Evaluation of Drilling-Fluid Filter-Loss Additives Under Dynamic Conditions

Abstract

KRUEGER, ROLAND F., MEMBER AIME, UNION OIL CO. OF CALIFORNIA, BREA, CALIF. Abstract Results are presented from tests of dynamic fluidloss rates to cores from clay-gel water-base drilling fluids containing different commercial fluidloss control agents (CMC, polyacrylate or starch), organic viscosity reducers (quebracho and complex metal lignosulfonate) and oil at several different levels of concentration. In the dynamic system the most effective individual additives to the clay-gel drilling fluid, based on cost-equalized concentrations, were found to be starch and the viscosity reducers. These results do not conform with the rankings determined by API fluid-loss tests, which indicate CMC, polyacrylate and starch to be the most effective and comparable. Generally, minimum dynamic fluid-loss rates were attained at cost-equalized concentrations of additive (including thinner) of about $1.00/bbl, or less.For chemically treated clay-gel drilling fluids, both the standard and the high-pressure API filter-loss tests were found to be inaccurate indicators of trends in dynamic fluidloss rates under the test conditions used, particularly for drilling muds containing viscosity reducers. From a practical field viewpoint, restrictions on the applicability of the API fluid-loss test are such that it is open to question whether or not results of this test can be used routinely with confidence as an indicator of control of downhole fluid loss under field treating conditions. Introduction The petroleum industry spends large sums of money during drilling operations to control the fluid-loss properties of drilling fluids based on the standard API filter-loss test, which is a static filtration system. Laboratory studies of dynamic filtration have shown that in a given time period filtrate loss from a circulating mud stream is greater than from a static system and that it is a function of linear mud velocity, pressure and the properties of the drilling fluid. Ferguson and Klotz and Horner, et al, observed thatthe dynamic fluid-loss rates for the drilling fluids used were not related to the extrapolated API filter loss andthe drilling fluids with the lowest API filter losses did not have the lowest dynamic fluid-loss rates. However, there has been no published information on the relative effects on dynamic fluid-loss rate as a given drilling fluid is treated with increasing amounts of chemical additive to reduce the API filter loss. Such information is economically important because drilling-fluid costs rise rapidly as chemical requirements increase.This paper presents the results of a study of dynamic filtration rates to cores from a clay-gel water-base drilling fluid treated with various commercial viscosity reducers and chemical fluid-loss control agents. The dynamic fluid-loss rates to cores are compared with the standard API filter-loss values at several different levels of additive concentration.Dynamic filtration rates were obtained in each experiment under two different simulated wellbore conditions:filtration just above the bit through a new mud cake laid down dynamically on a freshly drilled formation andfiltration up-hole through a mud cake formed by deposition of a static filter cake on top of the initial dynamically formed cake. The latter case corresponds to the bottom-hole conditions existing above the bit when mud circulation is restarted after a stand of pipe has been added or a round trip has been made to change the bit. Except for the short-duration, high-rate filtration beneath the bit where no mud cake can form, these two conditions probably represent the two extremes of dynamic filtration. Because thickness of a dynamic mud cake formed on freshly exposed formation is limited by the shearing action of the mud stream, the filtration rate for this condition is high. On the other hand, once circulation is stopped and a static mud cake forms on top of the dynamic cake, re-starting circulation has only a small effect on the cake properties and filtration rate is much lower thereafter. A discussion of the mechanics of mud-cake deposition and dynamic filtration is outside the scope of this paper but may be found in more detail in publications by prior investigators. APPARATUS AND EXPERIMENTAL CONDITIONS The test equipment used to simulate the dynamic flow conditions existing during drilling was a modification of that described previously by Krueger and Vogel. A schematic flow diagram is shown in Fig. 1. In general, a power-driven, high-pressure mud pump capable of delivering up to 60 gal/min was used to circulate drilling fluid parallel to the faces of 1-in. diameter sandstone cores mounted in a 2 3/4-in. ID high-pressure test cell. Pump rates were controlled by means of magnetic clutch to maintain an average axial fluid velocity of 110 ft/min in the annular space between the cell wall and a 1 1/2-in. rod positioned on the center line of the cell. The core specimens were Berea sandstone plugs sealed with plastic inside 1 1/8-in. OD tubes and were fluid-saturated prior to use. Burettes were used to accumulate fluid discharged from the cores.The mud sump shown was used for treatment and storage of the drilling-fluid samples during a particular test. JPT P. 90^