EXPERIMENTAL STUDY OF FORMATION DAMAGE IN HORIZONTAL WELLS BY SYED ZEESHAN JILANI A Thesis Presented to the DEANSHIP OF GRADUATE STUDIES. KING FAHD UNIVERSITY OF PETROLEUM & MINERALS DHAHRAN, SAUDI ARABIA SALEM MBAS brane In Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE In PETROLEUM ENGINEERING skiseiselseisy BisUMI Number: 1419523 INFORMATION TO USERS The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized ‘copyright material had to be removed, a note will indicate the deletion. UMI UMI Microform 1419523 Copyright 2004 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346, ‘Ann Arbor, MI 48106-1346KING FAHD UNIVERSITY OF PETROLEUM & MINERALS DHAHRAN, SAUD! ARABIA DEANSHIP OF GRADUATE STUDIES This thesis, written by SYED ZEESHAN JILANI under the direction of his Thesis Advisor and approved by his Thesis Committee, has been presented to and accepted by the Dean of Graduate Studies, in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN PETROLEUM ENGINEERING Thesis Committee: Thesis Advisor Het Gonoeon— Dr. Habib Menouar Co-Advisor Dr. Mohammad Aslam Khan wn / Ia sty > a: @. Dr. Abdulaziz A. Al-Majed Dr. Sidqi A \Abu-Khamsin Department Chairman 2 Member SS SSS Dr. Abdullah A.M. Al-Shehri Dean of Graduate StudiesThis thesis is dedicated to my Parents, Teachers, Colleagues & Friends who contributed to this work in ways they may never know. iACKNOWLEDGMENT In the name of Allah, Most Gracious, Most Merciful ‘Of knowledge, it is only a little that is communicated to you (O' men!)' (Qur'an 17:88). All Praise to Almighty Allah, Who gave me courage, patience, ability and time to accomplish this work. My deep appreciation goes to my Thesis Committee: Dr. Abdulaziz A. Al-Majed (Chairman), Or. Habib Menouar (Co-Chairman), Dr. Mohammad Aslam Khan and Dr. Sidgi A. Abu-Khamsin. Their enthusiastic participation, guidance and ‘encouragement have been critical to the success of this work. Acknowledgement is due to King Fahd University of Petroleum & Minerals for the Research Assistantship. | am thankful to all faculty and staff of the Petroleum Department who have in one way or the other enriched my academic and research experience at KFUPM | am also indebted to the following laboratory staff of the PETE Department and Energy Research Lab for their invaluable help: Mr. Mansour Al-Dhafeer, Mr. Iddris Abukari, Mr. Abdul-Rahim Muhammadain, Mr. Moussa Ali Moussa, Mr. Abdul Samad Iddrisu & Mr. Ahmad Al-Shuwaikhat (Petroleum Department) and Mr. Azam Baig (ERL). Finally, | gratefully appreciate the support of all my fellow graduate students and friends on the campus who provided me moral support and a wonderful company.Table of Contents Title page Final approval Dedication ‘Acknowledgement Table of contents List of Tables List of Figures Abstract (English) Abstract (Arabic) Chapter 1: INTRODUCTION . Chapter 2: FORMATION DAMAGE .. 2.1 Skin Damage and Formation Permeability 2.2 Formation Damage Processes «. 2.2.1 Formation Damage During Well Operations « 2.2.1.1 Drilling .. 2.2.1.2 Casing & Cementing 2.2.1.3 Completion . 2.2.1.4 Well Servicing 2.2.1.5 Well Stimulation ... Page iff iv2.2.1.6 Production . 2.2.1.7 Secondary Recovery Operations-Injection Wells ... 2.2.1.8 Enhanced Oil Recovery «+. 2.3 Formation Impairment: Horizontal Vs Vertical Wells .... 2.4 Mechanisms of Formation Damage ...+.-+ 2.4.1 Physical Mechanisms... 2.4.1.1 Invasion of Mud Solids ... 2.4.1.2 Wettability Alteration and Surface Adsorption Effects .. 2.4.1.3 Phase Trapping .. 2.4.1.4 Fine Migration 2.4.2 Chemical Mechanisms. 2.4.2.1 Clay Swelling ... 2.4.2.2 Formation of Insoluble Precipization .. 2.4.3 Thermodynamical Mechanisms ... 2.4.3.1 Temperature Aging .. 2.43. Electrolyte Influxes & Shearin; 2.4.4 Bacterial Mechanisms .....+e+eeee++ Chapter 3: PARTICLE & PORE SIZE DISTRIBUTION 27 3.1 Bridging Process ... 3.2 Critical Particle Size ... 3.3 Particle Concentration .. 3.4 Effect of Particle Size and Shape on Cake Permeability .. 3.5 Effect of Pore Size Distribution on the Oi Recovery .. 3.6 Particle & Pore Size Distribution & Fluid Design ....... viChapter 4: LITERATURE REVIEW 36 Chapter 5: OBJECTIVES & APPROACH OF THE STUDY 79 Chapter 6: EXPERIMENTAL SETUP & PROCEDURE 82 6.1 Apparatus «. 6.1.1 Core Holder .. 6.1.4 Transfer Cell. 6.1.5 Piston Pump .. 6.1.6 Back Pressure Regulator «. 6.1.7 Differential Pressure and Chart Recorder ... 6.1.8 Fraction Collector ... 6.1.9 Miscellaneous . 6.2 Pre-Experimental Procedures « 6.2.1 Fluids «.. 6.2.1.1 Oils. 6.2.1.2 Brine ... 6.2.1.3 Drilling Fluid .. 6.2.2 Core Preparation .. 6.2.2.1 Determination of Core Properties ... 6.3 Experimental Procedure .. 6.3.1 Absolute Permeability 6.3.2 Sw; & Effective Permeability vil6.3.3 Circulation of Mud .. 100 6.3.4 Retum Oil Permeability . 100 6.3.5 Apparatus Cleaning Process « 101 Chapter 7: ULTRASONIC INSPECTION AND ANALYSIS OF THE DEPTH OF MUD INVASION 102 7.1 Introduction ... sone] 2 7.2 Objective & Approach of the Study ... 103 7.3 Ultrasonic .. 104 104 7.3.1 Longitudinal / Compressional Waves « 104 7.3.2 Shear / Transverses Waves 7.4 Influence of Petro-physical Characteristics on the Compressional Wave Velocity.. 108 7.4.1Control of Mineralogy on Velocity 105 7.4.2 Control of Grain size & Microstructure on Velocity .. 7.4.3 Influence of Density on Velocity 7.44 Control of Porosity on Velocity 110 7.4.5 Directional Variation of Velocity .. 110 7.4.6 Effect of Water Saturation on Velocity . ALL 7.4.7 Biffect of Temperature on Velocity 117 eeedT 7.4.8 Effect of Viscosity on Velocity .. 