This paper summarizes a series of hydraulic fracture tests done to observe fracture initiation and growth in a variety of multiple-perforation and well orientations. A transparent brittle-elastic polymeric material is used to allow visual monitoring of the entire initiation and fracture growth process using a variety of fluids and injection rates. Several general observations from the roughly 40 constant injection rate tests can be made: Breakdown Pressure: only weakly increases with injection rate (~doubles for rate increase by factor 30-90); Perforation Initiation Sites: fractures tend to initiate from few perforations at low rate with water, while fractures initiate and grow from almost all perforations at high rate and with high viscosity fluids; Initial Fracture Orientation: far-field stresses have little influence on the of the initial fractures at the perforations; Final Fracture Orientation: primarily determined by the far-field minimum in-situ stress; Transition to the Final Orientation: alignment to preferred final orientation occurs quicker at high injection rate. More significantly, the general pressure behavior of these tests is different from the breakdown and monotonic pressure decay with growth typically assumed and modeled. The generic behavior observed in these tests shows a sharp pressure peak with rapid decay following fracture initiation and initial growth, then a roughly steady pressure ?plateau? with further fracture growth. Such pressure plateau behavior has also been observed in other hydraulic fracture tests on rocks and some field data as well, and is incompatible with typical Griffith-type fracture growth assuming zero or constant fracture toughness, but can be characterized by an increasing fracture toughness with size. Such increasing fracture toughness with fracture size is typically observed in many engineering materials (so called "R-curve" behavior), and is associated with the growth of a process zone (damaged or plastic deformation zone) with fracture propagation. Such increasing fracture toughness with fracture size has also been observed by the authors in concrete.
1. INTRODUCTION
This paper summarizes a series of hydraulic fracture tests done to observe fracture initiation and growth in a variety of simple geometries representative of possible field conditions. The objective of these tests is to directly examine the initiation and growth of fractures from a variety of perforation geometries and injection rates, and also containment at stress or stiffness contrasts. The work was motivated following an unpublished literature review of available information on fracture initiation. In particular, while there is available experimental data on fracture initiation in a variety of materials (e.g. see [1], [2], [3], [4]), almost none of the data are on transparent material allowing direct detailed visualization of the initiation and early growth (a notable exception is ref [5]). By using a transparent brittle elastic polymer, the tests presented here can provide such observations.
2. EXPERIMENTAL SETUP DESCRIPTION
This equipment is designed to provide true-triaxial stress control on a sample, controlled fluid injection rate, and continuous visual monitoring of hydraulic fracture growth from a vertical or inclined wellbore. Figure 1 is a labeled photograph of the entire experimental setup.
