14.6 BET. Brunauer, Emmett, and Teller theory; HPMIP. high pressure mercury intrusion porosimetry. a The pores larger than 1 μm were not counted to eliminate the effect of microfractures. bϕN2–HPMIP=(V1+V2)/VHPMIP)×ϕHPMIP, where VHPMIP is the total pore volume obtained from HPMIP, ϕHPMIP is the porosity obtained from HPMIP.
Table 2. Pore structure information obtained from HPMIP and N2 adsorption for Longmaxi Formation shale samples
Figure 8. Mercury saturation (%) vs. pore-throat diameter (μm) obtained from HPMIP (high pressure mercury intrusion porosimetry) for Longmaxi Formation shale samples.
The whole-aperture pore size distribution curves of four shale samples were obtained by combining mesopore information from N2 adsorption and macropore information from HPMIP (Fig. 9). It could be seen that the mesopore volume was significantly higher than macropore volume for all these four shale samples and there were several distribution peaks located in the pore diameter range of 3 to 6 nm. And the proportion of mesopore volume in total pore volume was 85.3%, 86.5%, 85.5% and 90.0% for samples B1, B5, B12, B14 respectively.
The absolute methane adsorption curves of shale samples
with different water contents were shown in Fig. 10. For dry shale samples, the absolute methane adsorption capacities of four shale samples increased with increasing pressure and the absolute MACs were in the following order: B14 > B12 > B5 > B1. With the increase of water content in shale samples, the absolute MACs of all the samples gradually decreased. However, the absolute MACs of samples B1, B5, B14 no longer significantly decreased with the increase of water content when the relative humidity was higher than 76% while the absolute MAC of sample B12 continued to decrease significantly with the increase of water content when the relative humidity was higher than 76%.