7.5 Experimental Setup & Procedure 118 7.6 Ideal Delineation of Velocity Distribution 123 villChapter 8: EXPERIMENTAL RESULTS & DISCUSSION 129 8.1 Results of Ultrasonic Method ..s.sssssessssceseeeesseeseseseseneesseseesnel SD 8.2 Overbalanced Pressure 147 8.2.1 Regain Permeability ...sessesssecsesssecsssseneesseesecsssanseneeseel 7 8.2.2 Invasion Depth .. 8.23 Skin Variation .. seseseneneneeed 53 164 8.3 Contamination Time .....sesssseeeeeeeeeee 22169 8.3.1 Regain Permeability ... sesenseeneseseneseeeesseneeenseeess 169 174 8.3.2 Invasion Depth ssssssesseesenenee 8.3.3 Skin Variation .. 8.3.3.1 Skin Profile along the Well Length ... Chapter 9: CONCLUSIONS & RECOMMENDATION 185 9.1 Conclusions .. cr seneneseneseseeeesneeeessee] 85 9.2 Recommendations .. seesssesssecceccesesssseseases e187 Chapter 10: REFERENCES 188 APPENDIX-A 198 APPENDIX-B 200 APPENDIX-C 201Table 21 61 62 63 81 82 83 List of Tables Page Characteristics Of Authigenic Clay Materials Water Based Polymeric Drilling Fluid Composition ‘Mixing times for Drilling Fluid Components Water Based Polymeric Drilling Fluid Properties Summary of the Experiments conducted with variable parameters ‘Values of constant parameters. ‘Summary of the mud invaded depths measured by ultrasonic testing method, 23 90 91 92Figure 24 22 31 61 6.2 63 6.42 6.4b 6.4c Ta 7.1b 72 13 14 15 List of Figures Page A Schematic of a Well with a Damaged Zone (Skin Damage) 8 Effect of Skin Factor on Well Flow Pressure. 8 Invasion of a permeable formation by mud solids. 30 Hassler Type Core Holder 83 Schematic of Dynamic Leak-off apparatus 37 Pore Size Distribution for Berea core. 9s Particle Size Distribution of Bridging Particles by Image Analyzer 96 Image of Bridging Particles by Image Analyzer 97 Image of Bridging Particles by SEM 98 Plot of mean velocity versus mean bulk density for dry samples 109 \Velocity/water saturated relationship in Massillon sandstone vs. confining pressure (resonant bar)(after Nur et al., 1980). ae Velocity/water saturated relationship for different materials 14 “Characteristic” behavior of velocity as a fumetion of water saturation for consolidated sediments and a confining pressure of 35 Mpa (after 1.5 Gregory, 1976). Influence of temperature on P-wave velocity in Venezuelan sands for different saturating fluids (ultrasonic measurements) after Tosayaet 116 al., 1985). Schematic of Ultrasonic mud depth measurement method. 120 xi76 1a 1.70 78a 78d 78c 81 82 83 84 85 8.6 87 88 89 8.10 8.11 Scanning position of the core sample Ideal Delineation of Velocity Distribution Invasion of Mud Solids & Filtrate in the Cylindrical Core. Dry Sample Sound wave follows zig-zag path. Oil Saturated Sample. Sound wave follows straight path Mud Invaded Sample. Sound wave follows straight path Velocity Profile for sample J-12; Overbalanced Pressure=100 psi; Contamination Time=4 hrs Velocity Profile for sample J-15; Overbalanced Pressure=100 psi; Contamination Time=12 hrs Velocity Profile for sample J-14; Overbalanced Pressure=100 psi; Contamination Time=24 hrs Velocity Profile for sample J-16; Overbalanced Pressure=100 psi; Contamination Time=48 hrs Velocity Profile for sample J-17; Overbalanced Pressure=200 psi; Contamination Time=24 hrs Velocity Profile for sample J-13; Overbalanced Pressure=300 psi; Contamination Time=4 hrs Velocity Profile for sample J-7; Overbalanced Pressure=300 psi; Contamination Time=12hrs Velocity Profile for sample J-7A; Overbalanced Pressure=300 psi; Contamination Time=24 hrs Velocity Profile for sample J-8; Overbalanced Pressure=700 psi; Contamination Time=4 hrs Velocity Profile for sample J-9; Overbalanced Pressure=700 psi; Contamination Time=12 hrs Velocity Profile for sample J-10; Overbalanced Pressure=700 psi; Contamination Time=24 hrs 121 125 126 128 128 128 136 137 138 139 140 141 142 143 144 145 1468.12 8.13 8.14 8.15 8.16 8.17 8.18 8.19 8.20 8.21 8.228 8.22 8.22¢ 8.23 8.24 8.25 8.26 Effect of Overbalanced Pressure on Regain Permeability after 4 hours. Effect of Overbalanced Pressure on Regain Permeability after 12 hours. Effect of Overbalanced Pressure on Regain Permeability after 24 hours. Effect of Overbalanced Pressure on Regain Permeability after 48 hours. Effect of Overbalanced Pressure on Regain Permeability. Effect of Overbalanced Pressure on Mud Invasion Depth after 4 hours circulation. Effect of Overbalanced Pressure on Mud Invasion Depth after 12 hours circulation. Effect of Overbalanced Pressure on Mud Invasion Depth after 24 hours circulation. Effect of Overbalanced Pressure on Mud Invasion Depth. Effect of Back Flow Pressure on Permeability Recovery of 12-inch Jong Berea core exposed to circulating Drilling Fluid. Bridging mechanism of the mud par for AP=100 psi. at the pore throat, Bridging mechanism of the mud particles at the pore throat for AP=300 psi. Bridging mechanism of the mud particies at the pore throat for AP=700 psi. Skin Variation with Overbalanced Pressure maintained for 4 hours. Skin Variation with Overbalanced Pressure maintained for 12 hours. ‘Skin Variation with Overbalanced Pressure maintained for 24 hours. Skin Variation with Overbalanced Pressure. xiii 148 149 150 151 152 156 157 158 159 160 163 163 163 165 166 167 1688.27 8.28 8.29 8.30 8.31 8.32 8.33 8.34 8.35 8.36 8.37 8.38 Effect of Contamination Time on the Regain Permeability under 100 psi Overbalanced Pressure. Effect of Contamination Time on the Regain Permeability under 300 psi Overbalanced Pressure. Effect of Contamination Time on the Regain Permeability under 700 psi Overbalanced Pressure. Effect of Contamination Time on the Regain Permeability, Mud Invasion Depth profile with Contamination Time for Overbalanced Pressure=100 psi. Mud Invasion Depth profile with Contamination Time for