BACKGROUND OF THE INVENTION ^•

1. Field of the Invention This invention relates to a method and apparatus for obtaining information from a borehole. In particular, this invention relates to a method and apparatus for measuring the fluid flow profile in a wellbore penetrating a petroleum reservoir.

2. Description of the Prior Art In oil field injection or production operations, it is important to know the injectivity profile of the formation adjacent to the wellbore. For example, during injection of fluids into a wellbore of an oil reservoir, it is important to determine the fraction of the total injection rate that enters each of the per .eable zones of the interval targeted for fluid injection. It is also important to detect zones that are not targeted to receive injected fluid, but which are likely to have fluid flow because of high permeability. Measurement of injection flow profiles is extremely important when the injected fluid is an enhanced oil recovery fluid such as a micellar/polymer or polymer-thickened water. This is because the injected fluids are extremely expensive and the success of the operations is often dependant upon achieving fluid injection into the desired reservoir zones.

Crucial information for proper planning of enhanced oil recovery operations includes the vertical conformity of the producing formations as well as their horizontal permeability and uniformity. This information is often obtained by an evaluation of the direction and speed of formation fluid flow by a borehole in the field. By obtaining such information at a sufficient number of boreholes throughout a field, a mapping of the total flow throughout a reservoir may be constructed to help plan the injection of chemicals or water in the enhanced oil recovery process.

In enhanced oil recovery operations, it is also critical to now the flow dynamics of the injected fluid through a given injection well borehole into the formation. Typically, an injection well is cased and the casing perforated at the levels of those zones into which fluid is to be injected. As fluid is pumped down the injection well, varying proportions of the fluid pass through the perforations into the different zones. The patterns of fluid flow into the various zones, including the proportion of fluid passing into each zone, are affected by the permeabilities of the various zones.

Several methods have traditionally been used to measure flow profiles in wellbores. Spinner surveys, which utilize a spinner rotor in the logging tool to detect flow can be used when the fluid flow rate is high. An example of a spinner flow meter is the Schlumberger Production Combination Tool, described in the publication "Schlumberger Engineered Production Services" published by Schlumberger Well Services of Houston, Texas. Unfortuantely, spinner survey tools currently available are often ineffective at the low flow rates normally encountered in wells containing viscous fluids. For example, in many wells the maximum injection rate per foot of reservoir interval flooded with viscous fluids is 5 barrels per day. For a 20 foot thick zone, the maximum flow rate in casing above the zone, prior to any fluid entering the reservoir, is 100 barrels per day. To accurately detect fluid loss to the reservoir zone, the logging tool must be able to detect a difference in flow rate of 5 barrels per day over a vertical distance within the wellbore of 1 foot. This is equivalent to a change in fluid velocity of approximately 0.16 ft/min if the hole diameter is 5 inches and tool outer diameter is 1.5 inches. Conventional spinner surveys are simply not sensitive enough to detect such low flow rates.

A logging tool commonly used to measure fluid flow profiles in low flow rate wells is the radioactive tracer flow logging tool. This is a very old and well known device. For

exaπrple, U.S. 2,453,456 to R. G. Piety on November 9, 1948 describes one of the earlier such radioactive tracer flow logging tools. In operation, two or more radioactive detectors are commonly suspended in a wellbore in a fixed, spaced relationship. A sample of radioactive material (often Iodine 131) is injected into the flowing wellbore fluid at a fixed point above the uppermost detector, and the arrival time of the radioactive sample at each of the detectors in the string is recorded. ^" If a large amount of tracer is flowing into the reservoir, differences will be detected in radioactivity level between the two detectors. Conversely, if little tracer is flowing into the reservoir, there will be little difference in radioactivity level between the two detectors. From this information, the flow rate of the radioactive sample between detectors can be calculated to generate an injection profile.

A current exacple of a radioactive tracer flow logging tool is the Nuclear Flolog (TM) or Tracerlog (TM) offered by Dresser Atlas. The Dresser Atlas tool is briefly described in a brochure entitled "Production Services" published in January, 1981 (reference number 9307) by Dresser Atlas, Dresser Industries, Inc.

While this tool is useful, it has been discovered that serious problems occur in measuring relatively low flow rates of high viscosity fluids, flow rates as low as those mentioned where spinner surveys cannot be used and where ordinarily the radioactive tracer log would commonly be used. Inaccurate -and completely misleading results are achieved when the fluid flowing is viscous or non-Newtonian. In one example, using the conventional Nuclear Flolog (TM) to measure flow rates of a viscous polymer/brine solution, an essentially zero flow rate measurement was obtained. Yet, other conventional logging measurements indicated a positive flow. In other cases, it has been found that a flow rate substantial lower than the true flow rate is indicated. For example, using a conventional Flolog (TM) tool in a wellbore fluid containing

1,000 ppm xanthamonas gum (fluid viscosity = 30 cp at 11 sec shear) a maximum flow rate in the casing was measured to be 58 barrels per day. However, the true flow rate as measured by a surface meter was 123 barrels per day. These difficulties clearly reduce the utility of radioactiv tracer flow logging tools.

