Advanced Rheological Techniques for Optimizing Borate-Crosslinked Fracturing Fluid Selection and Performance
- David J. Power (U. of Melbourne) | Lincoln Paterson (CSIRO Petroleum) | David V. Boger (U. of Melbourne)
- Document ID
- Society of Petroleum Engineers
- SPE Drilling & Completion
- Publication Date
- December 2001
- Document Type
- Journal Paper
- 239 - 242
- 2001. Society of Petroleum Engineers
- 3.2.3 Hydraulic Fracturing Design, Implementation and Optimisation, 2.5.2 Fracturing Materials (Fluids, Proppant)
- 0 in the last 30 days
- 262 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
The complex rheological behavior of crosslinked fracturing fluids is a function of shear history, temperature, and chemistry. Understanding the relationship between these variables and the downhole properties of the fracturing fluid is a challenging task. Rheological measurement techniques are presented that unravel some of the mystery associated with crosslinked fracturing fluids. The concept of a physical gel-point for fluids undergoing crosslinking is introduced and shown to correlate strongly with the proppant- carrying ability of the fluid. The gel-point variation with temperature and chemistry is discussed. This variation can be studied in the laboratory to provide an in-situ field performance evaluation without the need for expensive proppant-transport flow loops. The limitations for newly developed, low-concentration polymer fluids are also discussed. The fluid system chosen for analysis was the pH-activated, borate-crosslinked hydroxypropylguar (HPG) fluid.
Crosslinked polymer fluids, in particular the borate-guar (or HPG) fluid, have been proven as highly successful fracturing fluids for many years.1 Recently, new generation, low-polymer fluids have proven to be effective fracturing fluids in low-temperature reservoirs.2 While these fluids have demonstrated field success, methods for understanding the crosslinking mechanism and the physical properties that lead to their success are still not complete. Rheological techniques that use oscillatory testing provide an ideal method for studying the liquid-to-gel transition that occurs in crosslinked fracturing fluids.3,4 Oscillatory rheometry allows the network structure of the gel to be studied in a nonintrusive manner. In this way, it is possible to monitor the effect that polymer concentration, crosslinker concentration, and temperature have on the development of the gel structure and, subsequently, evaluate the potential performance of specific fluids for use in the field.
The gelation process of the borate-crosslinked HPG fluid is pH controlled. At low pH values, the fluid is a liquid, while at high pH values, the fluid is in the gel phase. The key components of the system are the HPG polymer and a borate source. In this instance, boric acid is used as the borate source. Controlling the pH of the system controls the concentration of borate ions in solution. The borate ion is essential for crosslinking to occur. A more detailed description of the HPG-borate chemistry is given by de Kruijf et al.5 Rheological data for steady-state fluid samples allow the approach to the gel point to be characterized as well as denoting the exact point of transition from the liquid phase to the gel phase. Once the fluid is in the gel state, it is able to suspend and transport proppant.
A systematic approach to fluid preparation was used to avoid problems associated with sample variation. The required volume of water was heated to 140°F, at which point the HPG polymer was added gradually while applying high shear to the fluid. Once all the polymer was added, shearing was continued for 15 minutes. At this point, the sample was immersed in a water bath and held at 140°F for 1 hour. The sample was then placed on a bench-top roller to blend for 24 hours. A concentrated NaOH solution (20 wt% NaOH) was used to adjust the fluid pH. After each addition of NaOH, the sample was again heated to 140°F and rolled for an additional 24 hours. After loading a sample into the rheometer, a minimum of 15 minutes was allowed before testing to permit the crosslinks to reheal and reach equilibrium. In all fluid samples tested, the boric acid concentration was 3 lb/thousand gal (0.36 g/L).
All rheological data were obtained with a Weissenberg controlled-strain rheogoniometer with a cone-and-plate test fixture. Steady-state fluid samples (with a constant pH) at varying stages of crosslinking were tested at strains that were within the linear viscoelastic response region. A detailed discussion of the testing procedure is given elsewhere.6 The key parameter for gel-point determination is tan d. Tan d represents the ratio of the loss modulus (G'') to the storage modulus (G'). The storage modulus represents the elastic, or solid-like, behavior of a fluid, while the loss modulus represents the viscous, or liquid-like, behavior of a fluid. These properties are traditionally measured as a function of the oscillatory frequency. A detailed discussion of the storage modulus and loss modulus is given by Ferry.7
The tan d vs. the frequency data for fluids ranging from pH values of 6.25 to 11.29 are given in Fig. 1. The polymer concentration of the fluid given in Fig. 1 is 40 lb/thousand gal (0.48 wt%), with a boric acid concentration of 3 lb/thousand gal (0.36 g/L). The molecular weight of the HPG sample was estimated to be 3.5×106.6 For fluids in the liquid phase (low pH), the slope of tan d vs. the frequency curve is negative. As the pH is increased, the magnitude of the slope decreases and approaches zero. For fluids in the gel phase (high pH), the slope of the curve is positive. The exact pH corresponding to a zero gradient in tan d vs. the frequency is referred to as the gel point.3 Conceptually, the gel point represents the point at which all the polymer molecules in the solution are interconnected by at least one crosslink. The effective molecular weight of the polymer at this point approaches infinity.
To accurately determine the gel point, a first-order linear regression is plotted through the data points for each individual fluid. In Fig. 2, the gradient of each regression curve is plotted as a function of pH. A clear distinction can then be made of the transition from liquid (negative gradient) to gel (positive gradient). Thus, this analysis technique provides a simple method for determining the gel point of a crosslinked polymer fluid. For the fluid system depicted in Fig. 2, the gel point corresponds to a pH of 8.7 at 71.6°F.
Simulating the proppant-transport conditions expected downhole in a fracture has traditionally required expensive procedures and equipment.8 For this reason, a simplified approach has been pursued. Single particle-settling experiments were performed with a glass bead falling through a cylinder that contained the fracturing fluid. Fig. 3 depicts the apparatus used to quantify the settling rate of a 5-mm-diameter glass bead in a 50-mm-diameter cylinder that contained the test fluid. The settling rate for steady-state fluids at varying degrees of crosslinking was measured and correlated with the measured rheological properties of the fluid. Fig. 4 presents the settling rate vs. the fluid pH, where the fluid pH is used as an indicator for the degree of crosslinking of the fluid.
|File Size||198 KB||Number of Pages||4|