OTC 19484

Kikeh Development: Sand Control Selection, Design and Implementation of ESS Tamara Webb, Jusni Omar, Rusty Desormeaux, Pat Moran, Kasim Selamat, Murphy Sabah Oil Co., Ltd., Steve Beare, Colin Jones, Kevin McWilliam, Weatherford International

Copyright 2008, Offshore Technology Conference This paper was prepared for presentation at the 2008 Offshore Technology Conference held in Houston, Texas, U.S.A., 5–8 May 2008. This paper was selected for presentation by an OTC program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect any position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Offshore Technology Conference 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 OTC copyright.

Abstract Discovered in 2002, the Kikeh field is located 120 km northwest of the island of Labuan, offshore Sabah, East Malaysia. The water depth at Kikeh, Malaysia’s first deep water development is 1,320m. The field development consists of a dry tree unit (Spar) and two subsea production hubs, all tied back to an FPSO. Two of the three productive intervals in the development were determined to require sand control, representing about two-thirds of the field’s production. This paper documents the selection and design process for a multi-zone sand control completion with zonal isolation to meet the requirements of both the open hole oil producers and the cased and perforated injector wells on Kikeh. The primary objectives of the system being: • Maximize productivity/injectivity • Provide efficient sand retention • Minimize cost and time • Design for well life to minimize workovers and interventions • Have a large bore to maximize the capability for future selective completions

After extensive testing and evaluation was performed on the seven appraisal wells across the field, including well tests with sand control, expandable sand screens (ESS) were selected as the primary sand control method. The evaluation, coupled with extensive local ESS installation experience in both Southeast Asia and Malaysia, gave the confidence that the installation objectives would be met. This paper describes the evaluation of sand control methods and why ESS was determined to be the best choice for the Kikeh field development. It also describes the details of each ESS type (water injector, producer, and intelligent completion) and the testing required to qualify each system. Swelling Elastomer Packers were used to provide zonal isolation in both the injector and producer wells. The qualification and selection process for these packers is also discussed. Significant application learnings will be discussed and, lastly, well productivity and skin results will be reported. Introduction Murphy discovered the Kikeh field offshore Sabah in 2002. A seven well appraisal program was conducted, which included well tests (two) with sand control. It was identified at this stage that one of the many challenges that the Kikeh field posed, was to find a sand face completion that could control the relatively unconsolidated, and poorly sorted sands present in Kikeh’s geology. The system would have to be truly flexible in its multi-zone capabilities, to safely and efficiently exploit the highly laminated sands in Kikeh, while isolating the shales, throughout the expected 20-year life span of the field. As part of the design case, the system would have to provide a suitable interface for a future selective completion allowing for reservoir management as required.

2 OTC 19484

From an economics standpoint, one of the main goals was to find a system that could be safely and efficiently installed with the minimum personnel and equipment. To realize this goal, it was identified from the outset that the manufacture, preparation and installation of equipment would have to be closely controlled by robust quality systems and procedures. Preceding the Kikeh well tests utilizing ESS technology, Murphy had previously completed the West Patricia field offshore Malaysia with the ESS technology, which yielded good productivity results and gave Murphy confidence in the ESS solution, as the optimum system for Kikeh. Reservoir Summary Kikeh reservoir structure is composed of a multi layered sandstone formation, which is highly laminated with shales throughout. It has seven distinct horizons – H110; H115; H120; H130; H136; H145 & H150 (Figure 1). These reservoir groupings have been chosen such that reservoir pairs have similar rock properties, reservoir fluid properties, and are in the same pressure regime. Following extensive testing and modeling, it was identified that the H110 through to the H136 horizonswould require sand control. Sand Control Selection Process It was intended that all production wells in the 120/136 horizons would be completed with an open hole sand control solution, with the non-hydrocarbon bearing sands and shales being isolated by way of open hole annular isolation. Open hole completions presented the best method to open laminated sections to flow in 120/136 packages. The reservoir modeling completed, it identified that the Kikeh reservoir would require pressure support of the formation through water injection from day one of production. This would also serve to water flood (sweep) hydrocarbons out of the pore space and into the production wells. Considering this, it was concluded that the injectors would require to be completed in a cased and perforated wellbore; this was due to concerns with shale sections within the reservoir and instability with injected water over a long time. However, during shut-in periods, cross flow and water hammer effects cannot be controlled at the surface, thus large pressure differentials can occur causing sanding, and long-term weakening of sands from water dissolution, meaning that sand control would be required in the injector wells. Based on the above, the selection process began to identify the optimum sand control systems for the producer and injector wells. When selecting and designing a sand face completion for a project as big as Kikeh, there are many factors that need to be considered, and prioritized. Some are industry generic, and others field specific. For Kikeh, the first priority was to find a system that would maximize production and injection, while providing the required sand retention. Throughout the early stages of the selection process, various sand control solutions were investigated. Among them were open hole gravel-pack, frac-pack, stand-alone screen and Expandable Sand Screens. Open Hole Gravel-pack: One of the drawbacks with a gravel-pack system was the difficulty of obtaining 100 percent gravel-pack placement in wellbores containing large amounts of shales with water based fluids. The water-based fluid also posed emulsion issues with the Kikeh crude. Compared to ESS, it posed a higher cost option, and was operationally more complex. The gravel-pack completion would also restrict the internal diameter of the wellbore, and isolation between intervals within a co-mingled sequence could be extremely difficult. Frac-pack: Frac-pack posed disadvantages on high cost, and being typically operationally complex, with a low level of local experience. Again it restricted the ID of wellbore. As with the gravel-pack, there would be issues with the emulsion characteristics of oil with various fluids, and isolation of the intervals would be difficult to achieve. Stand-alone Screens: The major disadvantage that stand-alone screens posed was that they provide no support to stabilize the wellbore. Sand production will fill the screen annulus, and reduce near wellbore permeability and increase the skin. As with the previous options, the screens will leave a restricted wellbore ID, and would pose difficulties for future repairs and intervention.

