A discussion on the decrease of unconfined compressive strength between saturated and dry rock samples

The unconfined compressive strength (UCS) of a rock is a basic parameter for many characterization systems, strength criteria and calculation methods. It is well-known fact that it depends on the water content of the samples, and decrease when the water content increases. The paper discusses the possible causes of this reduction. From published data by Vasarhelyi and co-workers and others authors some empirical tentative guidelines for this reduction are proposed, which can be used in rock engineering problems where changes in water content occur regularly (dam and bridges foundations, harbors...). .Hsu and Nelson (1993), in a preliminary research for the not built Super Collider, correlated the unconfined compressive strength of many types of shale (from Canada and USA) with the water content. Their results (fig 2), show a marked negative correlation between water content and compressive strength Fig 2 Unconfined compressive strength vs. water content for clay shales (Hsu and Nelson, 1993) Ballivy and Colin (1999) have analyzed the increase in triaxial strength related to changes in the dielectric constant of the fluid saturating the rock. In a propane storage cavern in shale the tension strength of the rock increased 150-200% due to the change in the dielectric constant, with a reversal to the prior strength when the propane evaporated. In their opinion changes in the saturation fluid cause changes in the effective stresses, a result already stated by Vutukuri (1974). In the same paper they show increases in the compressive strength of 20% when testing gneiss saturated with salt water (with a small decrease in the dielectric constant) over the same test saturated with distilled water, and increases in the compressive strength of 25-50% of dry samples over the same test saturated with distilled water. These results show a clear trend but cannot be generalized due to the small number of tests done. The respective dielectric constant are: 80, distilled water; 74, salt water; 0 dry state Lashkaripour and Passaris (1993) compiled a data base with selected values of shale rock properties. Fig 3 shows data from two coal mines. There is also a marked negative correlation between water content and compressive strength 3 CAUSES OF THIS DECREASE In strong indurated rocks of low porosity the compressive failure is preceded by the growth of cracks from the border of existing micro pores. The cracks coalesce into growing cracks finally extending to the sample dimension and failure happens. Fig 3. Unconfined compressive strength vs. water content for two shales (Lashkaripour and Passaris, 1993). (a) Linton Lane coal mine, (b) Rye Hill coal mine According to Vasarhely and Bobet (2000) there are three fundamental theories on crack initiation criteria: maximum tangential stress (Erdogan and Sih, 1963), maximum energy release rate (Hussain et al, 1974) and minimum energy density (Sih, 1974). Any of them can reasonably predict tensile crack initiation, both in tension and/or compression, but not in shear. In the simpler case the crack initiation occurs as “a progressive lengthening of the crack across the infinite plate” (Rummel, 1974).The mathematical formulation “involves a consideration of the energy change during the crack growth”. Following Rummel there are three energy terms to be considered: change in potential energy of the applied forces, change of strain energy due to the existence of the crack and change in surface energy. So the Griffith criteria for tensile fracture can be stated (in the simpler formulation) as σt = (2 E γ / π c) where: σt is the tensile strength of the material E is the deformation modulus γ is the specific surface energy c is the crack initial half length As has been shown by Ballivy and Colin (1999) the nature of liquid has a direct influence in the crack openings, a fact due to the decrease in surface energy of the crack borders when the pore is full of water. A similar explanation is offered by Vasarhely and Ledniczky (1999): “moisture diminishes the spread of free surface energy, i. e. it facilitates micro-cracks propagation by decreasing the elastic limit and the peak strength also” On the other hand the crack growth can be originated by increasing water pressures within the pores when the rock is saturated. Both effects can happen simultaneously In poorly cemented rocks the presence of water can affect to the cementation between the grains by different ways: solution, dispersion...Finally in soft argillaceous rocks the water diminishes the strength of the grains and/or the cementation So there are different causes which produce, together or unconnectedly, the reduction in strength 4.-SOME PUBLISHED DATA Steiger and Leundt (1990) gave some data extracted from an EXXON comprehensive research program on shale typical properties, shown in table 1 Table 1.-Data on UCS of typical shales (Steiger and Leundt, 1990)

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.