Previous rheological studies of water-base and oil-base muds have concentrated on fluid viscosities at elevated temperatures and pressures. This paper extends the data available to cold temperature regimes which can be expected to occur in the riser for deepwater wells. The rheology of eighteen different drilling fluids covering WBM, OBM and SBM have been evaluated between -1° to 90°C and 1 to 344.7 bar. The Herschel-Bulkley and Casson models both fitted the OBM and SBM rheograms very well. For the different WBM systems, the Herschel-Bulkley model also fitted the salt/polymer fluids and unweighted bentonite-based fluid rheologies, however, the Casson model best described the weighted bentonite-based fluid rheologies. The effect of the cold fluid viscosity in the riser on ECD for deepwater wells was assessed using a software model which uses a transient temperature simulator in conjunction with a hydraulic model which accounts for different mud viscosity throughout the wellbore. Comparing the ECD predictions with those calculated where the mud rheology is independent of temperature and pressure, the ECD was found to be underestimated by up to 6.1% for a SBM and overestimated by 3.1% for a WBM. If there are formations with small differences in the fracture and pore pressure gradients, then these variations in ECD could cause well control problems.

Summary This paper evaluates the performance of a standard oil-based mud (OBM) to drill horizontal wellbores, concentrating on its formation-damage characteristics and the flow-initiation pressures (FIP's) required for production to flow through the filter cake. For heterogeneous reservoirs, damage is relatively low in low-permeability rocks, but the FIP is high. Conversely, for high-permeability rocks, the FIP is low but formation damage is relatively high. If the drawdown pressure available from the reservoir is low, the scenario exists where inflow will occur predominantly from the higher-permeability formations, which could be damaged badly, but little inflow will occur from relatively undamaged lower-permeability rocks. In terms of maximizing production, this is obviously a less-than-optimal scenario. Evaluations of cleanup fluids were conducted to gauge their effect on lowering the FIP of OBM filter cakes. Various fluids were screened for their mud-removal performance, which would indicate potentially good OBM "chemical breakers." Mud parameters such as oil:water (O:W) ratio, base-oil type, and emulsifier content all affected the efficiency of the cleanup fluids. The best cleanup fluids were used then in a series of core tests to evaluate their effectiveness in reducing the filter-cake FIP. Reductions between 25 and 40% were possible, although parameters such as soak time and overbalance pressure were critical to their success. Introduction Increased inflow area offered by a horizontal wellbore over a vertical wellbore, and hence the greater productivity available, has led to a large increase in the drilling and completing of horizontal wells. These wells often are completed with open holes where screens either with or without gravel packs are used. Where the ultimate goal of the drilling and completion phase is to minimize the skin and maximize productivity, the drilling fluid can have a major impact on achieving this aim. Various laboratory evaluations of drilling-fluid performance, in terms of the formation damage and FIP, have been reported in the literature.1–7 Although both water-based mud (WBM) and OBM results have been reported, there has been a relative emphasis on the former, particularly sized-salt and polymer-carbonate drilling fluids. This paper aims to assess both the formation damage and FIP for an OBM applied to a variety of reservoir rock permeabilities and to assess the use of displacement/cleanup fluids. Performance advantages of OBM, such as the lubricity, shale stability, and fluid loss and filter-cake characteristics, can make them particularly suitable for reservoir-drilling applications. However, if the design of the completion phase involves the use of a water-based brine or gravel-pack fluid, then the engineering of the fluids and displacement procedures will have a significant impact on the overall success of the completion (i.e., minimum skin). Displacement efficiency will depend on the hydrodynamic characteristics of the drilling and displacement fluids as well as the chemical interaction of the drilling and cleanup fluids. This paper evaluates some properties of the drilling fluid that control the efficiency of these cleanup fluids. Cleanup fluids also are evaluated for their efficiency in reducing the FIP needed to initiate flow from reservoir rocks. Where the FIP of the filter cake is higher than the flowing pressure available from the reservoir, it will be necessary to reduce the filter-cake FIP to achieve inflow from as large a section as possible of the horizontal wellbore. OBM Formation Damage and FIP Experimental Methods. A variety of rock types covering a range of permeabilities were used as substrates to evaluate the formation damage and FIP for