Griffith's theory of brittle fracture, modified to account for the frictional effects due to crack closure in compression, has proved a reliable basis for the understanding of fracture initiation in hard rocks. It suffers from the disadvantage that it gives no information on the mechanism of fracture propagation under compressive stress conditions. This paper describes an experimental investigation in which the mechanism of fracture propagation in glass and rock specimens was studied. An important feature of this investigation is the accuracy of the stress distribution in the specimens, achieved by means of a specially designed loading machine. The results of this study suggest that fracture initiation can be accurately predicted on the basis of Griffith's theory but that the final fracture path depends upon the coalescence of an array of cracks and the shearing of interlocking grains along the crack path. A mechanism of fracture propagation is proposed and its practical importance is discussed.
La theorie de la rupture par fragilite de Griffith, modifiee de façon à tenir compte de l'effet de frotternent dû à la fermeture des fissures soumises à la compression, a ete reconnue comme une base sûre pour la comprehension de l'initiation de la rupture dans la roche dure. Cette theorie ne fournit cependant aucune information concernant le mecanisme de la propagation de la rupture sous les efforts de compression. On decrit une etude experimentale sur le mecanisme de la propagation de la rupture dans des echantillons de verre et de roche. L'avantage de cette etude est la realisation, grace à un chargeur specialement affecte, de la repartition precise des contraintes dans les echantillons. Les resultats de cette etude laissent sup poser que l'initiation de la rupture peut être prevue exactement suivant la theorie de Griffith, mais que la direction ultime de la rupture serait subordonnee à la fusion d'un grand nombre de fissures, ainsi qu'au cisaillement des grains enclenches Ie long de son cours. On propose un mecanisme de propagation de la rupture. dont l'importance pratique est discutee.
Die Griffithische Sprödbruchtheorie, die im Hinblick auf die Reibungswirkung bei sich unter Druck schliessenden Rissen abgewandelt wurde, hat sich als verlassliche Grundlage zum Verstandnis der Bruchbildung im Hartgestein erwiesen. Sie hat jedoch den Nachteil, dass sie keinen Aufschluss ueber den Bruchfortschritt unter Druckspannungsbedingungen gibt. Im vorliegenden Referat wird eine experimentelle Untersuchung zur Bestimmung des Bruchverlaufes in Glas- und Gesteinsproben beschrieben. Bin besonderes Merkmal dieses Verfahrens ist die Regelmassigkeit der Spannungsverteilung im Pruefkörper, die durch eine eigens zu diesem Zwecke konstruierte Pruefeinrichtung erzielt wird. Die Untersuchungsergebnisse lassen den Schluss zu, dass sich der Brucheintritt nach der Griffith'schen Theorie genau voraussagen lasst aber dass der weitere Bruchverlauf von dem Zusammenftiessen einer Schar von Rissen und von dem Abscheren ineinander verschrankter Körner entlang des Rissverlaufs bestimmt wird. Eine Erklarung fuer den Mechanismus des Bruchablaufs wird geboten und seine Bedeutung fuer die Praxis herausgestellt.
Griffith's theory of brittle fracture [1] [2]*, modified by MCCLINTOCK and WALSH [3] to account for the effects of crack closure in compression, has proved to be a reliable basis for the prediction of fracture initiation in hard rock under static stress conditions [4] [5]. While this theory gives information on the initiation of cracks in the rock material, it does not define rupture of the. specimen nor does it give information on crack propagation. This paper describes an experimental investigation into the mechanism of fracture initiation and propagation from a closed crack in a glass model. On the basis of the results obtained in this study and from observations on the failure of quartzite, a failure mechanism for hard rock is proposed.
One of the principal deficiencies of the original Griffith's theory, when applied to the prediction of brittle fracture initiation under compressive stress conditions, lies in the fact that it does not make provision for the closure of very flat cracks before the tensile stresses at their tips have reached sufficient magnitude to cause them to propagate **. The frictional forces which occur when the crack faces are forced into contact can exert a marked influence upon the subsequent crack behaviour. McCLINTOCK and WALSH [3]assumed that, under compressive stress conditions, the crack faces are forced into uniform contact over the entire crack length and that the frictional shear resistance resulting from this contact must be overcome before propagation of the crack can occur. BERG [6] has shown that the above assumption is mathematically correct in the case of fine elliptical cracks. The stress system acting upon a closed crack is illustrated in Figure 1. In this analysis it is assumed that the initial crack in an unstressed body is uniformly closed over its entire length, i. e. the normal stress au required to cause closure of the crack is zero.
