Gravel-packing of open-hole highly-deviated or horizontal wells is increasingly becoming a common practice, especially in deep water and sub-sea completion environments where production rates may reach up to 50,000 BOPD or 250 MMSCFD. In these wells, reliability of the sand face completion, in addition to other factors, is of utmost importance due to the prohibitively high cost of intervention or side-tracking and the very high hydrocarbon recoveries required per well. To date the norm in gravel-packing such wells is water-packing or shunt-packing with water-based fluids. With both techniques, filter-cake removal treatments are conventionally done through coiled tubing after gravel packing, pulling out of the hole with the service tool and running in with the production/injection tubing. Furthermore, because conventional gravel-pack carrier fluids are water-based (brine or viscous fluids), water-based drilling fluids are traditionally used to drill the reservoir section to ensure compatibility and improve wellbore cleanup, even if the upper hole is drilled with a synthetic/oil-based drilling fluid. In this paper, we discuss several novel techniques that can substantially improve return on investment in gravel packing of open-hole horizontal completions, through reduced cost and process time, improved fluid management practices, increased productivity and/or reduced risk of future interventions, so mitigating against the risk of sand face completion failure or under-performance. The proposed techniques include:Simultaneous gravel-packing and filter-cake removal with water-based carrier fluids when the reservoir is drilled with a water-based drilling fluid: laboratory data relevant to gravel-packing are given and field case histories are discussed in detail.Simultaneous gravel-packing and cake cleanup with either water or a synthetic/oil-based carrier fluid when the reservoir is drilled with a synthetic/oil-based drilling fluid: laboratory data on cake removal while gravel packing are presented for both water-based and oil-based carrier fluids along with data on kinetics of cake removal.a new service tool that utilizes wash-pipe as continuous tubing and thus allows spotting of breaker treatments immediately after gravel packing: detailed description of the tool and its operation is given.Gravel-packing of highly-deviated or horizontal wells above fracturing pressure. Benefits offered by each of the proposed techniques are discussed in detail along with their current limitations. Introduction A great majority of the highly-deviated and horizontal wells are being completed as open holes, primarily because of their much higher damage tolerance, higher well productivities at high mobilities (kh/µ) and lower cost compared to cased holes. Although most of these wells in areas requiring sand control have been completed with standalone screens, a rapidly increasing fraction of them are now being gravel packed, particularly in deep water, high production rate and/or sub-sea completion environments (currently ca. 40%, and projected to be ca. 60% by 2003/2004). The major drivers for this current trend are the prohibitively high cost of intervention and much higher reliability associated with gravel packs.1,2

Offshore reservoirs requiring sand control pose a major completion challenge because of extremely high cost and risk involved in remedial treatments, particularly in sub-sea completion and/or deep-water environments. It is therefore of utmost importance to ensure sand control without sacrificing flow conformance, recoverable reserves and well deliverability throughout the expected life of the completion. A major trend in these environments is towards open-hole, horizontal, gravel-packed completions. Although gravel packing stabilizes the wellbore, it can also entrap the filter-cake formed by the reservoir drilling fluid, potentially resulting in high drawdown requirements (flow initiation pressures) and/or low production rates (retained permeabilities). The cleanup procedures in the industry have varied significantly from no cleanup at all to complicated two-stage breaker treatments involving post-completion coiled tubing intervention, with no guidelines existing in the literature. In this paper, we present experimental results and field cases involving filter-cake flow-back through gravel packs with and without cleanup. Effects of various parameters, including gravel size (40/60, 20/40, and 12/20), formation permeability, drill-solids type (clays, quartz) and concentration, and the type of cleanup fluid have been investigated. Flow initiation pressure and retained permeabilities during flow back are reported as a function of these parameters. The experimental results show that the flow initiation pressure is a strong function of gravel size and the type of drill solids. It is concluded that, in clean (low-to-no clay content) formations of large grains and high permeabilities (~ several darcies) requiring large gravel sizes (e.g., 12/20), an enzyme or an oxidizer treatment is sufficient based on laboratory results and productivity predictions. This conclusion is also supported by several field applications as shown. In lower permeability (~ 100–250 md) formations of small sand sizes requiring smaller gravel (e.g., 40/60) elimination of both the fluid loss control agent (starch) and bridging agent (CaCO3) is necessary based on high flow initiation pressures and low retained permeabilities. In intermediate permeability (~ 500–800 md) formations of medium size sand-grains typically requiring 20/40 gravel, the results depend strongly on the type of drill solids: in clean formations (no clays in drilling fluid), an enzyme or an oxidizer treatment is sufficient, while in dirty formations removal of both CaCO3 and starch is necessary. These results are also supported by field case histories presented in the paper. Introduction Gravel packing has been gaining wider popularity in open-hole horizontal completions where sand control is required, particularly in sub-sea completion and/or deep-water environment. The cost of intervention in such cases makes risk mitigation a much more pronounced task. Until recently, a large majority of horizontal sand control completions have utilized standalone screens. However, because a substantial fraction of these wells have failed prematurely (either productivity loss due to screen plugging or loss of sand control due to screen erosion),1 many operators have changed their primary completion technique in these wells from standalone screens to gravel packing. This is particularly true in formations containing a large fraction of non-pay (shale, mudstone/siltstone) and/or have a wide particle size distribution.2