OBM in heterogeneous reservoirs. These are listed in Table 1. Core plugs 25 mm in diameter and 30 mm in length were used throughout this study. The cores were vacuum saturated with brine, then flushed with Isopar L, a highly refined, isoparaffinic kerosene. The cores are brought to residual water saturation using Isopar L at a flow rate of 7.67 mL/min. For the formation damage and FIP tests, the permeability of the cores was measured at imposed constant flow rates of 2, 4, and 6 mL/min. For all the flow rates, the pressure drop across the core was measured by a pressure transducer fitted to the inlet of the core holder. Permeability of the core plugs was calculated by plotting the pressure drop vs. flow rate and curve fitting the data points. After measuring the initial permeability of the core plug, drilling fluid was placed in the cell, and the core was exposed to the mud at a temperature of 180°F, a differential pressure of 500 psi, and dynamic filtration at 150 rev/min for 3 or 17 hours. After the mud-filtration phase, permeability to Isopar L was measured again, flowing in the production direction, using the same flow rates as used to measure the initial permeability. As backflow was imposed, a peak in the pressure was observed, which appears to correlate with cake rupture.1,2 This pressure peak has been used by some authors as an explicit value to signify the reservoir drawdown needed to initiate flow through the drilling-fluid filter cake.1,3 Others use the difference in peak pressure with the equilibrium flowing pressure in the damaged core and define this as the FIP.2,4 Following Alfenore et al.,3 in this paper we will use the overall pressure peak, FIPpeak, as the drawdown needed to initiate flow from the reservoir and FIPeq as the pressure where the FIPpeak is adjusted for the equilibrium flowing pressure (Fig. 1). The recovered permeability is the difference between the equilibrium flowing pressures before and after the mud-filtration phase and is a measure of residual damage. Table 2 lists the drilling-fluid formulations used. A standard OBM using a low-toxicity mineral oil was the base fluid. The drilling fluid had an O:W ratio of 75:25 and mud weight of 10.53 lbm/gal (barite-weighted). As well as the various mud products needed to make up the fluid, Rev Dust (calcium montmorillonite) was added at 15 lbm/bbl concentration to simulate drilled solids. The drilling fluid had an API plastic viscosity (PV) of 31 cp and yield point (YP) of 21 lbm/100 ft2.

The nature and content of suspended solids in drilling fluids directly affects its key properties, adversely impacting the drilling process, which in turn lowers overall well quality, augments costs and increases the environmental impact of drilling. To control the impact of the drilled solids on drilling fluid properties, the engineer at rig-site firstly requires an effective monitoring tool, allowing frequent and accurate determination of the solids content at various points in the circulating system. A technique based on X-ray fluorescence (XRF) has been developed which is being utilised at the rig-site to overcome the limitations of the traditionally employed methods. It is the aim of this paper to document the development of the XRF technique and its implementation at the rig-site. XRF spectra from a sample are input into a multivariate calibration model to predict the concentrations of solid phases (barite, low-gravity solids), liquid phase (water or oil and brine), as well as some ion concentrations (potassium and chloride in water based mud, chloride in oil based mud). The accuracies of the predictions for the solid phases are significantly better than the traditional methods. The ease and speed of use of the XRF technique facilitate multiple sampling from various points in the circulating system and effluent discharges. Two case histories are cited which demonstrate the use of the XRF technique for monitoring the deployment of solids control equipment at the rig-site and optimising its configuration. Introduction For any hole section to be drilled, the chosen drilling fluid has specific properties to be maintained for efficient, cost-effective, drilling. The fluid density, viscosity, gel strengths, filter cake properties, inhibition levels and lubricity are all adversely affected by the drilled solids entrained during the drilling process. As a consequence of the increased solids loading in the fluid, the performance of the fluid deteriorates resulting in decreased rates of penetration (ROP), decreased hole-cleaning efficiencies, increased chance of differential sticking and increasing the chance of solids plugging producing formations, therefore impairing future well productivity. In managing and controlling the solids content of the drilling fluid, the engineer at rig-site requires an effective monitoring tool, allowing frequent and accurate determination of the solids content in the circulating fluid and also effluent discharges from solids control equipment (SCE). A technique has been developed which is being utilised at the rig-site to overcome the limitations