SUMMARY OF THE INVENTION In accordance with the present invention, an improved method and apparatus for conducting radioactive flow logging in non-Newtonian fluids is provided which alleviates the above- mentioned problems. The invention comprises reducing the tracer injection velocity, compared to that required for Newtonian fluids, so that the tracer material is optimally positioned in substantially the center of the annular region of flowing non- Newtonian fluid between the tool and the wellbore casing. This may be accomplished by increasing the size of the tracer ejection ports, or by increasing the number of injection ports, or both. In part, the invention is based on the fact that the velocity distribution profile for non-Newtonian fluids is significantly different than the profile for Newtonian fluids.

BRIEF DESCRIPTION OF THE DRAWINGS FIGURE 1 schematically illustrates a radioactive flow logging tool positioned adjacent to a perforated casing interval. FIGURE 2 is a schematic representation of the fluid rheology on the velocity profile in a logging tool/casing annulus.

DETAILED DESCRIPTION OF THE INVENTION The present invention is an improved method and apparatus for generating an injection profile for non-Newtonian fluids flowing in a wellbore. The invention resides in a reduction of the ejection velocity of a radioactive tracer just sufficiently to place the tracer in substantially the center of the annular space containing flowing viscous wellbore fluids.

Referring now to FIGURE 1, a radioactive tracer flow logging tool 11 is schematically illustrated. The particular logging tool 11 is shown being used to measure injection profiles over an open hole completion interval, although obviously this is not a limiting factor. The tool 11 is shown suspended in a cased wellbore 12 through packer 13 from the surface by means of a wireline 14, and extends into an open hole completion in a reservoir 20 having a permeable wall 15. The tool 11 comprises two gamma detectors 16 and 17 in a vertically spaced relationship and radioactive ejector assembly 18 having ejector ports 19. Ejector assembly 18 contains a suitable radioactive dye, such as Iodine 131, which may be ejected by means of a suitable pump from the tool via ports 19. The general operation of the detectors and the ejector assembly is known, for example the Dresser Atlas Flolog (TM) and will not be further described. It should be noted that the present invention may also be applied to other logging tools having one or more radioactive tracer ejectors and having either one or two or more radiation detectors. Moreover, the present invention may be applied whether or not the tool is stationary, or moved up or down, to detect fluid flow following tracer injection (eg., the so-called "logging up" technique). The tool may be centralized or decentralized. However, if decentrali the radioactive tracer ejector must be directed toward the center of the wellbore. Conventionally, it has always been desired to eject the radioactive tracer at a maximum possible velocity in order to obtain increased dispersion. However, when ejecting tracer into a viscous non-Newtonian fluid, it has now been discovered that this gives completely undesirable results. This is because, with conventional ejector velocities as calculated for Newtonian fluids, the tracer material enters the viscous fluid as a discrete "blob" that is propelled into the annulus without dispersing into the flowing wellbore fluids and onto the casing wall where the fluid velocity is significantly lower (near zero) than the average

- ^y zE

fluid velocity in the annulus. Significantly, it has experimentally been observed that in non-Newtonian fluids the ejected tracer material remains undispersed, even though it passes through an annular region of high, velocity ordinarily sufficient to cause dispersion for ^" Newtonian fluids. Ejection of tracers into the wellbore under these conditions causes the flow rate in the flowing non-Newtonian fluid, as measured by relative movement past two detectors, to be much too low.

The present invention is for use in non-Newtonian fluids and requires that the tracer material be ejected at a low velocity compared to the ejection velocity into Newtonian fluids, such as brines. Referring to Figure 1, the ejector 18 is so constructed and arranged as to result in a tracer ejection velocity which is just sufficient to place tracer material substantially in the center of the annular region between the tool 11 and the open hole wall 15. In practice, for a non-Newtonian fluid flowing at a given rate, this means that the ejection velocity will preferably be at least 25% less than the ejection velocity into a Newtonian fluid flowing at the same rate as said given rate. Roughly defined, a Newtonian fluid is one whose viscosity does not vary with shear rate, while a non-Newtonian fluid varies with shear. Most enhanced oil recovery solutions exhibit non-Newtonian behavior.

The ejector assembly 18 may specifically be adapted to achieve the required velocity by: (1) increasing the size of the ejector ports 19, to preferably at least 25% greater cross- sectional area than conventionally used for Newtonian fluids; (2) increasing the number of ejector ports 19, preferably using twice as many ports as for Newtonian fluids; and/or (3) reducing the distance between the ejector assembly 18 and the first gamma detector 16, preferably reducing such distance to at least less than one-half (1/2) of the conventional distance.