Figure 1: Kikeh Main Field Area

OTC 19484 3

ESS: The ESS system was found to be the optimum solution to meet all of the following requirements: • Maximize productivity/injectivity. Industry data demonstrates low skins. Further to this Murphy has had a

successful track record with ESS in their West Patricia field, which backs up this data. • Provide efficient sand retention. The Kikeh sand size distribution is poorly sorted, and sand retention tests

concluded that non-compliancy of the wellbore would result in high delta P. The ESS system compliantly expands to the borehole, providing support, accommodating well irregularities and restricting sand movement.

• Minimize cost and time. Cost analysis completed showed that the ESS yielded a lower cost compared to gravel-pack and frac-pack. The installation of the ESS system is relatively simple compared to the other systems, requiring far less equipment and personnel. The ESS system chosen could be deployed and expanded in a single trip, which provided a huge saving in rig time, which is crucial when rig rates are at an all time high.

• Design for well life to minimize work overs and interventions. A critical design criterion for the ESS application has been conducted, and confirms that the screen withstands long-term production and depletion scenarios.

• Have a large bore to maximize the capability for future selective completions. The ESS renders a large ID after expansion, and does not restrict deployment of production packer for selectivity or smart application. The fixed ID of blank pipe provides guaranteed ID for packer/seal setting.

Fluid Selection Oil Producers: Drill in fluid selection process was focused heavily on oil-based mud solutions due to reservoir sensitivity issues as well as successful history of drilling Kikeh reservoirs with OBM fluids. The reservoir drill in fluid and the filter cake laid down during drilling had to be compatible with the 150 µ expandable screens selected for this completion. Besides issues related to completion compatibility, other considerations taken into account during the fluid design and selection process included;

• Full compatibility with the reservoir • Efficient rheological profile required to minimize circulating densities, and promote adequate hole cleaning, taking

into account bottom hole temperature, as well as the low temperature environment of the mud line (39 Degrees F). • Properly inhibit interbedded shale stringers within the sand zones • Maintain adequate bottom hole pressure with no tendency for solids settlement/sag • Effective clean up of filter cake to achieve low skin completions