of the currently employed methods. The concept of measuring the solids in drilling fluids by X-ray fluorescence (XRF) was presented in 1993 (ref. 1). This paper described the evaluation of this technique using a model calibrated with water based muds (WBM) comprised of barite, calcium carbonate to simulate drilled solids and brines of various salts. The model used XRF spectra coupled with certain sample specific inputs to predict the concentrations of high gravity solids (HGS), i.e. barite, low gravity solids (LGS), water and salt. Contamination experiments verified that substituting LGS phases of different chemical composition did not affect the predictive performance of the model. The quoted accuracy of the model was 0.32 volume percent (v/v%) for HGS and 0.8 v/v% for LGS. The resolution by which variations in LGS content could be detected was demonstrated to be approximately 0.3 v/v% absolute. It was estimated that the XRF technique was some ten times more precise than the traditional API retort technique for calculating solid volumes in drilling fluids. P. 167

The increased inflow area offered by horizontal wellbores, and hence the greater productivity potentially available, has led to an increase in the drilling and completing of such wellbores. The correct engineering of the reservoir drilling fluid is critical to achieving the ulitmate aim of minimising skin for the completed interval. This paper evaluates the performance of a standard OBM to drill such wellbores, concentrating on its formation damage characteristics and the flow initiation pressures (FIP) required for production to flow through the filter cake. For heterogeneous reservoirs, we show that in low permeability rocks the damage is relatively low but the FIP high. Conversely, for high permeability rocks the FIP is low but formation damage relatively high. If the drawdown pressure available from the reservoir is low, then we have the scenario where inflow will occur predominantly from the higher permeability formations which could be badly damaged but little inflow will occur from the relatively undamaged lower permebility rocks. In terms of maximising production, this is obviously a less than optimal scenario. With this in mind, evaluations were conducted of cleanup fluids to gauge their effect on lowering the FIP of the OBM filter cakes. Various fluids were screened for their mud removal performance which would potentially indicate good OBM ‘chemical breakers’. Mud parameters such as oil: water ratio, base oil type and emulsifier content all affected the efficiency of the cleanup fluids. The best cleanup fluids were then utilised in a series of core tests to evaluate their effectiveness in reducing the filter cake FIP. Reductions of between 25–40% were possible, although parameters such as soak time and overbalance pressure were critical to their success. Introduction The increased inflow area offered by a horizontal wellbore over a vertical wellbore, and hence the greater productivity available, has lead to a large increase in the drilling and completing of horizontal wells. These wells are often completed with open-holes where screens either with or without gravel packs are used. In this context, where the ultimate goal of the drilling and completion phase is to minimise the skin and hence maximise productivity, the drilling fluid can have a major impact on achieving this aim. Various laboratory evaluations of drilling fluid performance, in terms of the formation damage and flow initiation pressures, have been reported in the literature1–7. Allthough both water-based mud (WBM) and oil-based mud (OBM) results have been reported, there has been a relative emphasis on the former, particularly sized salt and polymer carbonate drilling fluids. It is the aim of this paper to assess both the formation damage and FIP for an OBM applied to a variety of reservoir rock permeabilities and assess the use of displacement/cleanup fluids. The performance advantages of OBM, such as the lubricity, shale stability and fluid loss and filter cake characteristics, can make them particularly suitable for reservoir drilling applications. However, if the design of the completion phase involves the use of a water-based brine or gravel pack fluid, then the engineering of the fluids and displacement procedures will have a significant impact on the overall success of the completion (i.e. minimum skin). The displacement efficiency will depend on both the hydrodynamic characteristics of the drilling and displacement fluids as well as the chemical interaction of the drilling and cleanup fluids. It is an aim of this paper to evaluate some of the properties of the drilling fluid which control the efficiency of these cleanup fluids. The cleanup fluids themselves are also evaluated for their efficiency in reducing the FIP needed to initiate flow from reservoir rocks. Where the FIP of the filter cake is higher than the flowing pressure available from the reservoir, it will be necessary to reduce the filter cake FIP to achieve inflow from as large a section as possible of the horizontal wellbore.