Overbalanced Pressure=300 psi Mud Invasion Depth profile with Contamination Time for Overbalanced Pressure700 psi. ‘Mud Invasion Depth profile with Contamination Time, Skin Profile with Contamination Time under Different Overbalanced Pressures. Skin Profile with Contamination Time under 100 psi Overbalanced Pressure. Skin Profile with Contamination Time under 100 psi Overbalanced Pressure for extended period. Skin Profile along the Well Length. xiv 170 mM 172 1B W7 178 179 180 181 182 183 184THESIS ABSTRACT NAME OF STUDENT: Syed Zeeshan Jilani TITLE OF STUDY: Experimental study of formation damage in Horizontal wells MAJORFIELD : _ Petroleum Engineering DATE OF DEGREE : March 2000 The influence of overbalanced pressure and long mud circulation time on the formation damage during drilling operations were studied by using an innovative ultrasonic method to measure mud invasion depth. Experiments were conducted on Berea sandstone core samples in a constructed Leak-off apparatus. Regain permeabilities and skin factors were determined by damage data generated through leak-off experiments and mud invasion depths were measured by an ultrasonic method, The results showed a decrease of regain permeabilities and increase in skin factors with overbalanced pressure and mud contamination time. Whereas the mud invasion depth decreases with increase in overbalanced pressures till it reaches a critical pressure. Beyond that, invasion depth increases with overbalanced pressure. A strong physical relationship between overbalanced pressure and pore & particle size distribution is observed which affects the mud fines and filtrate invasion depth. It is found that the permeability damage is more dependent on the severity of the formation damage than invasion depth. It is recommended to drill the well under low overbalanced pressure with high rate of penetration to produce maximum regain permeabilities and minimum skin factors. An example of application of the results to horizontal wells is given at the end to illustrate how the method can be applied to horizontal wells. MASTER OF SCIENCE DEGREE KING FAHD UNIVERSITY OF PETROLEUM & MINERALS Dhahran, Saudi Arabia xvLugl! paile get LES tee el LBW UY J i peal USM apie Aids: dba ylgie Saadeh spe ell pheee gules apetall Sault ASU be US etd le Upyb Ul LL gb basesrcilyall pot Leal oo JS tb of ASN ab 58 Gat ural Be Gp beg oY Hab plainly deeds LH ober «el US ce eA gS LT ell weer Sle gg LLU ne wate le bell cael ads stall BS Geb ge iy de JS Jb Wal opto ley Hii need Fuad Gall ule! sbpall Gb Obegll Uealy abl ob 32 ge TZU apt Glel ag ad Sle! 3 Well Saw Sy « aiLyatl pb Letall Baby np tall Jules toes Well 25U ya GLI Cay 1 ilps ad Lascll alge dis ib cn a5 Gee of LUT cig, Late « casi go pit ysl US ae etl bell ol be biel iy Leb I ee, PE tly Seal Ole wists «pl mes ail! oo ne SLE opts of ae A. etyall OU ape Gents LES tlh Olenee ow JS de al Gat de edly « Peal OU Se et Baht ge ge Stal cee Ipaall gle (jie) GLa lang sabes oll pie bie oF USN jie pe ob Lull pong sagt Osa Bly 25H bel ALAN Ls Gach LAS pings) LBW LSU Lal all es le Heeler Jie EU! Bhd a aglell oi singh! Zp0 oaledlg Jaci aad Put >Paty Figure 2.2: Effect of Skin Factor on Well Flow Pressure.2.2. Formation Damage Processes Formation damage is defined as any type of a process, which results in a reduction of the flow capacity of an oil-, water- or gas- bearing formation. Formation damage has long been recognized as a source of serious productivity reductions in many oil and gas reservoirs and as a cause of water injectivity problems in many waterflood projects. 2.2.1 Formation Damage During Well Operations Formation damage can occur whenever non-equilibrium or solid bearing fluid enters a reservoir, or when equilibrium fluids are displaced at extreme velocities. Thus, many processes used to drill, complete or stimulate reservoirs have the potential to cause formation damage. Some of these operations are; 2.2.1.1 Driling Mud Solids and Particle Invasion * Pore throat plugging * Particle movement Mud Fittrate Invasion * Clay swelling, Flocculation, dispersion and migration, * Fines movement and plugging of pore throats.10 © Adverse fluid-fluid interaction resulting in either emulsion/water block, or organic scaling. * Alteration of pore structure near welbore through drill bit action. 2.2.1.2 Casing & Cementing * Blockage of pore channels by cement or mud solids pushed ahead of the cement. + Adverse interaction between chemicals (spacers) pumped ahead of cement and reservoir minerals fluid. * Cement filtrate invasion with resulting scaling, clay slaking, fines migration and silica dissolution 2.2.1.3. Completion + Excessive Hydrostatic pressure can force both solids and fluids in to the formation. + Incompatibility between circulating fluids and the formation with resultant pore plugging. + Invasion of perforating fluid solids and explosives debris into the formation with resultant pore plugging. + Crushing and compaction of near-wellbore formation by explosives during perforation. + Plugging of perforation of extraneous debris (mill scale, thread dope, and dirt)," Wettability alteration from completion fluid additives. 2.2.1.4 Well Servicing Problems similar to those that can occur during completion. Formation plugging by solids in unfiltered fluids during well killing. Adverse fluid-fluid and fluid-rock interaction between invading kill fluid and reservoir minerals. Damage to clays from dumping of packer fluids. 