To meet these initial design requirements, a synthetic-based invert emulsion fluid was selected, which utilizes a treated micronized barite (TMB) slurry to achieve density requirements, and maintain full compatibility with the completion assembly. The utilization of this weight material, combined with sized calcium carbonate (sized specifically for the Kikeh reservoir), provided the fluid system with a solids package which would minimize any potential for formation damage by providing a filter cake which would remain external to the formation face, and minimize invasion into the reservoir. The remaining components of the fluid, emulsifiers, viscosifying agents, brine phase, etc, were selected to maintain the non-damaging characteristics, the compatibility with the completion assembly, and provide the required fluid properties for the drilling phase of the completion process. Water Injectors: The plan for the injectors was to use a filtered and treated calcium chloride brine system, weighted to provide a 300-500psi overbalance to the formation. However, due to the reservoir properties, it was identified in the planning stages that once the wells were perforated, the fluid losses to the formation could be extreme. The major issue with severe fluid losses is primarily maintaining well control, throughout operations. So the challenge was to find a loss control pill that could be spotted after perforating, and which could control losses until the sand control solution is in place, and a mechanical barrier can be installed as part of the completion. Extensive testing was undertaken to evaluate various solutions to confirm compatibility with the formation, sand control solution and relevant fluids. The pill that was chosen is a temporary blocking gel prepared with a low-residue cellulose polymer and a delayed cross-linking agent to form a clear, rigid gel. The delayed cross-linker allows placement downhole prior to cross-linking. The results of the testing indicated that the pill was reasonably stable and remained fully cross-linked over a four day period. Testing also identified a suitable breaker that could be spotted should there be a time requirement for the pill to be broken. Zonal Isolation Selection One of the requirements for Kikeh was to provide a zonal isolation solution as part of the sand face completion. This would allow for future selective reservoir management. The challenge was to find an isolation system that had a low installation risk, and did not require a dedicated trip to install or activate. Swelling Elastomer packers are a new and emerging technology, which uses the absorption of fluid into the rubber element, causing swelling of the elastomer, which will seal the annulus around the pipe in the well bore. Once the rubber contacts the well bore, as it continues to swell, it builds up a differential capability. There are a number of variables that affect the swell time and differential capability, temperature, hole size versus rubber OD, rubber volume, and fluid properties. Moving forward it was decided to conduct a qualification process, to identify and select the optimum solution for both injector and producer well types. The relevant well and fluid data was passed to the chosen swell providers, who completed simulations to confirm that their packers could meet the requirements.

4 OTC 19484

Open Hole Producer Cased Hole Water Injector

Running Fluid Oil Based Mud CaCl Brine

Well Temperature 140F – 175F

Well Fluids Kikeh Crude Treated Seawater

Hole Size 8.6” 9.625” (8.535” ID)

Base Pipe 7.625” 29.7# FJL

Metallurgy L80 13Cr L80 1%Cr

Required ∆P 1600psi 3000psi Table 1 – Well & Fluid Information

On completion of the simulations, it was identified that there were two possible provider, of oil swelling packers for the producers, and a single supplier of water swell packers for the injectors. Fluid samples were provided to allow fluid compatibility testing to be completed. This would confirm the following crucial criteria:

• Swell time in the running fluid, if the elastomer was to swell quicker than predicted, it could result in the string becoming stuck off depth during deployment

• To confirm the elastomer will swell adequately in well bore fluids • To confirm that all fluids that contact the elastomer will not affect it adversely • To provide accurate data for specific predictions and elastomer property selection

Testing was completed by immersing Elastomer samples in the required fluid. The sample is then inspected and measured at regular intervals; the results are recorded, and extrapolated to simulate results for Kikeh. Oil Swell Packers: For the oil swell packers, one of the providers had two distinct advantages over the other. First, the elastomer is built up with a three-layer construction process, which delays the start of swelling, allowing the system to be deployed to depth. The second factor was that at the time of testing, the swell provider had installed more than 400 packers. As can be seen from the differential pressure profile below, the required 1600psi differential requirement could be attained up to a 8.68” hole diameter (Figure 2).

Figure 2 – Oil Swell Packer Differential Pressure Profile

The swell time is another important factor which is greatly affected by wellbore temperature. Even with the delayed swell technology, the packers will start to swell upon entering the well bore. The timing needs to take into account the time not only to deploy to depth, but also the time to recover the string should a problem arise during deployment or hanger setting.

Contingency Installation Timings

Item Description (Task) Hrs : Min

1 Make Up & Rih 5.5" ESS Assembly 8.00

2 Make Up Liner Hanger Assembly 1.00

3 Make Up work string & Trip In Hole 10.00

4 GR-CCL Correlate on Depth 4.00

5 Contingency Time at Depth 8.00

6 POOH and Lay down completion 20.00

Hours 51:00

Days 2.13 Table 2 – Contingency Installation Timings

OTC 19484 5

As can be seen from the Oil Swell Packer Simulated Swell Times (Figure 3), the contingency installation timing of 2.13 days was safely within the limit because the estimated time to contact a gauge 8.5” borehole was approximately five days. Considering the analysis of all the above data, it was decided to proceed with the above oil swell packers.

