Geochemical cycle of hydraulic fracturing fluids: Implications for fracture functionality and frac job efficiency in tight sandstone

Abstract

The efficiency of the hydraulic fracturing processes is measured through its success rate in opening fracture space, fracturing fluid injectivity, and recovery rates of flowback fluid. This paper applies the volumetric ratio (RFF/FW) between recovered hydraulic fracturing fluid (FF) and formation water (FW) – besides gas production (GP) and flowback efficiency (FE) – as a third criterion and novel classification tool to quantify the loss of fracturing fluids, the inflow of formation water, and to define the type of involved fractures during hydraulic fracturing. Eight scenarios were designed to assess the performance and success rate of hydraulic fracking operations. For model calibration, geochemical time trends of flowback fluids from two horizontal wells were compared with the known composition of injected FF and local FW to quantify mixing ratios (RFF/FW) between both fluid types in produced water. The use of Na and Cl concentrations as nonreactive elements resulted in the most precise solution for endmember calculations. For the cased-hole scenario (C-1), low values of most operational parameters (gas recovery, injected fracturing fluid, recovered flowback volume, and FE of 0.37) reflect tight reservoir conditions with limited conductive capacity of induced fractures in the target zone. The elevated RFF/FW ratio (2.65) and dominant FF return during the first day of flowback suggest that injected fracturing fluids either remained close to the sealed borehole and returned immediately to surface during post-fracture production or were lost to the formation due to unconnected hydraulic fractures. In contrast, a complex fracture system at the open-hole completion (O-1) allowed the release of gas from micropores, reflected by relatively elevated ratios of 0.5 for FE and low RFF/FW ratios of 1.09. The interconnectivity between natural and induced fractures permitted the injection of larger volumes of FF with an elevated flowback of both fracturing fluid and formation water. A portion of 26% of the injected FF was recovered from well O-1 during a flowback period of 41 days, while an identical percentage was reached at well C-1 during 10 days. It is of practical importance for fracking operations that geochemical fingerprinting of flowback water can provide strategic decisions to optimize fracturing project performance beyond the capabilities of petrophysical data from well logging. In the present case, similar permeability and porosity characteristics for both targeted clastic intervals could not explain the contrasting performance of both frac jobs. Geochemical assessment can lead to avoiding water-pay zones to minimize the volume of required fracturing fluids for injection purposes, and to economize the recycling process for recoverable flowback fluids. An improved understanding of the functionality of the structural network of natural and induced fractures by the present classification is applicable to predict the success rate for upcoming frac jobs.