2.2.1.5 Well Stimulation Potential plugging of perforations, formation pores, and fractures from solids in the well kill fluid. Invasion of circulation fluid filtrate into the formation with resultant adverse interaction. Precipitation of hydrofluoric acid reaction by-products during acidizing. Potential release of fines and collapse of the formation during acidizing. Precipitation of iron reaction products. Plugging of pores and fractures by dirty fracture liquids, Inadequate breakers for high viscosity fracture fluids may cause blockage of propped fracture. Fluid loss or adverting agents may cause plugging of the perforation, formation pores, or propped fracture. Crushed proppants may behave like migratory fines to plug the fracture.12 + Inorganic/organic scales in the wellbore along with remnant debris may plug perforation, pores , or etched fractures. Fracture conductivity decline due to proppant embeddment. 2.2.1.6 Production Initiation of fines movement during initial DST by using excessive draw-down pressures. Inorganiclorganic scaling through abrupt shift in thermodynamic conditions. + Sand production in unconsolidated formation triggered by water encroachment into producing zones. + Screens of gravel packs can be plugged by produced silt, clay, mud, scale, ete. + Sand-consolidation wells may be plugged by debris, and sand-consolidating material may reduce reservoir permeability. 2.2.1.7 Secondary Recovery Operations - Injection Wells + Formation wettability alteration from surface-active contaminants in the injection water. + Impairment of injectivity due to suspended solids (clays, scale, oil and bacteria) in the injection water. + Formation plugging by iron corrosion products. + Inorganic scaling due to incompatibility of injected and formation waters.8 In pressure maintenance with gas injection, formation may be plugged by compressor lubricants that may also alter wettability. Reduced well injectivity from injected corrosion inhibitors in gas zones. 2.2.1.8 Enhanced Oil Recovery Fines migration, clay swelling, and silica dissolution initiated by contact of high pH steam generator effluents (condensates) with the formation rock during thermal recovery. Dissolution of gravel packs and increased sanding during thermal recovery. Inorganic scaling due to change in thermodynamic conditions during steam injection. Plugging due to carbonates deposition during CO; injection. Deposition of asphaltenes with CO2 contacts asphaltic crude oils. Potential emulsion formation during CO2 wag process. Fines movement due to hydrodynamic conditions of velocity and viscosity during chemical EOR process with surfactants and polymers.14 2.3 Formation Impairment: Horizontal Vs Vertical Wells Horizontal wells are much more susceptible to damage than their vertical counterparts due to a number of reasons, these being: 1. Substantially longer contact time with the driling fluid. in a vertical well, drilling fluid may only be in contact with the pay zone during few hours while in a horizontal well the time may be measured in weeks. 2. Most horizontal wells are not cased & perforated and remain as open hole completions. Relatively shallow damage, which would be easily perforated through a standard conventionally cased vertical well, remains a major source of permeability reduction in many horizontal wells, 3. Uniform drawdowns are difficult to obtain on horizontal welis due to the length of the well in the pay zone. This makes it much more difficult to clean up damage due to invaded fluid and/or solids except in selected zones. 4. The physical mechanics laws in the horizontal wells are substantially different from vertical wells due to the facts that the vertical and horizontal permeabilities in most formations differ. This generally results in equivalent amounts of damage causing substantially greater productivity impairment in horizontal Vs vertical wells.18 in comparison with vertical well, Horizontal well shows much more damage than vertical well due to additional drilling time but it can sustain more damage than vertical well with out a significant loss of well productivity. 2.4 Mechanisms of Formation Damage Due to the invasion of mud solids and filtrate into the formation, as well as the interaction between the used drilling fiuids and formation rock and /or the pore contents, the formation can be damaged in different ways. The depth of the damage depends on the near wellbore conditions, petro- physical characteristic & composition of the rock, mud filtrate and solids that flow with it under dynamic conditions. The mechanism of formation Damage or permeability reduction can be classified mainly into four categories. 2.4.4 Physical Mechanisms Physically, the formation can be damaged by 2.4.1.1 Invasion of Mud Solids: The invasions of artificial mud solids and un-removable solids produced by bit-rock interaction into the formation during overbalanced drilling conditions are the main cause of formation Damage. During drilling, the bridging mud solids16 ranging from larger to slightly smaller in sizes than the pore openings of the rock, build up on the wellbore face to form a low permeability filter cake. Particles that are smaller than the pore opening flow into the formation along with the mud filtrate. These solids eventually form intemal bridges at pore restriction between internal mineral grains. The colloidal particles in the driling fluid that might migrate and block flow channels include: clays, cuttings, weighting agents, various polymers and lost circulation agents. Much of the filter cake will be removed through the action of drill bit and circulating mud, but most of the internally bridged solids may be trapped. These solids create a skin effect around the wellbore, resulting in a significant formation pressure drop. If these particles are not flushed out completely when the well is put on production, they block the near region. The invasion of these colloidal particles in turn can be controlled by adding enough bridging particles of the right size to the drilling fluid. The type of damage (internal & external) as well as the bridging process is function of the particle/pore ratio, their shape, fluid velocity, particle concentration and charges on the particle and the pore walls. 