Summary Reservoirs requiring sand control pose a major challenge for selecting a suitable completion method. Horizontal openhole completions have been successfully used in such reservoirs to eliminate sand production while maximizing productivity/injectivity and well deliverability throughout the expected life of the completion and minimizing risk and complexity. Although horizontal, openhole, sand-control completions, ranging from preperforated/slotted liners to gravel packs, have been applied widely in the last decade and many case histories have been discussed in the literature, a systematic methodology for selecting these completion methods remains to be documented. It is the objective of this paper to propose such a design methodology by unifying the broad experience and understanding from a global, technically integrated perspective. The paper first discusses a generalized and unified methodology for determining when to install sand control, what to install for sand control, and how to install it in horizontal openhole completions. Specific factors recognized as affecting "when" are in-situ stresses, pore-pressure decline (sand prediction), expected well life, production rate, hydrocarbon and well type, gross product value, sand tolerance capacity, environmental limitations, and intervention capabilities, while the integration of all these factors has an impact on the overall risk analysis. In addition to many of the previous factors, critical drivers affecting "what" are identified as wellbore architecture, reservoir lithology and petrophysical properties, and sandface equipment reliability. Additional parameters impacting "how" are reservoir drilling fluid, displacement and cleanup methodology, screen type, operational implementation/assurance (risk management, operational timing, and location logistics), torque and drag analysis, and gravel-placement simulations. Secondly, examples of this methodology are presented in detailed case histories pertaining to different types of horizontal, openhole, sandface completions, including slotted liners, stand-alone screens (including expandable), and gravel packs, as well as various integrated cleanup methods, along with a summary of the lessons learned by each company. Introduction Horizontal openhole completions have been widely used in the oil and gas industry for effectively exploiting hydrocarbon reserves in both sandstone and carbonate formations during the last 2 decades. In sandstones, a major issue has been whether sand control is required during the life of a particular well, and if so, what technique to use to minimize overall completion and remediation costs, thus increasing profitability. Most recently, Bennett1 has developed a crossplot (see Fig. 1) that looks at the likelihood of wellbore failure with respect to formation quality and has used this to provide guidelines as to sandface completion methodology based on experiences gained in North Sea and Gulf of Mexico wells. Earlier, others2 developed a questionnaire-based approach for pooling industry experience and providing a database of events from which to learn and share. Unfortunately, because of the anecdotal emphasis this encouraged, little remains or has been developed since, especially in light of the fact that the industry's pace of technological development and understanding has once again accelerated in the last 3 to 4 years. While we see that a tremendous level of expertise and experience has been gained and although a large number of horizontal-well applications in sandstone formations and numerous publications exist, a well-defined set of guidelines for selecting the most suitable sand-control technique in openhole horizontal wells has not been published to date. It is, therefore, the objective of this paper to provide a unified set of guidelines based on five operators' and a completion service provider's experience and engineering expertise. The paper is organized as follows. First, we discuss the sand-prediction methodology that should be used to determine whether sand control is prescribed, and if so, when in the life of the well/reservoir it would be required. Once it is established that sand control is needed, the next step is to decide between gravel packing (GP) and stand-alone screens (SAS), for which we offer criteria based on current field experience, knowledge, and experimental data.3 For GP, the next step is selecting between two methodology techniques used in openhole horizontal wells - water and shunt packing. This is then followed by screen selection for the respective techniques (i.e., water-packing, shunt-packing, and SAS completions). Numerous case histories, both successes and failures, are given to support the selection methodology. Finally, suggestions for future work are made and conclusions are drawn.

TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractIn this paper, we present a novel approach for drilling and completing open hole horizontal wells with a fully compatible synthetic/oil-based fluid system utilizing shunt tube technology. The proposed RDF is a synthetic/oil-external emulsion that is reversible through exposure to a fluid of pH less than 7. A surfactant package included in the RDF waterwets the bridging/weighting agents (e.g., CaCO 3 ) upon reversal of the emulsion. The synthetic/oil external emulsion developed for gravel packing typically contains 50-75% by volume aqueous phase as the internal phase and is completely solids-free. The internal phase can either be brine or a pHreducer as well as a fit for purpose dissolver for the bridging agents. The pH-reducing property of the internal phase provides the required break mechanism for the S/OB-RDF emulsion remaining in the RDF filtercake under leakoff conditions allowing the bridging agents and drill solids to be water-wet, ensuring dissolution of the bridging agents.Laboratory data are provided for filtercake removal kinetics as a function of temperature, overbalance during gravel packing, gravel mesh size, and drill solids concentration in the RDF. Rheological data are given. Retained permeabilities and flow initiation pressures measured with the combined core and gravel-pack system are presented. Implications of the laboratory results on field practice are discussed.

Open-hole horizontal wells are increasingly used to improve reservoir exploitation and production rates by targeting specific zones and maximizing reservoir exposure. The drilling fluid of choice in many of these wells is "oil based" due to enhanced drilling rates with minimized friction as well as improved wellbore stability. However, in horizontal wells requiring gravel packs, the industry in general has been reluctant to use OB reservoir drilling fluids (RDF) for various reasons. Because the gravel pack (GP) carrier fluids that have been successfully used to date are all water-based and the use of OB-RDF would necessitate displacement of open hole to WB fluids prior to GP, the practice has been to switch to WB-RDF once in the reservoir section. This was due to concerns as to adverse fluid-fluid interactions resulting in sludging and difficulty in maintaining filtercake integrity while displacing OB-RDF in the open hole, leading to complex fluid management issues. An additional factor has been the perception that WB-RDF filtercakes are easier to remove should it be necessary, since most commonly used cleanup chemicals are water-based and the weighting/bridging agents used in the RDF are also water-wet if the RDF is water-based. In this paper, we present results from experiments conducted with OB-RDFs in the presence of gravel packs. We investigate two scenarios:the gravel pack carrier fluid is water-based, andthe gravel-pack carrier fluid is oil-based. In the first case, provided that no sludges are formed during displacement to water-based fluids, the retained permeabilities are comparable to or better than those obtained with WB-RDFs, although values lower than 0.04% can be expected in the presence of sludging. Another issue relevant to gravel packing wells drilled with OB-RDFs is the yield strength of their filtercakes in comparison to WB-RDFs. It is found through yield stress measurements of various RDF cakes that OB-RDFs have several orders of magnitude lower yield strength than their WB counterparts. This finding is consistent with the reported lower flow initiation pressures for OB-RDFs, and indicates that cake erosion during gravel packing is more likely with OB-RDFs. In order to optimize the sequence of fluids to obtain a good displacement of the RDF at field scale, we use a purpose-built numerical simulator. This simulator is a fluids mechanics code that can accurately calculate displacement fronts in field conditions: eccentric deviated annulus with as many fluids as necessary. Its main use is to detect unstable displacements such as channeling of the displacing fluid on the wide side of the annulus or slumping in horizontal portions. Furthermore, we provide data on a new oil-based gravel pack carrier fluid that can be used to eliminate fluid incompatibility and fluid management issues associated with the switch from OB to WB fluids. The laboratory and large-scale yard test results are presented, addressing critical considerations for oil-based GP carrier fluids. It is found that such emulsion systems can thicken or break (depending on the emulsifier concentration) at high shear rates unless the emulsion is made at the highest shear that it will be exposed to. The implications of these results on field practices are discussed along with recommendations on avoiding damage in gravel packed wells drilled with oil-based RDFs.

