New technology based on the use of potassium hydroxide to permanently 11 fix 11 clays in the near wellbore area is taken from the laboratory development status to the field. Injection well treatments using KOH are compared with the normal soak (potassium chloride) and coat (cationic polymer) approach in the same reservoir.

SPE Members Abstract Recent tests have shown that the conductivity of a 20–40 sand pack is increased by blending angular sand with the very round product currently used in hydraulic fracturing. The oil and gas industry requires fracturing sand to meet high roundness specifications, according to the Krumbein and Sloss chart, for the purpose of providing the optimum geometric pack in the purpose of providing the optimum geometric pack in the fracture. Indeed this will decrease point pressure on the sand grains and reduce crushing and fines generation in the sand pack, but tests indicate that the optimum geometric pack for crush resistance does not always offer the maximum conductivity through a sand pack. pack. Sand grains with a roundness factor of .5 (the minimum APT standard for frac sand is .6) were mixed in 50/50, 62.5/37.5, 75/25, 87.5/12.5 and 100/0 blends with sand grains having a roundness factor of .8. These combinations were placed in a conductivity cell and tested at closure stresses as high as 10,000 psi. Several blends proved more conductive than the 100% highly rounded grains used as a control sample. This paper will show how the increased void space of a round-angular sand combination overcomes its susceptibility to crushing and provides increased conductivity, the most reliable measurement of a sand's overall performance. Introduction In developing a study on 20-40 fracturing conductivity it is important to review three current principles generally accepted by the well stimulation industry.The key to a successful hydraulic fracturing treatment is to increase conductivity in the producing formation while minimizing permeability damage.The most important function of a propping agent in a hydraulic fracturing treatment is to create and sustain fracture conductivity.In using silica sand as a propping agent fracture conductivity can best be created with a very round and spherical product. We agree with the first principle. Indeed the industry's search for additional conductivity has prompted this and many other studies. We also agree with the second principle and believe that a reliable analysis of a proppant's fracture conductivity under simulated reservoir conditions would provide the most significant measure of its overall performance. The third principle we contend is not true all of the time. The American Petroleum Institute (APT) has recommended roundness and sphericity numbers for what would be considered an acceptable silica sand for hydraulic fracturing (Roundness is the smoothness of a sand grain's surface or the lack of rough edges. Sphericity is the degree a sand grain resembles a sphere or circle.). Those numbers are .6 for both roundness and sphericity according to Krumbein and Sloss (Fig. 1). We do believe that a high sphericity must be maintained because this eliminates sand grains with elongated shapes or protrusions and prevents their fitting into a natural void space and restricting conductivity. In the case of roundness, however, we do not feel that it is always advantageous for all of the sand grains to be perfectly round or smooth. We feel the same about the reverse situation of 100% angular grains for even though angular blends have the most porosity in a loose pack they possess surfaces that will break off or crush easily with any significant closure stress because of uneven pressure distribution on the surface points that touch each other. But if a certain percentage of angular grains are interspersed in a very round, spherical product the benefit of a highly porous angular sand can negate its prohibitive properties of easy crushing and produce prohibitive properties of easy crushing and produce more conductivity than either of the pure round or angular blends. This will be demonstrated with conductivity test results. The results of these findings could have a significant technological and economical impact on the usage of 20–40 silica sand for hydraulic fracturing.

Summary The Triangle "U" unit is located in Campbell County, Wyoming, in the Powder River basin. The field produces mainly from the Sussex A sandstone, with completions and limited production from the Sussex B. The flood recovered 12.8% original oil in place (OOIP) on primary before the waterflood, which began in March 1981. The Sussex A is relatively tight, with an average permeability of 15 md and porosity of 13.5%. The rock contains swelling and migrating clays, and the initial injection water source was fresh, leading to concerns about long-term injectivity. To stabilize clays, two different processes were applied. Earlier injection wells were treated with a combination of potassium chloride (KCl) and cationic polymer. Later injection wells were treated with potassium hydroxide (KOH). A recent comparison of long-term performance of the two groups of injection wells shows that the wells treated with KOH injected 476,437 bbls/porosity-ft more water than the wells treated with cationic polymer, in 1.4 years less time. This is an 83% increase in cumulative water injection. After KOH, all injection wells were put on a low concentration of imbibition agent to maximize in-depth penetration of water into low permeable rock. Cumulative oil recovery through March 1997 is 36.4% OOIP, compared to the original waterflood projection of 26.6% OOIP. A total of 37.7% pore volume (PV) water has been injected, and the water/oil ratio (WOR) is currently 0.71, for a fairly efficient flood in this tight, dirty sandstone. Introduction The Triangle "U" unit produced 12.8% OOIP on primary before initiation of a waterflood. Several methods of secondary recovery were considered for this reservoir. Gas injection was not feasible because of limited supplies, and micellar injection was too expensive and risky. Waterflood susceptibility testing in cores showed favorable displacement of oil by water, making this the most appropriate secondary recovery method. The waterflood was projected to recover an additional 13.8% OOIP. Polymer flooding was not considered, because the mobility ratio was favorable and the reservoir was relatively tight, with an average permeability of 15 md. There were two basic challenges to waterflooding. First, there was concern that clays would limit injectivity over time. Also, the rock exhibited a permeability variation of 0.65, which could lead to bypassing of recoverable oil as water tended to establish channels through more permeable rock. Clays can exacerbate channeling. SPE 53007 was revised for publication from paper SPE 39937, first presented at the 1998 SPE Rocky Mountain Regional/Low Permeability Reservoirs Symposium, Denver, Colorado, 5-8 April.

The Triangle "U" Unit is located in Campbell County, Wyoming, in the Powder River Basin. The field produces mainly from the Sussex "A" Sandstone, with completions and limited production from the Sussex "B". The flood recovered 12.8% OOIP on primary prior to the waterflood, which commenced in March, 1981. The Sussex "A" is relatively tight, with an average permeability of 15 md and porosity of 13.5%. The rock contains swelling and migrating clays and the initial injection water source was fresh, leading to concerns about long-term injectivity. To stabilize clays, two different processes were applied. Earlier injection wells were treated with a combination of potassium chloride and cationic polymer. Later injection wells were treated with potassium hydroxide (KOH). A recent comparison of long- term performance of the two groups of injection wells shows that the wells treated with KOH injected 556,000 BBLS/porosity-foot more water than the wells treated with cationic polymer, in 1.4 years less time. Following KOH, all injection wells were put on a low concentration of imbibition agent to maximize in-depth penetration of water into low permeable rock. Cumulative oil recovery through March, 1997, is 36.4% OOlP, compared to the original waterflood projection of 26.6% OOlP. A total of 37.7% PV water has been injected, and the water-oil-ratio is currently 0.71, for a fairly efficient flood in this tight, dirty sandstone. P. 307