2.4.1.2 Wettability Alteration and Surface Adsorption Effects Many drilling fluid additives used for mud rheology, stability, emulsion control, corrosion inhibition, torque reduction or lubricity etc. contain polar7 surfactants or compounds which can be preferentially adsorbed on the surface of the rock. The physical adsorption of these compounds can cause reductions in permeability by physical occlusion of the pore system, in the case of high molecular weight tong chain polymers, particularly in low permeability porous media where the small pore throats may be easily bridged by long chain polymer molecules. Numerous chemical additives such es emulsifiers, corrosion inhibitors, oxygen scavengers, alkalinity control agents and bactericides, normally used in various oil field operations can interact adversely with the reservoir fluid/rock, changes the interfacial tension and alters the preferential rock wettability from either water-wet to oil-wet or vice versa. Such a change reduces the effective hydrocarbon permeability in the invaded near wellbore regions of the reservoir which may adversely elevate producing water oil ratio for the well if the compietion is in a zone where a mobile water saturation is present. 2.41.3 Phase Trapping The loss of both water or oil based drilling mud filtrate to the formation in the near wellbore region due to leakoff during overbalanced drilling operations, or due to spontaneous imbibition in some situations during under-balanced drilling operations, can result in permanent entrapment of a portion or all of the invadingfluid resulting in adverse relative permeability effects which can reduce oil or gas permeability in the welbore region. 3.1. Waterblock The filtrate of water-based drilling fluids displaces the pore contents and change the saturation ratios in the invaded zone. To produce from these formations, the non-wetting fluids have to drive out the wetting filtrate from the pore spaces, for which a critical pressure or threshold pressure is required. This phenomenon is known as “Waterblock”. 2.4.1.3.2 Emulsionblock The filtrate of water-based drilling fluids and oil in pore spaces can form stable emulsion, which need even higher pressure to be drived out from the formation. 2443.3 Gasblock Reaction between the filtrate and pore contents and/or matrix materials can result in gas production. This additional phase decreases the relative permeabilities of other phases in the pore spaces.19 2.4.1.4 Fine Migration Fine migration is the motion of naturally pre-existing particulate matter in the pore system. This may be induced during the driling process by high fluid leakoff rates of water or oil-based mud fittrate into the near welbore region caused by elevated hydrostatic overbalance or excessively high under-balanced pressures. Fines migration is also identified as the major causes of permeability impairment in porous media. The fine particles may either come directly from the drilling mud or released by the invading filtrate from the pore walls. Many different types of migratable fines including the non-expanding authigenic clay minerals (Koalinite, chlorides, etc), expanding authigenic clay minerals (montmorilinite), quartz and carbonates are identified. These fines loosely adhere to the pore walis and are released dus to colloidal and/or hydrodynamic forces exerted by the invading fluids. The released fines in addition to the invaded solids, if present in sufficient quantities in the flowing fluid, plug the pore throat thereby reduce the permeability. The causes of clay dislodgment and mobilization have been investigated by a number of researchers. Two key factors identified are critical velocity and critical salt concentration.2.41.44 Critical Velocity For a given set of circumstances there will be a certain maximum velocity below which the fines in the pores will remain in position. For clays, the critical velocity depends on their morphology distribution and on the nature & strength of the bonds between the pore surfaces and clay particles. It is quite common for authigenic clays in the pore spaces to be only weakly bonded to the pore surfaces (e.g. by van der Waals forces). Sufficiently high flow rates alone can be enough to mobilize such clays. However chemical changes in the pore fluids can further weaken such bonds. 2.4.1.4.2. Critical Salt Concentration It has been discussed that changes in the chemistry of pore waters can increase the susceptibility of formation clays to mobilization. Khilar and Fogler" identified a critical salt concentration (CSC) beiow which permeability reduction arising from fines migration. Whether a particular clay-bearing formation will suffer damage when exposed to brine while salt content is below CSC depends ‘on many factors. Among these are (i) the type(s) of clay, (ii) whether the clays have an expandable component, (iii) the cation exchange capacity of the clays, (iv) the species of cations present in the induced and formation brines and (v) the degree of contrast between the induced and native fluids. in circumstances where the composition of the induced fluids is sufficiently dissimilar from that ofat itu clays, the pH of the pore fluids increases. This pH elevation weakens the forces that hold the clays to the pore surfaces. 