Hydraulic fracturing of rock samples and monitoring of the associated acoustic activity

INTRODUCTION Hydraulic fracturing is used to stimulate oil and gas wells in order to enhance production by increasing the layer permeability. Schematically, the reservoir is fractured by injecting under pressure a fracturing fluid, then a propping agent is injected to keep the fracture open. This technique has also been used in geothermal energy projects where the aim was to test the feasibility of developing a heat exchanger in a deep granitic formation. The hydraulic fracture being used as a fluid and heat exchanger between two wells and the formation, energy could be recovered by circulating water in the fracture. Other application of this technique was the creation of a permeable channel in a coal seam for gas production by in situ coal gazification. In these different projects, the problem is to know where the fracture developed. For oil and gas wells the fracture must stay confined in the productive layer. To test the feasibility of such an operation it is then possible to perform a "minifrac" and to map the corresponding fracture. Then the location of this fracture (orientation and extension) will help to design the main stimulation. The most promising method to know the extension and orientation of the fracture is fracture mapping based on the interpretation of acoustic activity observed during hydraulic fracturing. The principle is to locate the discrete acoustic events recorded during the fracturing phases i.e. creation, propagation and fracture closure. This acoustic activity can be recorded at the ground level, in neighbour wells or more often in the treated well. This last case is more common, due to the fact that for the depths investigated, the energy involved is too poor compared to the wave attenuation in the different layers: these factors hinder a good detection of the events at the ground surface. Depending on the type of formation, detection in a neighbour well implies that this we]] is not further than fifty to hundred meters at the reservoir level. This only exists in the case of an exchanger system such as for geothermal applications. In conclusion for oil and gas wells applications, acoustic activity associated to hydraulic fracturing is recorded in the treated well. In these conditions, events generated during fracture creation and propagation cannot be recorded because of the bad signal to noise ratio during fluid injection. Only events recorded after shut in can be treated. We already tested such conditions using a specific tool (Saxda et al. 1988). In the same time, hydraulic fracturing was realized in laboratory conditions on various rock samples: sandstone, limestone and chalk. The aim was to know if the hydraulic fracture could be mapped on the basis of an interpretation of the recorded acoustic activity.