You are on page 1of 6

SPE 68150

Novel Compact Oil/Water Separator Tested in Nimr


Larry Nnabuihe, SPE, Petroleum Development Oman (PDO)

Copyright 2001, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the SPE Middle East Oil Show held in Bahrain, 1720 March 2001. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972952-9435.

Abstract A novel compact oil-water separation concept using 2 hydrocyclones in series with recycle has been proven through laboratory testing and Nimr wellhead trials. The objective was to investigate the technical feasibility of using hydrocyclones at the wellhead to separate Nimr crudes prior to considering a downhole trial. With 'light' Nimr crude, good water separation efficiency was achieved with a residual OIW better than conventional separation. With more typical heavy viscous crude, similar separation efficiency was attained but OIW levels were significantly higher. The technology provides a viable option for separation at the wellhead, for example prior to reinjection in water flood projects or where surface infrastructure is flow constrained. Reservoir benefits through improved pressure support or sweep could make the business case for downhole separation more favourable; however, it poses challenges to downhole monitoring and well architecture. Background Producing reservoirs in Nimr area of South Oman contain viscous (400 cP), heavy (21 degAPI) oil underlain by strong aquifers. Field development is with a grid of horizontal producers, lifted by beam pump or Electrical Submersible Pump (ESP). The unfavorable mobility ratio of oil to water leads to rapid water breakthrough and production is required to high water cuts (98% BSW) to recover economic reserves. Increasingly large volumes of produced water result, with the 1999 rate of 150,000 m3/d, forecasted to double over the next 5 years.

The current favoured disposal method is re-injection at a central location into the aquifer leg, downdip of producing reservoirs. Injectivity is sustained by disposing under fracturing conditions. The capital and operating cost of water disposal is consequently significant owing to the high injection pressures involved. In-field dehydration either at surface or downhole using hydrocyclones offers a potentially attractive alternative. Produced water lifting; processing and re-injection might all be improved, reducing disposal costs. Following lab testing, [ref. 1], a surface hydrocyclone trial was carried out in order to investigate the technical feasibility of using a hydrocyclone to separate Nimr produced fluids. Two ESP-lifted wells were tested: a typical 400 cP heavy oil (Nimr B) and a lighter 30 cP oil in a potential remote water flood development (Al Burj). The trial was designed and executed by a multidisciplinary team from PDO, SEPTAR and Vortoil, who successfully bridged the technical and operational interface issues that were posed. Concept The hydrocyclone test unit comprises of a skid with 2 hydrocyclones in series with a recycle pump (Figure-1).
OIL BS&Wout WATER OIWout

WELL BS&Wwell

F1 F2

recycle

Figure 1: 2 stage hydrocyclone configuration

LARRY UBA. NNABUIHE

SPE 68150

The hydrocyclone unit can purify oil to a few percent water in oil and water to a few hundred parts per million independent of the inlet oil-concentration. It was envisaged that the level of separation efficiency will meet the water quality requirements for direct re-injection, whilst minimising the product flow. The technology when applied downhole or at the wellhead, will de-bottleneck the production system by creating ullage. In a similar manner, an early integration of the concept in a green field development plan could considerably reduce the inter-field flowline infrastructure and cost hence improves the project econimics. Figure-1 above illustrates the general concept where the oil production from a high BS&W well is fed into a dewatering cyclone (F1) operated such that the overflow contains oil as dry as possible whilst the underflow is fed to a de-oiling cyclone (F2). The operation under this mode is such that its underflow contains clean water. The overflow of the 2nd cyclone is recycled into the inlet feed to the unit for re-processing. In order for downhole dehydration to compete with deep water disposal, a significant reduction in water cut is required whilst maintaining an oil in water level in disposal streams, at least as good as current practices (approx. 200 ppm). Cyclone performance is presented in this article as: i. Water (Separation Eff) = BS&Wwell BS&Wout BS&Wwell ii. Residual oil in water in the outlet water stream, OIWout Justification The poor separation efficiency of Nimr type crude has been attributed to the high crude viscosity and the associated formation of tight emulsion. It was generally believed that it would be almost impossible to achieve a good separation efficiency using a 2-stage hydrocyclone. The following benefits were envisaged in the event of a successful trial: Provision of source water for waterflood and water injection projects. The option of shipping produced water back to the individual fields after separation in Nimr Production Station (NRPS) would be eliminated. As a result, the Value Investment Ratio (VIR) of future water injection projects will be greatly improved from savings in laying water pipelines and or drilling water supply wells. Creation of ullage in the production system (pipeline and vessels) Reducing the risk of pipeline corrosion as less water will be introduced into the system. Confirmation of the feasibility of downhole oil and water separation in the Nimr area. Note: downhole conditions are more favourable to the separation process due to lower crude viscosity resulting from higher downhole temperature as well as avoiding the

formation of emulsion across the ESPs and at the wellhead chokes. Reduced cost and exposure for Deep Water Disposal (DWD) expansion Reservoir pressure support

