Experimental Investigation on the Effects of Proppant Migration and Placement on the Conductivity in Rough Fractures

ABSTRACT Proppant conductivity was usually measured under static or designed proppant concentration. The ISO 13503-5 standard provides specific experimental procedures for measuring the long-term conductivities of proppants. However, it does not consider the proppant migration and placement during the slurry process of the hydraulic fracturing operation. In fact, the fracture was not fully propped and the proppant settlement shape highly dependent on the slurry viscosity, proppant concentration, and pump rate. This paper designed a dynamic sand placement apparatus to evaluate the dynamic proppant placement and its conductivity. Formation rock was cut and splitted into the API standard shape and size. Both smooth and rough fracture surface morphology quantitative data was obtained by 3D scanning technology to further evaluate the effect of fracture roughness on proppant migration and settlement pattern. Dynamic fracture conductivities were measured under different slurry viscosity, proppant concentration, and pump rate. Results showed that among the three influence factors of the dynamic flow conductivity, the effects of pump rate is greatest, then is the sand ratio and the viscosity of fracturing fluid. The dynamic conductivity of rough rock plate is lower than that of smooth rock plate with rough fracture conductivity decreasing about 10%- 15%. It is suggested that the proppant conductivity should be evaluated under real formation fracture roughness. The design of slurry operation parameters should use dynamic proppant conductivity measurement results to enhance the effectiveness of hydraulic fracturing. INTRODUCTION At present, the fracture conductivity is mainly divided into propped fracture conductivity and unsupported fracture conductivity (Wang et al. 2022; Chen et al. 2019; Naik and Singh. 2021; Li et al. 2020). The main evaluation methods of proppant conductivity are the API static sanding conductivity test and numerical simulation test (Zhu et al. 2022; Wang et al. 2022; Jia et al. 2020; Li et al. 2020), most researchers idealize the placement method as that proppant is evenly placed in the fracture to study the conductivity. This process fails to combine the settlement and migration process of proppant in the fracture, which is a certain gap from the field. Researchers published API RP 61 as the standard for static conductivity test in 1989 (Duenkel et al. 2019). With the improvement of the accuracy of conductivity measurement in the oilfield, Marpaung et al. (2008) developed the laboratory steps for dynamic conductivity test to simulate the reservoir conditions of the oilfield. The process of dynamic conductivity test is divided into three different units: the pre-fluid/sand-carrying fluid pumping unit to simulate the fracturing process, the gas backflow to simulate the backflow and production unit, and the proppant conductivity test unit. The simulation results showed that the internal friction angle and other mechanical properties of the proppant jointly determine the conductivity of the fracture in the support section, and the lack of interlock of the proppant will still lead to the closure of the fracture in the case of low differential stress. Well-classified proppant filling layers with different particle sizes can provide better conductivity and improve production in a longer time. During the study, the roughness of cracks and the settlement of proppant were not considered into the influence factors, and the stress was taken as the main influence factor (Mehdi et al. 2016; Li et al. 2019). Mollanouri-Shamsi et al. (2018) also got the conclusion that the well-graded proppant filling layer can obtain better conductivity. At the same time, the shape of proppant was studied, and the conclusion that the medium proppant with irregular shape may be better in practical application was obtained. Jia et al. (2016) studied the impact of dynamic fracture conductivity in the production process, and simulated the conductivity in the fracture by constructing the fracture closure and injection model (FIPM). The model is constructed based on the force acting on the proppant, fluid filtration, wall effect, proppant concentration and the interaction between proppants. The study found that the proppant distribution in the fracture is not uniform and the effect of fracture height yield and proppant distribution on the production is obtained. Kong et al. (2022) studied the fracture conductivity and stress sensitivity of fully supported split shale, and studied the change of fracture capacity under different particle size proppant and different stress. The study showed that the use of large particle size proppant in the near wellbore zone and the use of small particle size proppant in the far wellbore zone can effectively improve the conductivity. When the larger particle size proppant is used, the conductivity and fracture sensitivity will increase accordingly. The author used the split rough fracture core, but failed to take the influence of fracture surface into account. Kathryn et al. (2014) designed three groups of contrast tests for shale with different properties, including two groups of proppant concentrations: no sand, 0.15 kg/m2 and 0.5 kg/m2. It is found that under the condition of low concentration of proppant, the fracture conductivity largely depends on the rock property and the roughness of the fracture surface. The degree of debris shedding, sliding and self-supporting of rocks with different properties during fracture shear fracture is different, which has a great impact on the fracture conductivity under the condition of low closure stress and low sand concentration. Zou et al. (2021) carried out static conductivity test for rough cracks and found that the migration distance of support agent in rough cracks was significantly limited. With the increase of closure pressure, the conductivity decreased by more than 80%. Finally, the conclusion was put forward that the sand laying concentration should be increased and the conductivity of cracks should be increased by using proportionally small-size support agent. Yu et al. (2021) quantifies the degree of reduction and sensitivity of the proppant under different stresses and verifies that the values obtained by the long-term diversion method are not significantly different from those obtained by the short-term method. In most cases, the short-term diversion can be used instead of the long-term one to test the proppant to reduce test costs.