At the time of testing, the plan for Kikeh was to run 10 open hole ESS producer wells, one of which would be completed with an intelligent completion. The intention of the intelligent completion was to allow for independent zonal production, which would enable reservoir allocation. Following the installation of the ESS and the intelligent completion, the plan for this well was to perform a clean up and test, which would prove the differential capabilities of the packers in situ. The swell time calculations would be critical in ensuring that the packers had swollen sufficiently in time for the welltest, according to the simulations completed (Figure 3), the time for the packers to seal in was 8.5" & 8.68” was 2.7 days and 10 days respectively at 170 F. Based on this, it was decided to run 8.25” x 9m element packers, between the zones to be tested on

this well. The longer packer element allows for a greater differential capability. All the remaining packers for the producers would be 8.25” x 5m element. Water Swell Packers: The water swell packers posed a slightly different challenge, in that there was a very limited industry run history at the time of selection. However, the only supplier had completed a very thorough testing and qualification process. The injector scenario was also easier in that the cased hole well bore is a known ID (8.535”). One important piece of information that was gained from the fluid compatibility testing was the elastomer would swell extremely slowly while in the calcium chloride (CaCl2) completion brine. Tested figures showed, 0.08” growth over five days. The benefit of this was that the completion string could be safely deployed to depth, without risk of the packers swelling and causing the string to get stuck off of depth. As can be seen from the Water Swell Packer Expansion Chart (Figure 4), the required 3000psi differential requirement could be achieved in the 8.535” casing. To meet this requirement, it was required to run 3m x 8.25” elastomer elements, on the 7.625” base pipe. Considering the above information and analysis, it was decided to proceed with ordering the water swell packers for the injectors.

Figure 4 – Water Swell Packer Expansion Chart

Geomechanical Modeling For an expandable sand screen completion, one of the key areas must need to be addressed to ensure long term reliability is formation loading on the screen. An expandables wellbore stability model (EWBS)(REF) was used to calculate the deformation of the screen under the loading condition expected during the life of the field. The main important inputs to the modeling are the strength of the ESS, the strength of the rocks, the in-situ stresses and the drawdown and depletion. The strength of the ESS has been determined in large scale testing and FEA modeling. The rock strength was taken from detailed triaxial testing of cores from the appraisal wells. The rocks had a relatively low UCS around 900psi, with friction angles down to 18°.

Figure 3: Oil Swell Packer Simulated Swell Times

6 OTC 19484

Tectonics in the region are complex, so normal, thrust and strike slip faulting were all modeled. Further data honed in the models to a thrust fault regime. The results of the EWBS model (Figure 5) show approximately 10 percent reduction in wellbore ID for the worst case scenario. This is well within the deformation limits derived from large scale testing where up to 30 percent deformation was achieved without any loss of sand control.

5.00

5.50

6.00

6.50

7.00

7.50

8.00

2000 2500 3000 3500 4000 4500 5000 5500 6000Wellbore Pressure (psi)

ESS

ID (i

nche

s)

Wellbore Diameter Screen Yield500psi Overbalanced Reservoir Pressure20% Deformation Limit Expected Final Reservoir Pressure

Figure 5 - EWBS Model Results

Weave Selection A laboratory study was performed in 2004 to select the appropriate weave aperture for use in the Kikeh field. The sand samples used for retention testing were from two sources: 1st study performed for Kikeh H145 core (samples K, N & O) and, 2nd study using H120 through H150 core samples (A11-13 and A23) which were all taken from a core recovered in Kikeh 4ST-1. Samples were extracted from selected zones in these cores for both retention testing and also particle size measurement to confirm that the sands used for retention testing were representative. In 2006, significantly more particle size data became available and this data was also analyzed and compared to confirm the relevance of the retention testing already done. Once the weave selection had been finalized, the drilling fluid to be used for the reservoir sections underwent laboratory testing to identify the ease with which it could be conditioned to allow the screens to be run without plugging and to determine whether the filtercake formed would pass through the ESS when the well was brought onto production. Sand retention testing: Two types of sand retention test are typically performed in screen selection laboratory studies, and both were used for this project: • Slurry tests where the sand is suspended in a slurry and flowed through the screen • Sandpack tests where the sand is placed directly onto the screen and a wetting fluid is flowed through The test methods are fully described elsewhere (SPE82244 and SPE98308). Three sand samples were supplied from the previous study (K, N, and O sands), and in addition three slabbed preserved cores were available for sand extraction and psd analysis. The sand in the cores was finely interbedded with coal and required careful extraction with a scalpel for particle size analysis. There were two layers of sand which yielded enough material for retention tests and these samples were included in the study. The particle size distributions were measured by both sieving and Laser Particle Size Analysis (LPSA), and the data obtained indicated that the 150 micron ESS weave would be able to retain all the sands. Slurry test were performed in duplicate on all the sands, and, since typically slightly less sand passes through for sandpack tests compared to slurry tests, the two coarser samples were not used in the sandpack tests. Table 3 summarizes the total amount of each sand which passed through the weave in all the tests, and also gives details of LPSA size information for each sand. As a general rule, up to 0.3g of sand through is considered acceptable. The amount of sand passing through during the slurry tests is within the general retention guideline of less than 0.3g, and the results are in the order of the size and sorting of each sand as would be expected.

