Successful stimulation of an oil or gas well by a hydraulic fracturing treatment is largely dependent on obtaining a propped fracture with adequate fracture conductivity. The fracture conductivity is influenced by the reservoir environment following a hydraulic fracturing treatment. Many years of field experience have shown that it is usually necessary to use a gelled fluid to create fracture width and suspend the proppants. However, until only recently, laboratory testing methods did not realistically measure the damaging effects of these fluid systems on the conductivity of propped fractures.
As reported in recent publications, a few laboratories have constructed large scale models to simulate the pumping operations as well as the postfrac reservoir environment of a propped fracture. During pumping, treatment conditions are simulated by using a high shear flow loop to model wellbore tubular conditions, and a low shear, heated flow loop to simulate flow down the fracture. The fluid then enters a heated test cell where dynamic fluid loss in the fracture is modeled with core waters set apart to allow fluid to flow between them and create gel filter cakes. A proppont-laden slurry is then injected into the gap between the core waters and closure stress is applied. This completes the simulations of the fracturing treatment and the propped fracture. While holding the cell at the temperature of the reservoir being modeled, the fracture conductivity is monitored for long time periods at stress progressively increasing levels of stress.
This paper describes the equipment and procedures used to accurately simulate simulations treatment and downhole reservoir conditions. Resultant conductivity data of both sand and manufactured proppants are presented incorporating many of the commonly used fracturing fluid systems at different reservoir conditions of temperature and stress. Data comparisons are also presented whereby fracture conductivity data was obtained with both aqueous and hydrocarbon fluids as the flowing media.
References and illustrations at end of paper.
Results from recent fracture conductivity testing has significantly expanded the data base for fluids, temperatures, and proppants over any previous work, and also illustrates the limits of reproducibility of fracture conductivity measurements in the presence at fracturing fluid system filter cakes. Conductivity data using brine as a flowing medium is compared to data collected using kerosene as the flowing medium to simulate production from a liquid hydrocarbon reservoir. This comparison shows the effect of this important test parameter.
Using the test procedure described, considerable data has been generated at several conditions. The conditions investigated are for temperatures of 100 °F to 250 °F with fluid systems appropriate for each temperature condition. The proppants tested include 12/20 Brady sand, 20/40 Ottawa sand, and 20/40 ISP. KCI water and KNO3 water have been used to generate the data, considering brine as the flowing medium. KCI water was used for the earlier testing presented here. However, when corrosion of inlet and outlet filters began to occur, the flowing fluid was changed to KNO3 water. This change is considered acceptable because it keeps a potassium base fluid as the flowing medium and it eliminates the chloride corrosion problem.
