A liquid-liquid centrifugal separator has been modified for possible application as a downhole method for separating crude oil from produced water. Centrifugal separators of various sizes (from 2-to 25-cm rotor diameter) have been built and operated over the past decades at various U.S. Department of Energy facilities. These units have several characteristics that make them attractive for downhole applications, including excellent phase separation, reliability in remote applications with .20,000 h of operation prior to maintenance, and the ability to handle high volumetric throughput with a very low residence time. These separators consist of a rotating cylinder in which the two phases are separated and a stationary housing that collects the separated streams. This paper discusses some of the aspects of the alterations required for downhole operation. Specifically, we discuss modifications of the exterior housing allowing for greater flow through the system. The system presented here improves the performance of a standard separator by 140%.

No abstract

Large storage tanks may require mixing to achieve homogeneity of contents for several reasons:  prior to sampling for mass balance purposes, for blending in reagents, for suspending settled solids for removal, or for use as a feed tank to a process. At Oak Ridge National Laboratory, mixed waste evaporator concentrates are stored in ∼190-m3 (50 000-gal) horizontal tanks, about 3.7 m (12 ft) in diameter and 18 m (60 ft) in length. This tank configuration has the advantage of permitting transport by truck and therefore fabrication in the shop rather than in the field. A survey of the literature revealed no information on mixing large storage tanks with length-to-diameter ratios greater than 2. Jet mixing experiments were carried out in two model tanks:  a 0.87-m3 (230-gal) Plexiglas tank that was ∼1/6 linear scale of the actual waste tanks and a 95-m3 (25 000-gal) tank that was about 2/3 linear scale of the actual waste tanks. Mixing times were measured by the use of a sodium chloride tracer and several conductivity probes distributed throughout the tanks. Several jet sizes and configurations were tested. In the 0.87-m3 tank, jet diameters of 0.016, 0.022, and 0.041 m (0.62, 0.87, and 1.61 in.) were used. In the 95-m3 tank, jet diameters of 0.035 and 0.049 m (1.38 and 1.93 in.) were used. One-directional and two-directional jets were tested in both tanks. Mixing times for each tank were correlated with the jet Reynolds number and for the two tank sizes using the recirculation time for the developed jet.

This report has been reproduced directly from the best available copy. Available to the public from the National Technical Information Service, U.S. Department of Commerce, 5285 Port Royal Rd., Springfield, VA 22161. I IThis report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implii, or assumes any legal liability or responsibility for the accuracy, completeness. or usefulness of any information, apparatus, product, or process disclosed. or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily consti-M e or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The 2.3. 4. 5.6. 7. 8.Photograph .3 .4 . .6 . .8 . .10 . . .14 . Kaolin clay was used as a surrogate sludge in both of these tanks, and a chemical sludge that -xi was formulated to represent the constituents of the sludge in the MVSTs was also used as a surrogate sludge in the 0.87-m3 tank.The tests performed in the 0.87-m3 tank with kaolin and the chemical surrogate indicated that the two materials behaved similarly with respect to mobilization and that they -could be modeled by comparing the effective cleaning length (ECL) with the product of the nozzle diameter and the jet velocity @Yo).Mobilization experiments were conducted in the 95-m3 tank to obtain scaleup data.Bidirectional discharge nozzles were installed at three locations along the length of the tank.Suction legs were also installed at three locations h the tank. This arrangement provided versatility in conducting mixing and mobilization experiments. The depths of the suction legs were varied to determine whether their location had any effect on mixing.The mobilization data obtained fiom the 95-m3 tank were also fit to the previously mentioned ECGDY, model. It was determined that the model could be used to predict the mobilization effort within the range ofDYo values tested. Up to 81% of the kaolin clay was mobilized and mixed; however, the mobilization was limited by use of an existing pump with a capacity of 12.6 L/s (200 gal/&). It is believed that a higher nozzle velocity would mobilize more of the sludge.A comparison of the positions of the discharge nozzles along the length of the tank indicated that the nozzle position did not affect the quantity of kaolin mobilized. This finding was not surprising because the ECL of the nozzles was shorter than the distance between the nozzle and the end of the tank. A difference would be expected if the ECL was -longer than the distance between the nozzle and the end of the tank.Experimental data indicate that the depth of the suction lines did not affect the quantity of sludge mobilized, but evidence showed that the contents mixed faster when the suction leg was eithe...

A liquid-liquid centrifuge has been tested for possible application as a downhole method for separating crude oil from produced water. Centrifugal separators of various sizes (from 2-to 25-cm rotor diameter) have been built and operated over the past three decades at various US Department of Energy facilities. These units have several characteristics that make them attractive for downhole applications, including excellent phase separation, reliability in remote applications with >20,000 h of operation prior to maintenance, and the ability to handle high volumetric throughput with a very low residence time. In these studies, water-to-oil feed ratios of 10:1 to 1:19 were tested with a light Gulf of Mexico crude oil, and the separator operated efficiently for the full range of feed ratios. Air was added to the oil stream in one test to model the effect of gas in the oil. Air additions up to 20% of the ORDER REPRINTS feed flow rate (the maximum tested) did not have any impact on the performance of the separator. The separator also effectively processed a very viscous North Sea heavy crude oil. The heavy crude was used to determine the effect of higher temperatures on the performance of the separator. Increasing the temperature of the oil and water feed stream improved overall performance and decreased the concentration of oil in the water discharge stream.