OTC 19484 7

Sand through

sand D1 D5D10

D50 D90 D95<45µm%

UCslurry sandpack

H145 (K) 202 171 154 89 8.9 3.1 26 11.5 0.0985 0.0519

H145 (N) 218 194 179 115 23 6.7 7 5.6 0.0499

H145 (O) 447 324 285 185 54 18 10 3.8 0.0294

H120/136 (A11-13) 192 169 153 96 14.2 4 21 7.5 0.0626 0.0200

H150 (A23) 189 160 144 84 8.1 3 27 11.9 0.1085 0.0669

Table 3 Summary of the amount of sand passing though for retention tests The pressure data from the slurry tests (Figure 6) shows there is very good agreement between the duplicate tests. Generally the pressure begins to rise as the sand starts to build up on the weave. So the delay in the beginning of the pressure rise can be an indication of the amount of retention (the longer the delay in pressure build-up the less sand is retained). Jagged profiles suggest that sand bridges are breaking down. The pressure gradient is a reflection of the permeability of the sand, but may also indicate plugging, and separating a pressure rise due to the sand itself building up on the weave from a pressure increase due to plugging of the screen is not straightforward. The pressure rises are steepest with the K and A23 sands which are the finest and most poorly sorted. The O sand the coarsest sand gives the lowest pressure rise. As with the retention results, the pressure data is in the order of size and sorting of the sands.

Standard sandpack tests were run with three of the sands (K, A23, and A11-13), and the results are given in Table 3. Although these results show acceptable retention, the flow velocity in the Kikeh wells was expected to be twice the standard lab flow rates. Since flow rate can have a significant influence on retention in sandpack tests, further testing was performed at the higher rates to ensure that the amount of sand passing through the 150 micron weave tailed off to acceptable levels after the clean-up stage of well production. The high-rate tests were performed with sands A23, A11-13, and 2C. The 2C sand is very similar to the A23 sand, while the A11-13 sand is slightly different in that it is better sorted. The higher more realistic reservoir flow rate for Kikeh did increase sand production through weave, but the amount of sand through

settled down to a similar background rate as tests performed at half the flow rate. The rate of sand production when scaled up to a typical Kikeh well is 1lb/1000bbl, which is the worse case as it assumes all the well is composed of a fine, poorly-sorted sand, where in reality much of the reservoir is composed of coarser, better-sorted sands which will provide most of the production. Assessment of recent particle size data: A more recent coring exercise provided large amounts of additional particle size data from seven zones wells; H110, H115, H120, H130, H136, H145, and H150. This data was assessed together with permeability measurements to determine whether all the sands measured would contribute to production, and it was decided that zones with permeability less than 80mD could be ignored for sand control purposes. The remaining sand size data was then compared with the particle size distributions of the K, N, and O sands used in retention testing. Looking at a comparison with the H145 sands, (Figure 7), it can be seen that almost all of them are embraced by the range of the test sands. Similar results were obtained for all the new data, and so it was felt that the original retention testing was still valid and the 150 micron weave was selected.

