New Techniques for Measuring Rock Fracture Energy

Abstract

Abstract The use of closed-loop, servocontrolled test systems to obtain rock fracture energy measurements in uniaxial tension, torsion, and direct shear is described. Results of fracture energy tests performed on three rock types are presented. Analysis of the test results substantiates the validity of measuring fracture energy by the new test techniques. The advantage of these new tests, providing data potentially useful for input into analytical rock potentially useful for input into analytical rock response calculation, far outweighs the advantage -simplicity of test procedure-of the usual test methods. Fracture energy values for three rock types tested under four different load configurations indicate that fracture energy requirements depend upon the type of loading applied to the specimen. The ratio of fracture energy required in uniaxial compression to that required in uniaxial tension rages from 140 for Berea sandstone to 560 for Barre granite. Introduction The need for better means of characterizing rock fragmentation is becoming increasingly apparent to researchers in all fields of rock mechanics. A rock fragmentation characterization that eliminates rock/ machine interaction effects must be developed if fragmentation systems are to be understood and improved. Many drilling researchers use measurements of "specific energy" (compressive energy per unit volume) in their analysis of rock behavior per unit volume) in their analysis of rock behavior but such measurements combine rock characteristics and rock/tool interaction into a single measured value. Researchers are becoming more aware of the importance of fracture energy (i.e., all energy consumed in fracturing a rock of unit cross-sectional area) in the analysis of rock failure. Drilling and hydraulic fracturing are examples of load applications encountered in petroleum production where fracture energy must be known for the proper analysis of rock behavior. The practical use of rock fracture surface energy measurements was presented in an analysis of the hydraulic fracturing phenomenon almost a decade ago. Fracture surface energies were then measured on cleavage beams. Recent improvements in closed-loop, servocontrolled test systems made it possible to measure directly the energy required to cause rock failure under laboratory conditions for several different test configurations and stress regimes representative of field applications. The test methods that were investigated and that are reported here provide energy measurements for uniaxial tensile, torsional, direct shear, and uniaxial compressive failures. The purpose of this paper is to introduce fracture energy measurement techniques that the new to rock mechanics. A secondary purpose is to describe briefly the techniques of measuring fracture energy by obtaining "complete" load-deformation curves for rock in uniaxial tension torsion, and direct shear. To validate the test procedures, test results are presented for three rock types. Theory Griffith introduced the concept of energy balance as a hypothesis to his theory on rupture. Griffith's energy balance hypothesis was that a crack will extend as long as the potential energy available to the crack is greater than the surface energy required to extend the crack (generate a surface) in that system. Griffith then went on to calculate stress fields and to relate strength and failure through stress levels. We proposed to measure directly all energy consumed in rock fragmentation by conducting laboratory tests in such a manner that cracks can grow under stable conditions -- conditions, that is, where the potential energy within the test system is controlled so as to incrementally supply energy necessary to create new rock surface. For such conditions, a complete load-deformation history of a specimen loaded to failure can be obtained. The work performed on the specimen can directly be found from the load-deformation history. We have adopted "fracture energy" as a broad term that includes surface energy and all other dissipative energies associated with and necessary for the generation of new surfaces within a rock volume per unit area of cross-section.