Formation interface fracturing experiment: an in situ investigation of hydraulic fracture behavior near a material property interface. Final report

摘要

Two hydraulic fractures were created, one above and one below an ash-fall tuff-welded tuff formation interface. These formations have significant differences in their Young's moduli, Poisson's ratios and porosities. Sufficient cement was injected into each zone to create vertical fractures of 600 ft total length if the height was restricted to 50 ft; 256 and 117 bbls were injected at 6 bbl/min into the ash-fall tuff and welded tuff zones. Mineback along the interface has revealed a total fracture length of 150 ft; coring has shown that the heights of the fractures are about 100 ft for the fracture in the ash-fall tuff and 200 ft for the welded tuff fracture. The fracture which was initiated in the low modulus ash-fall tubb propagated upwards into the higher modulus welded tuff. No containment was observed. The results from the fracture subsequently initiated in the welded tuff are obscured because it propagated alongside the lower fracture and thus provided no definitive information on behavior at the interface. Widths in both the ash- fall and welded tuffs are consistent with design calculations. The in-situ stresses were found to have the greatest effect on fracture behavior and geometry. Variations in the minimum principal in situ stresses controlled the direction of fracture propagation and the final height of the fracture. The low stress measured in the welded tuff probably aided the propagation of the fracture into that region. Natural fractures in the welded tuff caused significant offsets in the induced fractures and resulted in filling of some of the natural fractures with grout and severe fluid Leakoff. These experiments show that material property interfaces should not, in general, be considered containment features. A more likely factor in controlling fracture height is in situ stress differences.