0

5

10

15

20

25

30

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50sand reaching screen (g)

pres

sure

dro

p (p

si)

150 ESS K sand

150 ESS K sand repeat

150 ESS A23

150 ESS A23 sand

150 ESS repeat A11-13

150 ESS repeat A11-13

150 ESS N sand

150 ESS N sand repeat

150 ESS O sand

150 ESS repeat O sand

Figure 6: Pressure profiles from slurry tests on Kikeh sands

H145 from core 5 Gsiz plots samples <80md removed

0

10

20

30

40

50

60

70

80

90

100

1 10 100 1000particle size (microns)

cum

ulat

ive

%

Figure 7: Comparison of H145 size distributions with those of the sands used in retention testing

8 OTC 19484

Fluid Testing The two aspects of drilling fluid compatibility with regard to installing sand control screens are conditioning the fluid such that it does not plug the screens when they are deployed, and producing the filtercake back through the screen on production start-up. Both these aspects were assessed by laboratory tests, and the test methods are described in SPE98287. A novel drilling mud with ultra-fine barite as the weighting agent was selected for this application to minimize the conditioning required, and, in fact, a freshly prepared laboratory sample passed the laboratory plugging test without any conditioning being necessary. Filtercake flowback efficiency is assessed by means of coreflood tests in which a mudcake is built up on a core plug at reservoir temperature and overbalance pressure, and then displaced through a section of the ESS weave by oil flow through the coreplug. The efficiency of mudcake removal is determined by reference to the core plug permeability before and after mudcake application and removal, and since the mud itself may cause permeability reduction within the rock, a baseline flood with no ESS present is performed for comparison. The tests performed on the 11.6 ppg WARP mud gave a return permeability of 43 percent after the cleanup oil flow with the 150 micron ESS present, rising to 74 percent after a high rate oil cleanup flow. Comparable results were obtained in the baseline test with no ESS present and showing that the 150 micron ESS would not significantly influence mud cleanup. Injector Erosional Analysis One of the key considerations for the water injectors was to analyze and understand the risk of erosion to the sand control filter media during injection. Extensive modeling was done to not only understand the situation, but also to use the data to select the optimum perforating system. Injection flow velocity through the perforations must be less than critical erosional velocity of screen, which is 2.49ft/s. The conclusion of the modeling done (Figure 8) was that with an 8 SPF system, creating a 0.56” hole diameter, would be well within the limits. The perforating company designed the gun system around this perforating strategy.

Figure 8 – Erosional Analysis

Well Type Summary Open Hole Producers: The design conclusion for the open hole producers was to run and cement the 9.625” production casing in place above the top of the reservoir. The 8.5” reservoir section was then drilled through the production zones to TD, with a conventional 2D J profile. The reservoir drill in fluid is constantly conditioned to ensure that the fluid in the well bore and the filter cake will not plug the 150 micron filter media on the ESS. The cleanliness of the mud is constantly checked on site using a production screen tester (PST). The test entails passing a litre sample of mud through a 150 micron ESS weave sample, to ensure it does not plug. At the outset of the project planning, the management and conditioning of the reservoir drilling fluid to meet the above criteria was seen as one of the biggest challenges of the entire operation. This was due mainly to the massive riser volume, and the inherent difficulties with cleaning and conditioning such a large fluid volume. Research was done to look at previous similar installations of ESS, to review and implement any lessons learned.