Wellbore cleanup in horizontal, open hole sand control completions has been the subject of many publications in recent years. Although a large majority of horizontal wells have been standalone screen completions, an increasing number of these wells are being gravel-packed, particularly in deep water, sub sea environment where reliability of the sand face completion is of utmost importance due to prohibitively high cost of intervention. In such wells, increased significance of "doing it right the first time" further necessitates an emphasized consideration of wellbore displacement and filter cake removal treatments. Although a substantial amount of laboratory data on filter cake cleanup are available in the literature, a great majority of these data are relevant to non-gravel-pack completions. In field practice to date, cake cleanup in GP completions has almost exclusively been done after gravel packing and typically involved coiled tubing. Although several new methods have recently been proposed and successfully practiced in several applications (e.g., inclusion of cake-breaking chemicals into gravel-pack carrier fluids, post-GP breaker treatments immediately after GP w/o requiring coiled tubing), laboratory data directly applicable to such conditions have been scarce. In this paper, we present laboratory data relevant to gravel-packed completions. We show that the cake removal time scales in the presence of a gravel-pack are longer compared to absence of a gravel-pack on top of the filter cake. The degree of delay is shown to depend primarily on carrier fluid viscosity and whether the breakers are included in the carrier fluid or introduced as a post-GP treatment. It is further shown that including breakers in brine during water packing must be exercised with extreme caution since even slow-reacting breakers can yield premature screen out due to increased losses should external cake erosion occur. Introduction A large fraction of wells drilled in reservoirs requiring sand-control are being completed as horizontal open holes. Although a large fraction of these wells have been completed with screens-only, an increasingly higher number of these wells are being gravel packed, particularly in deep water and sub-sea completion environment where reliability of the sand face completion is of utmost importance due to prohibitively high cost of intervention. Furthermore, because the costs associated with filter cake cleanup treatments are typically marginal compared to potential intervention costs, a thorough cleanup treatment is considered an integral part of the completion in such wells, in order to maximize well productivity and longevity, and provide more uniform production profile and avoid premature water or gas breakthrough, although the actual cleanup methodology, including the type of chemicals and the placement techniques may vary significantly. Cake cleanup in gravel pack completions has traditionally been done after gravel packing, and typically with coiled tubing.1 Several new techniques have recently been proposed.2,3 These involve incorporating breakers into the gravel-pack carrier fluid to place the cake-breaking chemicals into the wellbore during gravel packing, as well as a modified service tool that can be used to displace wellbore fluids and spot breaker solutions by allowing circulation down through the wash pipe and up through the wash-pipe/base-pipe annulus.2

Initial Open Hole Gravel Pack (OHGP) completions that have been installed in Greater Plutonio to date have all achieved complete annular packs and zero mechanical skin factors, resulting in well productivity indices that are significantly greater than expected. The success of the Greater Plutonio OHGP completions has been attributed primarily to the rigorous design and field application of the fluid systems used at all stages of the well from drilling the reservoir through to the gravel pack itself and subsequent completion. An integrated approach was adopted for the design of the fluid systems involving extensive formation damage and fluid compatibility testing. To translate the robust design into a fluid system which can be applied effectively in the field, a thorough, fit for purpose QA/QC system for all drilling and completion fluids was developed, requiring extensive fluids testing and reporting at the well site. The paper describes in detail the reservoir completion philosophy, drilling and completion fluids' systems and overall operational practices used in the Greater Plutonio OHGP completions. It also discusses the fluids design phase of the project and the QA/QC processes implemented in the field. Finally the paper presents the well productivity data from the wells completed to date. Introduction The Greater Plutonio Development is a 5 field deepwater project, located in Block 18, offshore Angola (Figure 1). All 43 planned development wells are subsea in water depths ranging from 1200 to 1500m. Of the 43 wells, 20 are producers and 23 are injectors. The development drilling programme began in 2005 with the aim of drilling and completing 15 wells prior to first oil in 2007. All the 5 fields produce from poorly consolidated Oligocene, turbidite reservoirs and consequently sand control is required in all development wells. Each of the 5 fields is composed of multiple stacked reservoirs separated by shales. The formation sands are high permeability (Kaverage between 800 to 1500md) with low viscosity oil (0.5 to 0.8 cP). GOR ranges from 700 to 1100 scf/stb. Formation Sand Particle Size Distribution Laser Particle Size (LPSA) and sieve analysis was performed on the whole core recovered from the exploration and appraisal wells. The results of the analysis conducted on these wells are summarized in Table 1. These results are typical of all producing formations in the Greater Plutonio Development. The LPSA data for the main reservoir, the Plutonio O73, are shown in Figure 2. The majority of the sands are well sorted with low fines content. There are however some poorly sorted sands with higher fines content which result in the Uniformity Coefficient (D40/D90) ranging from 2 to 16 and the fines content from 1 to 15%. Overall the formation sands are poor to well sorted with low to high fines content. This particle size distribution data (PSD) combined with the limited amount of whole core data available for 5 fields, required the selection of a sand control system capable of providing well bore stabilization in all production wells. Shale Characteristics The deepwater Angola fields are located in shallow, immature sediments which typically have reactive shales that are incompatible with water based fluids. The experiences from the analogue fields in the basin clearly demonstrate the reactivity of the shales. The initial Open Hole Gravel Pack (OHGP) wells in an offset Angola deepwater block, were drilled with oil based mud and the hole was displaced to brine prior to running screens 1. Significant problems were initially encountered running screens due to shale instability and as a result the current system in the offset block deploys the screens in oil based mud.

No abstract