Rock Engineering: Where is the Laboratory?

Abstract

This paper is based on an invited lecture presented at the ARMA (ARMA-American Rock Mechanics Association) 2018 Rock Mechanics/Geomechanics Symposium, Seattle, WA. June 17–20, 2018. At the time of the invitation, the author was preparing an “Appendices” to a book (Sikora (2018) Charles Fairhurst—The Long Shadow published (via Amazon) by Itasca Consulting Group) to be published by Itasca Consulting Group—but page limitations required that the “Appendices” be eliminated. The invitation from ARMA provided an ideal opportunity to present the essence of the notes to professional colleagues. This paper attempts to expand on the presentation in Seattle. At the International Society for Rock Mechanics (ISRM) in 1962, Dr. Muller emphasized the central importance of large-scale discontinuities and anisotropy in rock engineering, and the need to establish a discipline distinct from the continuum fields of elasticity and plasticity. With the benefit of over 50 years of hindsight, it is clear that both elasticity and plasticity are important in rock mechanics—but discontinuities, especially on the scale of engineering projects, can be critical. Having been involved with the development of rock mechanics at the University of Minnesota since the late 1950s; with Professor Cundall as a faculty colleague since 1972; and with the founding of Itasca Consulting Group in 1981, the topic of discontinuities in rock has been a prominent long-standing concern to the ‘Minnesota group’. Theoretical developments in mechanics are often stimulated by experimental observations in classical ‘bench-scale’ laboratories. Thus, elasticity theory was stimulated by Hooke’s experiments (1678) and plasticity by Tresca’s experiments (1864). Even if it was possible to construct a laboratory to test ‘specimens’ of a rock mass on a scale sufficient to include large discontinuities, separation from the rock mass would remove in situ forces from the specimen, resulting in unknown changes to the specimen. What are the options to establish the constitutive behavior of the rock mass? Where is the laboratory? This paper discusses past attempts to answer this question and suggests a direction for the future.Rock in situ is unlike any other material encountered in engineering. Typically, it will vary in age from several hundreds of millions to as much as a few billions of years. Rocks of different composition and mechanical properties are often adjacent to each other. Subject to changing tectonic forces and gravity over this period, the rock mass is mechanically complex, and usually contains systems of fractures and mechanical interfaces, varying from grain boundaries to tectonic plate boundaries. Within this range, discontinuities comparable in size to the dimensions of engineering projects in rock are of particular concern to designers. The International Society for Rock Mechanics [Recently re-named International Society for Rock Mechanics and Rock Engineering. The acronym (ISRM) has been retained] was formed in 1962 to focus attention on the need to develop mechanic-based design procedures to give due consideration to such discontinuities. The paper reflects on approaches taken to address this concern in the almost 6 decades since formation of ISRM. Early efforts concentrated on testing of large physical models in a laboratory, plus a variety of efforts to incorporate discrete discontinuities into continuum mechanics. Particular attention is given to the development of the Discrete Element Method (DEM), introduced by Cundall (Proc Symp Int’l Soc Rock Mech 2:129–132, 1971). Examples of the application of DEM to practical design problems and conclusions drawn from them are discussed. In some cases, results show important differences with the current procedures and empirical rules. Although most of the examples shown are drawn from mining, references are made to applications in other engineering fields, especially Civil Engineering and recent developments in Enhanced Geothermal Systems. Currently, the principal limitation to widespread application of DEM to rock engineering design problems is computational speed. This problem is one faced in many scientific and engineering disciplines, so it is anticipated that solutions will be developed in the coming several years. In the meantime, simpler representations of discrete fracture systems are used to develop valuable general insights to inform practical designs. Recent leadership by the US Department of Energy in rock mechanics research through FORGE and SubTER [Frontier Observatory for Research in Geothermal Energy (FORGE); Subsurface Science, Technology, Engineering, and R&D Crosscut (SubTER).] offers hope that the importance of subsurface engineering to the US is being recognized. A problem, mentioned recently by Hoek (2018) that needs to be addressed in the United States is that of developing an engineering workforce capable of applying analytical and numerical techniques sensibly to design in rock.