The stress dependent elastic properties of thermally induced microfractures in aeolian Rotliegend sandstone

Abstract

A B S T R A C TThe impact of thermally induced microfractures on the stress-sensitive elastic wave properties of aeolian Rotliegend sandstone samples is analysed. It is found that to identify the effects of the microfracture contribution accurately, a correction must first be made to account for water loss (representing a mass loss of 4-6%) from the pore throats and clays due to the heating process, despite care being taken to ensure that the thermally fractured samples re-adsorb room moisture. Both the original and thermally fractured rocks are stress-sensitive at the ultrasonic wave frequencies of the laboratory. However, a distinct shift in the estimated distribution of internal rock compliance indicates that the population of thermal microfractures differs in nature from that caused solely by core-plug extraction damage. In particular, the ratio of normal to tangential compliance is observed to be higher for the thermally generated microfractures than for the broken grain-grain contacts created by extraction unloading. This can be explained by the intragranular thermal-fracture surfaces being smoother when compared to the intergranular boundaries. Mechanical hysteresis is observed between the up-and downgoing test cycles for both the original and, to a greater extent, the thermally fractured rock. This indicates that there is compactioninduced movement of the fractures in the samples during application of stress in the laboratory.

Exploring trends in microcrack properties of sedimentary rocks: An audit of dry-core velocity-stress measurements

“…10 red line), we also observe an increase in the range of B N /B T ratios, with values between approximately 0.0 and 1.9. These results show some similarity with those of Exploring trends in microcrack properties of dry rocks MacBeth and Schuett (2007), Figure 7c, except that our estimated means are significantly larger. Potential sources of discrepancy between our results and those of MacBeth and Schuett (2007) are theoretical differences in methodology, errors introduced from digitizing the graphical data into discrete velocity versus stress data and applying the high stress assumption to estimate the background elasticity.…”

“…The inversion results show initial crack densities ranging between approximately ǫ 0 ∈ (0.08, 0.8) and initial aspect ratios ranging between a 0 ∈ (0.0005, 0.005). The inversion results for the thermal damage experiment of MacBeth and Schuett (2007) confirm the expected result that the initial crack density term is sensitive to core damage. Although qualitatively instructive, the range of values obtained from a limited number of sandstone samples is not statistically compelling.…”

“…However, the initial crack density and aspect ratio remain poorly constrained. Verdon et al (2008) estimate the initial crack density and aspect ratio of several sandstone samples from the Clair reservoir (Kendall et al, 2007) and several thermally damaged sandstone samples (MacBeth and Schuett, 2007). The inversion results show initial crack densities ranging between approximately ǫ 0 ∈ (0.08, 0.8) and initial aspect ratios ranging between a 0 ∈ (0.0005, 0.005).…”

“…For ν ∈ (0.1, 0.4), one would predict the range of the B N /B T ratio for dry (or gas filled) discontinuities to be (0.75,0.95). However, recent laboratory estimates of B N /B T ratio for sandstone (MacBeth and Schuett, 2007) and limestone (Lubbe Exploring trends in microcrack properties of dry rocks , 2007, 2008) dry core samples have found values significantly lower than unity, with ratios ranging between 0.0 and 0.6. MacBeth and Schuett (2007) also note that for (thermally) damaged samples the ratio may increase to values above unity.…”

“…For comparison, we estimate B N /B T using the velocity-stress data for the aeolian Rotliegend sandstone samples presented in Figures 3, 4 and 6 of MacBeth and Schuett (2007). The results for the undamaged samples (Fig.…”