2.4.2. Chemical Mechanisms The formation can be damaged by the reaction between the filtrate and pore contents and/or matrix materials. Chemical causes for formation damaged are essentially swelling or dispersion of clays, precipitation by the reactions between mud filtrate and pore contents as well as solution of salts and materials from the matrix, 2.4.2.1 Clay Swelling Almost all sandstones contain clays, which can be affected by the filtrate of the water-based drilling fluids, reducing the permeability of the formation. Detrital clays have been imbedded in the matrix during the sedimentation phase. ‘Theso clays normally form an integral part of the supporting rock matrix and therefore, are not mobile. Diagenetic clays have been transported into the pore channels after the sedimentation phase by underground circulating waters. The last mentioned clays can remain as loosened particles in the pore channels or can cover their surfaces. The kind and content of the electrolytes in the filtrate influence the mechanisms of clay-hydration. The Detrital clays can swell and narrow the pore22 channels. Diagenetic clays can be loosened from the pore surface and transported by filtrate further into formation. They can lodge themselves into the narrow necks of capillaries in other areas of the formation and block them. Clay minerals that are smaller in size and structurally plately, have their surface areas larger and thereby tend to react with injected fluids. The clays most commonly associated with formation productivity impairment are Smectite, ilite, koalinite, chlorite and mixed layer. These clay minerals are sensitive to composition, pH and ionic strength of the surrounding water. Smectite is considered to be the most potentially damaging of these clay groups. This clay will hydrate and swells 20 times its normal size and reduce the radius of flow in a pore, in which it is located and also facilities the migration of particles. However swelling of clays also weakens rock matrix structurally, lowering their specific critical velocity. Illite and Kaolinite, while considered to be non-swelling are hydratable and contribute to the migration 's within the formation. Susceptibility of a particular expandable clay to swelling depends on the mineralogy and chemistry of the clay itself, the chemistry of the fluids with which it is normally in contact and the nature of the introduced fluids. Table-2.1 shows the characteristic of some authigenic clays.23 -ui8ju9 ¢1°9 st vase aDeyins ZyxENg ‘WostsEduIOD 10,1 yjiqvauuad Suronpas sarod sso19e eon snowaueyyy 004-001 Oke] POX, asp pe soo ty ede Sete “Aamqwonuod | “quo Kove 30 6qgom pue Aysorod-ororur Jo sso] sasue:) | ‘SI99YS popu Kawa sedan | OFT OMIV!S (Od SANTV)EN POUT) oot amnoauig, yREpURED %OOL “SAHISNDS 2316/4 eee “ea ygepta says 20 somds oreo im noi, oot ty 40 onosor peg -a8eqquo ‘quooxauoy ‘sare (HO) ["O"AV "(oa ‘Iv SW) 001 eHONY) 4 sso pu Burn sions Ssteo Jeon a1od aif yw oyeataou0 | “says 10 ad poyons 10) [08 Iv 0 areyory pi erin ede yearg : maou wosasrivorvn | aboTONaMON Sinaia woinai van ee ang | STRANI spay doy) amasiyiny SO sonspiapoviny) 307 # AIAN L la24 2.4.2.2 Formation of Insoluble Precipitate The majority of formation damage problems also arise from fluid-fluid and fiuid-rock interactions. Damaging results due to adverse chemical reactions between the introduced fluid either with reservoir fluids or with the rock mineralogy. Differences in chemistry between the invading filtrates and the reservoir fluids could lead to adverse reactions which results in the formation of organic (paraffanic and asphaltenes) and inorganic (CaCO3, BaSO.z, SrSO., and FeCOs) scales. 2.4.3. Thermodynamical Mechanisms There are no stable water-based drilling fluids. By addition of polymers to clay suspension, these fluids can be converted into meta-stable system, in which the phase separation is decreased so that there are virtually no system changes. The stability of water-based drilling fluids is on the one hand affected by electrical charges and by mechanical barriers between the dispersed particles in the system, and on the other hand affected by temperature aging, electrolyte influxes and shearing.25 2.4.3.1. Temperature aging in the borehole, drilling fluids are exposed to dynamic and static temperature aging. The rise in temperature increases the kinetic energy of both the particles in the system, and the molecules incorporated into the drilling fluids. Higher temperature increases also the dissociation of water, so that the pH-value of the drilling fluid decreases. The temperature influences drilling fluids by affecting their hydratable components. Increasing temperature results in splitting of the water-dipole-polymers-linkages and/or water-dipole-clays-linkages and also in degrading polymers. Due to the splitting of water dipole from the hydration layers of the clays and the polymers, the amount of available free water in the mud. increases significantly. The degradation of hydration layers of clays and polymers also results in the formation of a highly porous and permeable filter cake. 2.4.3.2 Electrolyte Influxes & Shearing The osmotic swelling of the clays and the swelling of the dissociable polymers are function of the ion concentration difference between the hydratable particles and the bulk solution. The treatment of the drilling fluids with electrolytes result in a decrease in ion concentration difference, and thus in a degradation of the hydration layers of clays and polymers. The amount of the free water in the drilling fluid increases. The mud shearing causes stretching of the polymer molecules, splitting of water dipoles from the hydration layers of the polymer and26 clays and finally splitting of the polymer filaments by crossing its critical shear stress under extremely high shearing. As discussed above, higher temperatures, electrolyte treatments and extremely high mud shearing can cause the degradation of hydration layers of polymers and clays. Due to the formation of a highly porous and permeable filter cake, higher amounts of filtrate dissipate into the formation and displace the original pore contents that cause more formation impairment. 