To further identify tool performance in viscous fluids, tests were performed in a visual flow loop containing viscous

fluids. A tool was placed in a vertical flow loop containing 6 inch diameter and an 8 inch diameter plexiglass pipe. A solution containing xanthamoπas biopolymer in water with a viscosity of 30 centipo-Lse at 11 sec was pumped through the pipes. Food dye was substituted for radioactive tracer in the tool ejector assembly, and ejections of tracer were made using a conventional Dresser- Atlas Flolog (TM) tool having a single 0.025 inch ejector port spaced at 4.5 feet from the top gamma detector. This conventional tool provided perfectly accurate flow measurements using flowing Newtonian fluids, such as brine. Fluid flow rates of 100, 200, and 300 barrels per day of the viscous biopolymer solution were maintained in the pipe during the tests. It was visually observed that, regardless of the flow rate, ejected material was not dispersed in the flowing fluid, and in most cases the ejected tracer moved as a discrete volume (or "blob") through the annular region of high velocity to the wall of the pipe where.it flowed at a significantly lower velocity (near zero) than the average fluid velocity.

A better understanding of the present invention may be obtained by referring to Figure 2. The tool 11 is schematically illustrated in the center of a wellbore. The velocity distribution of a Newtonian brine is illustrated as compared to a pseudo- plastic xanthamonas biopolymer solution (a non-Newtonian fluid). A biopolymer solution such as is the one previously described having a viscosity of 30 centipoise, would exhibit the non- Newtonian velocity distribution. As used in Figure 2, is the tool radius, R is the radius of the tool to the center of the v annular space between the wellbore and the tool (or to the point of maximum velocity), R is the radius of the wellbore (the subscript c standing for casing, although the invention does not require a cased wellbore), and V z stands for the velocity distributio in the axial direction.

As may be seen, the velocity distribution for a non- Newtonian fluid is substantially different than that for a Newtonian

fluid. The distribution indicates the critical importance of properly ejecting the tracer material at a velocity which places the material at about R . The test suggested that in order to accurately measure a flow profile for flowing viscous fluids, the tracer ejection velocity must be adjusted so that the tracer will move closer to point "B" in Figure 2. Point "A" and point "C" have too low a velocity to carry the tracer material past the detectors. Moreover, and significantly, simply causing the ejected tracer material to pass through R (or point "B") will not achieve the benefits of the present invention.

Velocity profiles for Newtonian or non-Newtonian can be calculated from theory. Different components of the fluid velocity can be experimentally measured if the tracer is placed at different radial distances from the wellbore centerline. By adjusting the tracer ejection to place the tracer at specific radial distances, specific velocity components can be measured. Of special interest is the maximum fluid velocity at point "B" of Figure 2, which can be correlated with the average fluid velocity for any fluid reology. Appropriate fluid flow theory is found in A. G. Fredrickso and R. B. Bird, "Non-Newtonian Flow in Annuli", Industrial and Engineering Chemistry, volume 50, number 3, March 1958 (pages 347-352).

In order to accurately measure the flow rate for viscous fluids, the tracer ejection velocity must be reduced just sufficient so that the tracer material moves preferably to the center of the tool/casing annulus. While too high a velocity will place the tracer at the casing wall, which is a low velocity region; too low a velocity will place the tracer just next to the tool, another low velocity region. For the viscous fluids described previously, it was experimentally determined that satisfactory velocities could be achieved by using three 0.032 inch diameter ejector ports, instead of one 0.025 inch port. The distance from the uppermost detector to the ejector ports was 14 inches, rather than 4.5 feet.

Thus, placement of the tracer as desired can be achieved by a number of means. The volume of tracer ejected for a given number of ejection ports of fixed diameter can be varied, or the number of ports or the diameter of the ports can be varied for a fixed volume of tracer ejected. Multiple ports having different diameters can be used to place shots at different radial distances from the wellbore centerline, thus allowing simultaneous measurement of different velocity components. In addition, the ejector ports may be placed closer to the uppermost detector in order to more accurately and quickly detect the tracer velocity at low flow rates.

For the tool illustrated in FIGURE 1, the amount of fluid flowing from the wellbore into a given vertical interval of . reservoir is measured by the following steps: (1) placing the tool 11 at a stationary position at a known depth; (2) ejecting a volume of radioactive tracer (usually radioactive iodine, iodine 131, dissolved in water) at a low rate just sufficient to place the tracer in substantially the center of the annular region between the tool 11 and the wall 15; (3) monitoring the radiation intensity moving past the two detectors 16 and 17 placed a known distance apart; (4) using the computer travel time of tracer between detectors to determine fluid velocity; (5) from the measured fluid velocity and known dimensions of of the wellbore and logging tool, computing the fluid flow rate; (6) repeating steps (1) to (5) at various depths over which fluid flow into the formation is being measured; (7) by determining differences in fluid flow rates at various depths determining the amount of fluid entering the formation over various depth intervals.

Various modifications of this invention will be apparent to those skilled in the art without departing from the spirit of the invention. Further, it should be understood that this invention should not be limited to the specific embodiments set forth herein.

-^TR