The Brent field is located in the Viking Graben of the northern North Sea and produces from the Brent Formation and the deeper Statfjord Formation. Virgin reservoir datum pressure in 1976 was approximately 5,655 psi at 8,700 ft TVDSS. Pressure support was maintained until the 1998 since when reservoir pressure has been depleting at about 500 psi per year. The current datum pressure in the Brent units is approximately 1600–1700 psi. Significant lost circulation problems started to be experienced in the late 90's and a study identified the cause as the narrowing of the mud weight window as reservoir depletion gradually lowered the fracture gradient [ref 1]. Mud weights have been lowered to mitigate against the costly lost circulation events. The mud weights currently used are typically 700–900 psi less than the shale pore pressure (i.e. underbalance). Two case histories from sub-horizontal wells last year illustrated that the shales can be drilled over 900 psi underbalanced with no indication of shale failure, however, when running in liners they both hung-up where shales had collapsed. Based on this experience, the minimum mud weight for drilling the sub-horizontal reservoir sections has been set at 700 psi underbalance relative to the interbedded shale pore pressure. Reservoir depletion has reached the point where the (static) minimum mud weight for shale stability is almost equal to the fracture propagation pressure (FPP) for the reservoir sands. In the first half of 2002, 5 of the 7 wells drilled experienced lost circulation. The average losses volume per well was over 5,000 bbls with nearly 19,000 bbls being lost on one well. Of more significance is the NPT associated with the losses, this averaged over 300 hours per well. The average monetary cost of the lost time and mud volume was close to £1 MM per well. In an environment drilling low cost sidetracks, accessing reserves of 1–2 MM boe, this situation was unsustainable. A task force set about finding possible solutions to combat the lost circulation problems. After technical review, including laboratory testing, a recommendation was made for a size and concentration of graphite to be added to the OBM. In 10 field trials to date all the sections reached TD without inducing lost circulation, however, 4 of the 10 sections did experience losses when running the liner and/or circulating prior to cementing. With the addition of the graphite the average losses per well has dropped from 5,238 bbls to 621 bbls and the associated NPT has dropped from 302 hours to less than 1 hour per well. Recent field data suggest that the graphite could add approximately 1000 psi to the fracture breakdown pressure. This opens a host of possibilities for future infill drilling in depleting reservoirs. Introduction The Brent field is located 186 km northeast of the Shetland Islands in the UK North Sea and has a STOOIP of 3.8 billion stb and a GIIP of 7.5 Tscf (figure 1). The field was discovered in 1971 and was brought on production in 1976, with annual production peaking in 1984 at 410 Mbbl/d. Since the 1980s, oil production has been experiencing decline, but because of the high solution GOR (ranging from 250–980 v/v) substantial gas reserves remain, dissolved in the residual and by-passed oil. In 1992 the decision was taken to depressurise the Brent Field to recover an additional 1.5 Tscf of gas and 34 MMstb of oil, so extending the end of field life by 5–10 years. This required a £1.3 billion redevelopment of three of the four Brent platforms to install pressure facilities for low pressure operations, to reduce operating costs, to implement safety upgrades and to refurbish the facilities.