Description of Trial Equipment

Figure 2: Skid-mounted two-stage hydrocyclone unit Hydrocycloes are conical devices that make use of centrifugal forces which are develop due to the geometric design of the system to separate the light phase from the heavy phase. Figure-3 below illustrates how a hydrocyclone operates. The equipment has no moving parts. The Nimr 2stage hydrocyclone trial was carried out with a new purpose built test skid. The test skid /pilot unit (Figure-2) comprises two standard hydrocyclones (production K-liners) adapted to allow easy connection and mounted in series, with an option for single stage operation. The Vortoil hydrocyclone separator is a liquid/liquid unit capable of separating immiscible, insoluble liquid/liquid mixtures utilising a number of specially designed liners. The flow capacity of one hydrocyclone is low compared with the total flow of a well hence are arranged in series. The well production fluid enters the K-liner unit (hydrocyclone) through a tangential involute inlet (orifice). The flow is directed into a vortex without disrupting the reverse flowing core. As the flow is forced down the liner it takes up a helical form along the liner inner wall. It is accelerated in the concentric reducing section to the high velocity required to create the strong centrifugal forces that promote rapid separation. The velocity is maintained along the liner, and frictional losses overcome, by a gradual reduction in cross sectional area throughout the tapered section.

SPE 68150

NOVEL COMPACT OIL/WATER SEPARATOR TESTED IN NIMR

L IG H T O VER FLO W

FEED IN L E T

A C C E L E R A T IN G H E L IC A L FLO W PATH

R E D U C IN G S E C T IO N

L IG H T E R P H A S E M IG R A T E S TO CENTRAL CO RE

Due to the high viscosity of the Nimr type crude, the production stream contains a high quantity of very small oil droplets (less than 8 micron) held in tight emulsions. As a result, it was considered necessary to enhance the system performance by the injection of chemicals. The chemicals are designed to increase the size of the oil droplets and, hence, aid separation. In liaison with the PDO Field Chemist, an optimum dosage was established for the Baker Petrolite RP-6392. However, before considering the use of the chemical, the upstream performance had to be optimised. A Baker Petrolite chemical injection skid (seen in the right hand side of Figure-2 and the demulsifier chemical were supplied. Candidate Well Selection Criteria The following criteria were used in the selection of the candidate wells: Stable production records without surging tendencies High rate ESP wells with approx. 0 kPa annulus pressure to facilitate dumping of separated fluid back into the well annuls. High BS&W producers (BS&W > 70%) Wells with tubing head pressure between 15 - 20 bar. High and low crude viscosity wells Two candidate wells emerged from the final screening viz. 1. Nimr-234 with a gross production of 531 m3/d 84% BS&W viscosity (400cp, 21 deg API) 2. Al Burj-26 with a gross production of 900 m3/d 79% BS&W viscosity (30cp) Well Preparation and Engineering Modifications Engineering modifications were carried out on the flowline at the wellpad areas in order to accommodate the hydrocyclone skid mounted unit and the chemical injection skid as follows. Construction of tie-in points from where split flow could be diverted to the hydrocyclone unit. Provision for the separated fluid to be dumped back into the well annulus. Connection of chemical injection skid at the wellhead with 1/4" pipe connection. Electricity supply from the Centrilift ESP power supply panel Instrumentation connections for the coriolis meter etc.

BACK PRESSURE D E V IC E

DENSE PHASE U N D E R FLO W

Figure-3: Schematic of a hydrocyclone unit The denser fluid (water) moves to the wall of the liner and is removed at the downstream fluid outlet (underflow). The less dense fluid (oil) is drawn into the low-pressure core and, by applying a back pressure to the outlet, flows back up the liner to be removed at the upstream outlet orifice (overflow). The vortex and reverse flowing core extend down the tail section of the liner, increasing the residence time and allowing smaller, slower separating droplets to migrate to the core. The total residence time in the liner is a few seconds. The centrifugal force generated within the liner is in the order of 1000G, which makes the unit insensitive to motion, and orientation. A range of overflow sizes (1.5, 2, 2.5, 3.5 and 4.5 mm diameter) was available for different overflow rates for each of the K-liner units. The K-liners are rated for a flow range of between 0.5 and 4.1m3/h dependent on the pressure drop and interconnecting pipework. The skid unit is also equipped with a pump which recycles the second stage overflow back into the inlet stream of the 1st stage hydrocyclone for re-processing. The pump is a double plunger, positive displacement type and is nominally rated for 35 barg with a flowrate of 0.013 - 0.21m3/h. Pressure gauges (PI-01 PI-06) were fitted on the inlet and outlet of the underflow/overflow of each of the Kliners. Micro-Motion Coriolis flowmeters were supplied by SIEP (Shell International E&P), The Netherlands for the measurement of temperatures, densities, mass flowrates and volume flowrates. Sample points were provided on the inlet and outlet overflow and underflow of each hydrocyclone unit.

