Experimental Study of Fracturing Fluid Loss

Abstract

Abstract This paper describes experimental work for the filtration characterization of common hydraulic fracturing fluids. The concept of filter-cake resistance is introduced and elements of visco-elastic theory are used for the interpretation of observed phenomena. Important conclusions from this work include relationships between filter-cake leakoff coefficient and polymer loading and predictive models for filter-cake leakoff behavior for specific polymer loads, temperature and differential pressure. The results can be incorporated into any robust model to account for fluid leakoff during hydraulic fracture execution. Introduction Fracturing fluid loss is a composite of several effects. Of these, the filter-cake properties play a critical role for leakoff control during a fracture treatment(1–7). The objective of this work was to investigate properties of filter-cake leakoff and ultimately their impact from the pressure history of a fracture treatment. This can permit a rational basis for combining the effects of different fluid-loss control mechanisms. In order to isolate filter-cake properties, experiments using filter-paper were done. Experiments conducted in accordance with previous work(1,2) using constant pressure differentials across filter-cakes have revealed no essential contradictions on the influence of temperature, pressure and polymer loading. Relationships between the leakoff coefficient and fracturing pressure can thus be developed. Experiments, though, with pressure histories simulating both the pumping and closure phase of a fracture treatment yielded findings which required a new interpretation. A generalized empirical interpretation for describing the cake filtration is offered here covering both the fracturing and closing periods. Visco-elastic theory hints towards this interpretation. For clarity and brevity, the filter-cake leakoff coefficient will be denoted as wall coefficient. Procedures and Basic Relations The experiments were conducted with a high-pressure, high temperature filter press commonly used for static fluid-loss testing. A zirconate-crosslinked HPG-polymer solution, prepared according to standard industry procedures, was used as the fracturing fluid system. Filtrates were collected continuously and crosslinked polymer was clear and had the Newtonian properties of plain water. Evaporation was prevented by cooling the effluent. Sufficient duplicate experiments were performed to ensure the quality of the resulting data. Basic leakoff theory for filtercakes defines the filtrate volume By(3): Equation (1) (Available In Full Paper) and the wall coefficient by: Equation (2) (Available In Full Paper) where V is the filtrate volume per unit area, t = time, Sp = spurt loss, k = cake permeability, b = constant of deposition, Δp = pressure differential and µ> ƒ = viscosity of the filtrate. The constant of deposition is defined as: Equation (3) (Available In Full Paper) where V> ƒ the filtrate volume and Ve is the cake volume. A plot of the filtrate volume vs. > √r results in a straight line where the wall coefficient is directly related to the slope by: Equation (4) (Available In Full Paper) The correction of measured filtrate volumes for spurt loss ensures that all straight lines have the same point of origin, which is advantageous for a clear comparison of the straight-line slopes. This procedure was employed for the analysis of filte