2.4.4 Bacterial Mechanisms The formation can aiso be damaged by the colony of bacteria and their precipitated products, blocking the pore channels. Bacterial formation damage can occur both with and without oxygen present. A particular type of bacterial family (sulphate reducing bacteria) has been responsible for the souring of many previously sweet gas reservoirs. HS gas generated can be very corrosive and extremely lethal to humans.CHAPTER 3 PARTICLE & PORE SIZE DISTRIBUTION As observed in the previous chapter, pore plugging due to solid invasion or fluid-fluid / fluid-rock interaction is one of the main factors involved in formation impairment. In this chapter, the influence of mud particle and pore size distribution on formation damage is discussed in detail. Formation damage from drilling fluid solids is strongly dependent on the pore and pore throat size distribution of the formation, particle size distribution in the drilling mud and wellbore overbalanced pressures. Their effects become more pronounced if fissures and natural fractures are also present in the vicinity of the wellbore. Measurements of particle size distributions of solids in the muds and calculations of rock pore sizes can be used to design competent muds. 2728 Adding of proper sized bridging materials to drilling muds can minimize solids invasion and formation impairment The effects of drilling mud particle size and rock pore size distribution on regain permeability of the formation were investigated by many researchers ©1771 Their results are summarized below: © Invasion and formation damage occur with all muds. © The depth of invasion and level of impairment can be controlled, to certain degree, by designing the mud to include bridging material. © The effectiveness of the bridging material in reducing invasion is a function of the concentration and particle size of the material and of the pore sizes of the formation rock. * Damage is most likely to occur in higher permeability formations because large pore & throat size favors solid invasion. 3.1. Bridging Process During drilling operations, initially, there is a mud spurt at the start of mud circulation across the formation face before filtration properly begins and thereafter filtration volume becomes proportional to the square root of the time interval. Mud spurts may be larger when filtration takes place against the more permeable rocks. In fact, they can be infinite (i.e. circulation is lost) unless the29 mud contains particles of required size to bridge the pores of the rock and thus establish a base on which the filter cake can form. Only particles of a certain size relative to the pore's size can bridge. Particle larger than the pore opening cannot enter the pore, and are swept away by the mud stream. Particles considerably smaller than the opening invade the formation unhindered; but particle of a certain critical size stick at bottlenecks in the flow channels, and form a bridge just inside the surface pores. Once a primary bridge is established, successively smaller particles, down to the fine colloids, are trapped, and thereafter only the filtrate and ultra-fine colloids invade the formation. The mud spurt period is very brief, matter of a second or two at the most. As a result of the process just described, three zones of mud particles are established on or in a permeable formation as shown in figure 3.1 and presented below; 1. An external filter cake on the walls of the borehole. 2. An intemal filter cake, extending a couple of grain diameters into the formation. 3. Azone invaded by the fine particles during the mud spurt period, which normally extends about an inch into the formation.30 ‘vo FLOW ——> BRIDGING ZONE Figure 3.1: Invasion of a permeable formation by mud solids.a Experimental results reported by Krueger and Vogel""* suggest that initially fine particles (during spurt losses) do not cause much permeability impairment, but do so afer filtration had proceeded for some hours, presumably because of migration and consequent pore blocking. 3.2 Critical Particle Size With regard to the critical size required for bridging, it was shown by Coberly! that because of jamming, particles down to one-third the size of a circular screen opening would bridge that opening. Abram®! also reported that particles whose median diameter was about one-third the median pore size of a 5-darcy sand pack would bridge that pack. In order to form an effective base for a filter cake, a mud must contains primary bridging particles ranging in size from slightly less than the targest pore opening in the formation about to be drilled, down to about one-third that size. In addition, there must be smaller particles ranging down to colloidal size, to bridge the smaller formation pores and the interstices between the coarser bridging particles. Contrary to above, Mally”) pointed out that one third bridging rule does not always apply. He stated that at