OTC 19484 9

During the drilling of the section, extreme care is taken to ensure that the hole diameter is as close to gauge as possible. This is for two reasons, first to ensure that the swell packers have the required hole size, which will enable them to provide the expected differential (Figure 2). Second, to ensure that the hole size is in the compliancy range of the ESS. As discussed earlier, compliancy is crucial for controlling Kikeh sands. On completion of a wiper trip, the open hole is displaced with PST quality mud, and the drilling bottom hole assembly is then pulled from the well. The logging while drilling tools are then downloaded; this data is evaluated, and then used to identify the hydrocarbon bearing zones. The space out is then completed, which will leave the ESS across the producing intervals, blank pipe across the shales, and swell packers positioned as required for zonal isolation. The space out is reasonably complex, this is due to the number of production intervals, to date six and seven respectively. This is because the shales are wherever possible isolated with blank pipe sections. The flexibility of the ESS system really provides that ability to fully exploit the reservoir, without exposing the screen unnecessarily to potentially weak zones. A mechanical cleanup trip is then run to clean up the casing and the riser, and ready for the deployment of the sand control system. This is a critical step in the success of the operation., Basically, the crucial factor is ensuring that the fluid is of PST quality, and will not cause the screens to plug during the trip in hole. If the screens were to plug, this would impair production, and, in extreme cases, could cause hydraulic collapse of the screens. The sand face completion (Figure 9), deployed was a 5.5” 150 micron ESS single trip system. The system is then picked up according to the predetermined space out; the expandable top connectors, and expandable bottom connectors are used as crossovers between the ESS and blank pipe, and provide the full flexibility of the system. The ESS equipment as standard is all 316L, and the remainder of the ancillary equipment is L80 13Cr. The EXR single trip hanger is picked up last, the hanger system is assembled with the single trip setting tool, and the ESS expansion tools. The completion is then deployed to depth using the rig work string, once on pipe tally depth; Gr-ccl is then run to confirm correct space out. With such a tight space out, this is critical to ensure that the completion is spaced out exactly on depth. The hanger is then set, and the setting tool is released from the hanger. The string is then lowered to position with the expansion tools above the top ESS zone. Primary expansion is completed with a solid retrievable 6.72” cone ring and weight is slacked off from surface, which expands the screen. The zones are expanded sequentially, top to bottom. On completion of the bottom zone, the string is pulled out of hole to position the expansion tools above the top zone again. Secondary compliant expansion is achieved by the ACE tool; fluid is pumped from surface which achieves the required pressure in the ACE tool, activating the expansion rollers. Weight is then slacked off from surface to expand the ESS against the well bore. On completion of the lower zone, the string is pulled out of the hole with the operation complete. The completion can then be run, which can be either a single, or intelligent completion as required. Cased Hole Injectors: The plan for the cased hole injectors is to drill a 12.25” section through the injection interval. A 10.75” x 9.625” production casing is then run and cemented in place.The well bore is then mechanically cleaned, and displaced to filtered calcium chloride brine. The injection intervals are then perforated under balance using an oriented TCP gun system. The benefit of the oriented system is that it is oriented to shoot at 90 and 270 degrees, thus minimizing fines dropping into the well bore during shut in periods. An injectivity test using treated seawater is then performed to ascertain baseline injectivity data. The well is then monitored to establish a fluid loss rate. Once confirmed, the loss control pill is spotted across the perforations. The well is then scraped and cleaned to remove any perforating debris and burrs from the casing. It was identified that the burrs could potentially damage the ESS, during thermal movement resulting from temperature change during injection. The ESS completion is then picked up (Figure 10), the string is run with a wireline entry guide, and this allows for fines entering the well bore during shut in periods to drop into the well’s sump. The ESS zones are picked up as required, with blank pipe placed between zones, and swell packers positioned as required. The equipment and installation process is fundamentally the same as the producers; the only key difference is the metallurgy. The weave on the ESS has been upgraded to inconel 825. This was as a precaution against oxygenated corrosion, which may result from oxygen excursions during water injection.The ancillary equipment is all L80 1%Cr. The sand face completion is then deployed to depth, with the ESS spaced out across the perforations, with a minimum 3 meter screen overlap. The hanger is then set, the screens selectively expanded, all exactly according to the procedure used on the producer wells. On completion of the expansion process, a breaker for the loss control pill can be spotted through the expansion string, to accelerate the breakdown of the gels. The well is then ready to be completed.

10 OTC 19484

Figure 9 – Typical Oil Producer ESS Schematic Figure 10 – Typical Water Injector ESS Schematic Implementation and Learnings From the outset of the project, Murphy has placed the greatest emphasis on quality and safety, to attain operational excellence and absolute efficiency. Through all stages of the operations, opportunities for offline and simultaneous operations were utilized, thus ensuring that efficiencies are always maximized. Murphy has an in-house ESS Project Manager and a dedicated offshore ESS Team Leader assigned to the project, interfacing with the Murphy Engineers during the planning and installation phases. This provides a free flow of information between all sides of the team.

OTC 19484 11

Before the manufacture of any components for the ESS contract on Kikeh, a manufacturing quality plan was created from each of the equipment supply plants, and Murphy’s dedicated quality inspectorate identified the critical steps that were required to be overseen and approved. This would ensure that all equipment was of the highest standard, and that any non-conformance issues would be highlighted to Murphy for review. The next step was to create a district workshop quality plan to cover the receipt, preparation and load out of all Kikeh equipment. Again, this was reviewed and approved by Murphy’s quality inspectorate to identify the areas that their quality engineer would require to witness and approve. The last major component was to produce installation procedures, that identify clear operational and installation limits, for the deployment and expansion of the ESS system. The above were all put in place to provide absolute clarity on the whole operation, and to provide personnel with clear limits to work within, and to share the responsibility. Any issues that arise are reviewed and discussed as a team, and an informed decision can be made knowing that all the facts are in place. Results and Conclusions At the time of writing this paper, Murphy has installed the ESS system in two open hole producers, one of which was completed with an Intelligent Completion, and three cased hole water injectors. The performance of all the wells completed to date has met Murphy’s requirements and expectation, with production and injection targets for the wells being met. The two producers completed have yielded skins of less than 1, and the three injectors have skins from -1 to 5 at mid range of injection rate and 11 at maximum injection (Table 4 and Table 5).