Flow Measurements The inlet flow to the hydrocyclone is usually determined from the measurement of the pressure drop across the hydrocyclone i.e. between the inlet and water outlet streams. A field proven relationship exists between this pressure drop and flow. In order to verify these flows and gather as many data as possible during the trial, 3 coriolis massflow density meters were installed in the test unit. The individual meters directly measure the mass flow and density of process fluids. The coriolis meters were connected at a safe distance

LARRY UBA. NNABUIHE

SPE 68150

from the unit due to their sensitive nature. The table below shows the location of the coriolis meters within the skid as well as the maximum flow rates: Meter No. Location within skid FR1 2nd stage (de-oiler) cyclone overflow FR2 2nd stage (de-oiler) cyclone inlet FR3 1st stage (pre-separator) cyclone Range 0.5 m3/hr 3m3/h 5 m3/h

The RFT9739 transmitters were fitted on a temporary installation during the trial. Data collected from the meters were used in evaluating the project. Trial Layout and Hook-up Details The hydrocyclone unit was connected to the tie-in point provided on the flowlines downstream of Nimr-234 and Al Burj-26 wellheads as shown in figure-4 below. The gross flow rates from the wells are in excess of the maximum throughput of the test unit (4.2 m3/hr). The maximum rate attained during the trial was 2.2 m3/hr, possibly due to the inter-connecting pipework in the skid. As a result, only a split flow could be diverted through the unit. However, the tie-in point was inadvertently constructed at the top of the flowline in the 2 candidate wells. At NMR-234, the feed pipework to the test skid from the flowline was equipped with a ball valve located on the top of the main flowline (Fig-4). A tie-in point at the side should ideally give a more representative sample of the flow. However, in this situation where the wells are producing via ESPs, fluid leaves the pump downhole under turbulent flow condition. This coupled with further agitation at the wellhead chokes and bends leaves no room for water dropout within the few seconds of flow in the wellpad area. Both the ESP and the wellhead chokes will tend to aggravate the formation of tight emulsion. At Al Burj-26 the feed pipework to the test skid from the flowline was via a needle valve also located on the top of the flowline. As part of efforts to ensure that representative samples are collected, a sample probe was installed inside the main flowline at the tie-in point. (See Figure-5). Figure-4
Trial Layout in NM-234 & ABJ-26
Electricity supply panel

Figure-5: Hydrocyclone tie-in point showing sampling probe (ABJ-26 wellpad area). The 4 outflow points from the unit (overflows and underflows) were connected into single manifold pipes which disposed the separated fluids into the well annulus. Each of the termination points has valves located within the unit to control flowrates. The decision to dispose the separated fluid via the annulus was part of cost saving measure as it was noted that the cost of renting a storage tank and disposal vacuum truck was beyond the budget cover of the project. Sampling & Data Gathering Separated water, dry-oil and oily-water mixture samples were taken from the various sampling points on the test skid during each run. The samples were taken from the inlet, the overflows and underflows of each hydrocylone stage. The samples were analysed in the Nimr Production Chemistry Laboratory for either BS&W or oil-in -water (OIW) depending on the quantity of oil in the sample. Other crude properties derived from the analysis include the crude density and viscosity values. Separation Modes Three separation modes were investigated: Single hydrocyclone separation - De-oiling mode Two hydrocyclones in series without recycle Two hydrocyclones in series with re-cycle

Wellhead

Sampling point Coriolis meter Overflow/Underflow disposal lines to annulus Hydrocyclone Chemical injection skid Location valve 34 m Tie-in point with sampling probe (Split flow) Flowline
1m

30 m

SPE 68150
st

NOVEL COMPACT OIL/WATER SEPARATOR TESTED IN NIMR

Single Hydrocyclone (1 Stage De-Oiling Mode) The hydrocyclone was initially run in single stage de-oiling mode, to determine the optimum settings for reducing water cut in the oil (dry) outlet stream:
Well Well BS&W (%) Oil overflow BS&W (%) Water separation efficiency (%) NM-234 84 20 75 ABJ-26 79 14 83

Figure-6. Photograph of optimised samples taken at Al Burj26.

