Hydraulic fracturing is necessary to develop low-permeability reservoirs, where a large volume of fracturing fluid is pumped into the horizontal wellbore to generate fractures with each fracturing stage. After the pad is pumped, proppants are added into the fracturing fluid and pumped into the created fractures to prevent them from closure. Currently, ceramic proppants and silica sands are two major types of proppants applied in the field. Ceramic proppants have higher strengths and lower crushing rates comparing to silica sands; laboratory measurements have indicated that the conductivity of ceramic proppants can be 5 to 10 times of silica sands at low to moderate closure stresses, and ceramic sands may loss the conductivity at a high closure stress of 60 MPa (Kurz et al., 2013; Zheng et al., 2018). However, ceramic proppants have a larger density, which is typically 1—1.5 times of silica sands. Since the settling rate of a proppant in the fracturing fluid changes linearly with its density according to the Stocks equation, the silica sand can migrate further than the ceramic proppant under the same pumping condition, and thus generating longer propped fractures (Gidley et al., 1990). Similar observations have been published in studies on proppant settling and migration in rough fractures or non-Newtonian fluids (Liu and Sharma, 2005; Clark, 2006; Blyton et al., 2018; Alotaibi and Miskimins, 2019). To reduce the settling rate of ceramic proppants, the lightweight proppant has been developed, whose density can be reduced to about 75% of silica sands, and the conductivity is enhanced by 40% compared with silica sands (Jackson and Orekha, 2017). However, its high price limits its field application in China. Besides, laboratory studies have also shown that the fracture roughness and tortuosity can also affect the proppant settling and transportation (Liu and Sharma, 2005; Sahai and Moghanloo, 2019; Qu et al., 2020). In viewing the heterogeneity of tight conglomerates in Mahu reservoirs, it is necessary to compare the performance of different types of proppants in the field, so as to explore the conditions when silica sands can partially or completely replace the ceramic proppants.
During hydraulic fracturing, proppants are delivered to the created fracturing by the fracturing fluid, whose viscosity also affects the settling rate of proppants. Laboratory measurements have shown that proppants are almost suspended in the crosslinked gel without settling, which is far slower than the settling rate predicted by the Stocks equation based on the Newtonian fluid assumption (Liu and Sharma, 2005; Hlidek and Duenckel, 2020). The crosslinked gel increases the fluid viscosity and thus can increase the fracture height (Olsen and Debonis, 1988; Fisher and Warpinski, 2012; Yue et al., 2019). However, the crosslinked gel cannot completely degrade and its residues can adsorb on the rock surface, thus inhibiting the hydrocarbon flow from the reservoir rock to the created fractures; meanwhile, the residues can also plug the propped fractures and reduce the fracture conductivity (Sarwar et al., 2011; Xu et al., 2011; Zhang et al., 2016). Laboratory measurements have shown that the conductivity reduction due to gel residues can be as large as 50% – 60% (Kim and Losacano, 1985; Wang et al., 2020). Slickwater is made of water and the friction reducer, whose concentration is typically about 0.1% (Liang et al., 2017). The friction reducer can reduce the friction of fracturing fluid by more than 70%, thus allowing a significantly larger pumping rate than the crosslinked gel; meanwhile, this large pump rate can slow down proppant settling in the slickwater and carry proppants further towards the fracture tips. Therefore, the crosslinked gel and the slickwater have their advantages and disadvantages, where the former has a better proppant-carrying ability and larger conductivity, and the latter has a lower price and causes less formation damage after fracturing. Since this is difficult to comprehensively evaluate the performance of two types of fracturing fluid, it is necessary to conduct field experiments to explore the best fracturing design for the future development of Mahu tight conglomerate reservoirs.
When the hydraulic fracture propagates, the adjacent reservoir rocks are compressed, which can increase the minimal horizontal stress and change the direction of fracture propagation. This is called the “stress shadow effect”, and can cause the non-uniform propagation of fractures within one fracturing stage (Vermylen and Zoback, 2011; Wu and Olson, 2016; Zhou et al., 2018). Recently, along with the downhole characterization techniques including the tracers, downhole camera, the distributed temperature sensing (DTS) and the distributed acoustic sensing (DAS), it has been found 80% production may come from only 20% clusters (Lecampion et al., 2015; Haustveit et al., 2017; Trumble et al., 2019; Wheaton et al., 2016). Observations from reservoir core samples from the Hydraulic Fracturing Test Site (HFTS) have also indicated that hydraulic fractures form the “fracture swarm”, where proppants unevenly distributes among them; when silica sands are used, thick fractures can be observed, but 80% propped fractures are less than 2 mm at a distance of 40—60 m away from the wellbore (Elliott and Gale, 2018; Maity et al., 2018; Maity and Ciezobka, 2019). For low permeability reservoirs, decreasing the fracture spacing can potentially increase the estimated ultimate recovery rate (EUR), but this can intensify the stress shadow effect and reduce the conductivity of created fractures. Therefore, the impact of fracture spacing on chosen proppant type is also necessary to be revealed in field experiments.
Since the year of 2017, 74 wells have been chosen in Ma-131 and Ma-18 plays for field experiments, where different types of proppants and fracturing fluids, slickwater ratios, and fracture spacings are tested. This paper summarizes and analyzes observations from field experiments during four years, and the learnings can help guide the optimization of hydraulic fracturing parameters for future wells in the Mahu reservoirs.