Injectivity Data- Water Injectors

Well Event Skin/PI max

injection rate

WX-11 H110/H115/H120 sands

Subsea Cased ESS initial injectivity skin of -1.05 at 15bpm, skin -0.57

on fall off 15bpm

WX-08 (H110/H120 sands) Subsea Cased ESS initial injectivity skin 3.9 at 2bpm , 5 at 8bpm, 11 at

13bpm 13bpm

WX-10 (H130/H136 sands) Subsea Cased ESS initial injectivity noisy data-not interpretable 12 bpm

Table 4 Summary of injectivity performance of completed ESS water injectors

Productivity Data-Producers

Well Event Skin/PI max

injection rate

PX-01 H150 sand PBU PI of 68, skin 1 -Source EOWR BU

data n/a

PX-02 Smart well PBU H115/H120 PI 47, skin 1 source EOWR BU data

H110/H115/H120 H110 PI 38, skin -3 PX-03

H130/136 commingle, openhole ESS

PBU PI of 8.2, skin 1 -Source EOWR BU data

Table 5 Summary of productivity performance of completed ESS producers With the information available at this time, it seems that the oil swell packers are providing zonal isolation. However, there have been some complications with zonal communication, which may be attributed to the SMART completion, but this cannot be confirmed until the completion of the investigation. The use of ESS technology has proven to be the correct choice for the Kikeh development with the objectives being met to date. The focus of the project has always been to maintain the highest standards of quality, and operational excellence. These measures, coupled with the efforts of all involved, are keys to the successes that have been achieved to date.

12 OTC 19484

Acknowledgements The authors wish to thank Petroliam Nasional Berhad (PETRONAS0, Petronas Carigali Sdn Bhd (PCSB) and Murphy Oil Corporation for permission to publish this paper, and Weatherford International Ltd for their encouragement to do so. Nomenclature ACE = Axial Compliant Expander EBC = Expandable Bottom Connector ESS = Expandable Sand Screen ETC = Expandable Top Connector EWBS = Expandables Wellbore Stability Model PST = Production Screen Tester SPF = Shots per Foot TCP = Tubing Conveyed Perforating References 1. Ballard, T and Beare S.: "Media Sizing for Premium Sand Screens: Dutch Twill Weaves", SPE 82244, SPE European

Formation Damage Conference, The Hague, Netherlands, 13-14 May 2003. 2. Ballard, T and Beare S.: " Sand Retention Testing – The More You Do, The Worse It Gets", SPE 98308, SPE International

Symposium and Exhibition on Formation Damage Control held in Lafayette, L.A., 15–17 February 2006. 3. Hampshire, Ken, et al.: "Kikeh ESS® Well test – a Case History of a Deep water Well test, Offshore Malaysia" SPE 88564,

SPE Asia Pacific Oil and Gas Conference and Exhibition, Perth, Australia 4. Jones, C., Tollefsen M, Somerville J.M., & Hutcheon R.: “The Prediction of Skin in Openhole Sand Control Completions,”

paper SPE 94527 presentation at the SPE 6th European Formation Damage Conference, Scheveningen, The Netherlands, 25-27 May 2005.

5. Webb, T, Selamat, K, Omar, J, & Desormeaux R.: “Kikeh Development: Delivering World Class Completion

Performance,” paper OTC 19467 presentation at the Offshore Technology Conference held in Houston, Texas, U.S.A., 5–8 May 2008.

List of Illustrations Fig. 1 Kikeh Main Field Area Fig. 2 Oil Swell Packer Differential Pressure Profile Fig. 3 Oil Swell Packer Simulated Swell Times Fig. 4 Water Swell Packer Expansion Chart Fig. 5 EWBS Model Results Fig. 6 Pressure profiles from slurry tests on Kikeh sands Fig. 7 Comparison of H145 size distributions with those of the sands used in retention testing Fig. 8 Erosional Analysis Fig. 9 Typical Oil Producer ESS Schematic Fig. 10 Typical Water Injector ESS Schematic