Particulate Invasion From Drilling Fluids
- Louise Bailey (Schlumberger Cambridge Research) | E.S. Boek (Schlumberger Cambridge Research) | S.D.M. Jacques (Birkbeck College) | Tony Boassen (Statoil) | O.M. Selle (Statoil) | J.F. Argillier (Inst. Français du Petrole) | D.G. Longeron (Inst. Français du Petrole)
- Document ID
- Society of Petroleum Engineers
- SPE Journal
- Publication Date
- December 2000
- Document Type
- Journal Paper
- 412 - 419
- 2000. Society of Petroleum Engineers
- 1.6 Drilling Operations, 1.2.3 Rock properties, 2.4.3 Sand/Solids Control, 4.1.5 Processing Equipment, 1.11 Drilling Fluids and Materials, 2 Well Completion, 5.1 Reservoir Characterisation, 5.3.2 Multiphase Flow, 2.7.1 Completion Fluids, 1.6.9 Coring, Fishing, 5.3.3 Particle Transportation, 4.3.4 Scale, 1.8 Formation Damage
- 4 in the last 30 days
- 1,046 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
We investigate the invasion of solids and their mobility during cleanup. We study the effect of weighting agent particle size on rock substrates of varying permeabilities. We find permeability damage increases but flow initiation pressures decrease with increasing substrate permeability.
We obtain quantitative profiles of solids invasion by scanning electron microscopy/x-ray mapping and synchrotron energy-dispersive x-ray diffraction tomography. We compare these profiles to core sectioning data. We find invasion profiles drop steeply but fines are observed deep within the core. We examine the effect of backflooding on the invasion profile. Near-surface damage is reduced but deeply invaded fines are unaffected by backflow.
We develop a deep bed filtration model for solids invasion and consequent permeability reduction. This model is compared to the profiles obtained in the invasion experiments. We find that we can fit the experimental invasion profile for monomodal particles using a single trapping coefficient. Backflow is modeled by reversing the flow rate. We postulate a phenomenological rate of erosion to untrap particles in line with experimental observations (30%). When erosion is included in the model, a peak in the backflow pressure is found. This peak may be correlated with the experimentally observed flow initiation pressure.
Well productivity is critically important if oil and gas reserves are to be developed economically. With the change in economic climate and the maturation of many existing fields has come an emphasis on reduced production costs and optimized productivity. The trend towards openhole completions places additional emphasis on damage avoidance. Near-wellbore permeability impairment from drilling and completion fluids can have a substantial, yet potentially avoidable, impact on well productivity. The proper design and engineering of fluid systems to minimize productivity impairment is therefore important.
The fundamental mechanisms involved in formation damage by drilling fluids and its remediation have been investigated as part of a research collaboration between Schlumberger Cambridge Research and the Institut Franc¸ais du Pe´trole and Statoil under the auspices of the EC-Joule Programme.1 In this paper we report studies of particulate invasion and its impact on production.
Particulate invasion is one of the primary causes of formation damage from drilling fluids. During the initial stages of filter cake growth particles are forced into the formation, building an internal filtercake which plugs the near surface pores. Removal of this internal cake can be difficult, and can lead to reduced permeability.
The importance of minimizing internal filter cake is widely recognized. Most attention has focused on the selection of an appropriately sized weighting agent to bridge across surface pores, thereby minimizing spurt loss.2 Alternative bridging approaches such as the use of structured fluids3 (for example, mixed metal hydroxide-bentonite systems) have also been explored. Despite these efforts, the minimum spurt is not zero and some internal cake does form. This invaded material can be quite tenacious. Francis4 reported significant damage from shallow invasion even after removal of external filter cake. Better understanding of the properties of both the internal and external cake is needed for improvements in drill-in fluids.
Visualization has been seen as an important step in understanding the mechanisms of permeability impairment determined through core flood testing.5 However as van der Zwaag observed,6 quantitative analysis of invasion is a much more powerful tool for assessing formation damage, giving information which can ultimately be used to develop predictive models. A number of different approaches have been pursued,5-8 ranging from electron microscopy and x-ray analysis, to chromatic tomography (CT) scanning and nuclear magnetic resonance (NMR) imaging. All have their advantages and disadvantages; some are destructive, while nondestructive techniques often cannot resolve features at the individual pore level. Here we report studies of particle invasion using two techniques: scanning electron microscopy energy-dispersive spectroscopy (SEM-EDS) x-ray mapping, and energy-dispersive x-ray diffraction tomography using a synchrotron source (synchrotron EDD-T).
In this paper we report experimental studies on invasion and bridging with particulate weighted fluids and the effect on permeability damage. We locate and profile the damaged zone using different analytical techniques and quantify the effect of cleanup by backflow. Finally we describe a deep bed filtration model of solids invasion and consequent permeability reduction. This model is compared to the invasion experiments.
Formation Damage Measurements.
Formation damage measurements were made with three different test configurations. Detailed descriptions can be found in Refs. 1, 3 and 8. Each permitted the in-situ determination of the core permeability before and after exposure to drilling fluids. Two are based on API high pressure-high temperature (HPHT) filtration cells modified to take core samples,3 L×d is either 8×38 or 32×25.4 mm with a flat-blade impeller to provide shear for dynamic filtration conditions. The third8 uses long cores of L×d of 190×40 mm. Pressure taps along the core holder at ?5 cm intervals allow direct spatial resolution of the permeability impairment during and after filtration.
Before use, cores were vacuum saturated with brine simulating connate water from the Heidrun reservoir. Permeability to brine was measured at a minimum of three flow rates. Unless stated, filtration was conducted for invasion of 1 pore volume under 20 bar differential pressure (in the opposite direction to the permeability determination) and at ambient temperature (23°C). After filtration, permeability to brine was again measured to give the percent of retained permeability. As backflow is imposed, a peak in the pressure is observed. This is defined as the flow-initiation pressure (FIP).9 The accuracy of measurements on short and long cores (32×25.4 and 190×40 mm) was ?1% and reproducibility 5%. Tests with the thin rock slices were less accurate; 10 to 20%.
Rock substrates with different permeabilities were used. The majority were sandstones with minimal clay content. Our primary substrate was Clashach sandstone. A polymer-based fluid was used; its formulation is given in Table 1, with different grades of barite and carbonate weighting agents, and particle size data given in Table 2.
We used a Philips XL30 SEM equipped with a motorized sample stage and a Noran Voyager energy-dispersive spectrometer. The external filter cake was removed and cores were carefully fractured before air drying for analysis.
|File Size||588 KB||Number of Pages||8|