Two-Hydrocyclones in Series, Without Recycle Using these optimised settings, a second stage of processing was introduced with the objective to minimise residual oil levels in the water outlet stream with the following results:
Well Well BS&W (%) Oil overflow BS&W (%) Water separation efficiency (%) Water underflow OIW (ppm) NM-234 84 18 78 5000 ABJ-26 79 16 80 75

From left to right: 1. Dry-Oil Outlet Sample (approx. 14% BS&W) 2. Package Inlet sample 3. 1st Stage Underflow Sample 4. 2nd Stage Overflow sample 5. Clean Water Outlet (75 PPM) Figure-7 below shows summary results of the Nimr hydrocyclone trials.
Summary Results
ABJ-26 NM-234

Two-Hydrocyclones in Series, With Recycle The regulated recycle was then added to further improve performance as shown in the table below.
Well Well BS&W (%) Oil overflow BS&W (%) Water separation efficiency (%) Water underflow OIW (ppm) NM-234 84 14 82 5000 ABJ-26 79 12 87 60

100000
outlet OIW (ppm)

10000 1000 100 10 1 0

Trial Evaluation
Al Burj-26 The 30 cP Al Burj oil was successfully handled by the hydrocyclone. A water separation efficiency of over 80% was achieved, although this showed little improvement when a second stage of separation or recycle was introduced. The OIW level of 46 - 100 ppm improved slightly with the introduction of a recycle, and was significantly better than the target value of 200 ppm. Nimr-234 The 400 cP Nimr oil proved a more demanding test. Water separation efficiency improved with the introduction of a second stage and a recycle from 75% to 78% and 82% respectively. OIW however, remained disappointing at 5000 ppm. Demulsifier injection was not found to have any significant impact on performance, probably due to the limited surface residence time. The photograph in Figure-6 shows an optimised sample taken in Al Burj-26 with 75-PPM OIW in the clean water sample, while Figure-7 summarises the results of the hydrocyclone trials for 2 different Nimr crudes.

20 40 60 80 water separation efficiency (%)

100

Figure7: Summary Results Scope for Implementation The concept of compact oil water separation using hydrocyclones at a wellhead has been successfully demonstrated. Hydrocyclones provide a viable option for local or infield water separation prior to reinjection in waterflood projects and to reduce the impact of flow constrained surface infrastructure. The residual OIW obtained from heavy Nimr crude, however, makes disposal economics slightly less favourable as compared to current deep-water disposal practices. It is however envisaged that the high OIW content of separated water can be improved further under downhole separation conditions (DOWS) in a pull-through configuration (i.e. separation upstream of the ESP) as the tendency to form tight emulsion across the ESP and wellhead chokes will be avoided.

LARRY UBA. NNABUIHE

SPE 68150

Additional benefits from downhole dehydration may arise if reservoir sweep and pressure support can be realised. Waterflood projects in fields with favourable crude characteristics could benefit from this emerging technology of compact oil/water separation specially when applied infield at the wellhead or downhole. Key areas for technology development in a downhole application in Nimr are downhole measurement and control of the performance of hydrocyclones and associated ESPs in order to minimise OIW levels and maximise ESP run-lives. In addition, well architecture challenges are presented by the need to provide a cost effective reinjection zone, sufficiently isolated from the horizontal (openhole) producing zone. Learning Points The following learning points were established after demonstrating the novel compact separation in Nimr field: 1) The technology offers a cheap/economic alternative source of injection water in waterflood projects through infield dehydration (i.e. using surface hydrocyclones at the wellhead) as against dedicated water supply wells. 2) Application of the technology downhole will translate to Reservoir Engineering benefits accruing from improved sweep and RF, which could make subsurface field development project-economics more robust. 3) Th compact separation technology at wellhead or downhole could significantly reduce the inter-field flowline infrastructure and debottleneck flow-constrained process facilities. 4) The favoured environmental disposal method of reinjection into the aquifer leg of the producing reservoir is demonstrated through downhole oil and water separation. 5) Application of the technology downhole using compact hydrocyclones is an attractive disposal option since water lifting, processing and disposal cost might all be reduced.

Acknowledgement The author wish to thank Petroleum Development Oman (PDO) and the Ministry of Oil and Gas (MOG), Sultanate of Oman for permission to publish this paper. The joint effort and enthusiasm of the Nimr Team, PDO Service Providers and Contractors (including Baker Process, Rugby, UK - the manufacturer of the hydrocyclone unit) is acknowledged. The author also thank the following colleagues for their dedicated efforts and comments: Dave Munro (PDO, Oman), John Barron (PDO, Oman), Paul Verbeek (SEPTAR, Holland) and Dave Bennett, Field Service Engineer, Baker Process (Rugby, U.K.)

References
1. "Tandem hydrocyclone with recycle A novel approach to oilwater separation downhole or at wellhead" by P. G. Witte and P. H. J. Verbeek (SIEP) 2. "Subsea and Downhole Separation SystemsThe latest advances" by T. kjos, Read Group et al. 3. "Joint Industry Development of the Downhole Oil Water Separation System-Field Case Study" by Peter J. Schrenkel REDA, Camco International Company.

You might also like