a relative low fiow rate, even larger particles but smaller than the existing pore size, can be produced one after another with out bridging and plugging. Yan and Jiang" in a research study also reported that32 optimal effect of bridging occurred when particle diameter is 1/2 ~ 2/3 of pore size. Thus, in order to obtain optimum particle size, the best way is to determine primary-bridging sizes by making trial and error tests on cores of the formation of interest. In this context, researchers®#225 found that particles less than 2 microns in diameter will bridge rocks of permeability less than 100md; 10 micron particles will bridge consolidated rocks of permeability between 100 and 1000 md; and 74-micron (200 mesh) particles will bridge sands up to 10 darcies. A mud containing a suite of particle sizes upto maximum of 74-micron should bridge and form a filter cake on all formation except those with macro-openings, such as gravel beds and formation with open fractures. 3.3 Particle Concentration In 1994, Ismail during an experimental study regarding the effect of solids concentration on formation damage, showed that the fluid loss increased as solids concentration in the mud increased. Whereas Abram" recommended that the concentration of the bridging solids must be at least 5 percent by volume of the solids in the final mud mix. He also explained that the greater the concentration of bridging particles, the quicker bridging will occur, and the less will be the mud spurt. On consolidated rocks with permeabilities in the range of33 100 to 1000md, about 1-Ib/bbl (2.8 kg/m) of the required size range is sufficient to prevent the mud spurt from invading further than an inch into the rock. In general, increasing the solids concentration in the mud appears to improve permeability recovery when the cores were backflushed. To reduce formation damage and improve recovery, suitable size, shape and concentration of solids in the mud must be properly selected. 3.4 Effect of Particle Size and Shape on Cake Permeability Mud filter cake permeabilities decrease sharply with mud particle size. In 1943, Krumbein and Monk? investigated the permeabilities of filter cakes of river sand by separating the sand into ten size fraction and recombining them to obtain two sets of mixtures. In one set, the mixtures had increasingly large mean particle diameters, but all had the same range of particle sizes. In the other set, all the mixtures had the same mean particle diameter, but increasingly wider ranges of particle sizes. The results showed that cake permeability decreased (1) with mean particle diameter, and (2) with increasing width of particle size range. One might expect minimum cake permeabilities with an even graduation of is of Bo et al! showed that the minimum particle sizes. However, the experi permeabilities were obtained when there was an excess of particles at the fine end of the scale, and not when the size distribution curves were linear. It would34 appear therefore that a uniform graduation of particle sizes is of secondary importance, but obviously there must be no major gaps, or the finer particles. would pass through the pore openings between the larger ones. Mostly drilling fluids contain substantial amounts of colloids, whose size may range down to less than 10° microns. The permeabilities of their filter cakes depend almost entirely on the proportion and properties of the colloidal fraction. Cake permeability is influenced by the kind of colloid as well as by the amount and particle size. For instance, filter cake of bentonite suspension in fresh water have exceptionally low permeabilities because of the flat, filmy nature of the clay platelets, which enables them to pack tightly normal to the direction of flow. 3.5 Effect of Pore Size Distribution on the Oil Recovery Primary oil recovery is affected by the wettability of the system and a wateravet system exhibits greater primary oil recovery. The relative flow of fluids is a function of pore size distribution. Therefore any change of this distribution due to blocking would change the relative permeability curves. Furthermore, formation having smaller pore sizes exhibits an increase in the irreducible water saturation and the residual oil saturation in water-wet rocks thus decreases the mobile oil saturation.35 3.6 Particle & Pore Size Distribution and Fluid Design Particulate invasion is one of the primary mechanisms of formation damage caused by drilling fluids. High spurt losses, deep solid invasion and poor permeability recovery are observed if muds are used without the addition of solid particles. However, it is possible to mini solids invasion and formation impairment by adding bridging material to the muds. Bridging of pores by particles is required to initiate filter cake formation. This bridging material should be chosen by matching its size to the formation rock pore sizes. A bridge may be initiated when two large particles (< pore size diameter) start to move into an opening at the same time and lodge against each other. Other smaller particles may bridge the opening between the larger, previously bridge particles. If the proper particle sizes are present, this process may continue until the openings become too small for any contained solids to penetrate. It is at this time that only the filtrate and ultra-fine colloids flow through the filter cake. The shape and concentration of particles also play an important role in the bridging process. Therefore, wellbore fluid should be designed so that the solids bridge on the surface of the formation rock rather than insides the pores and form a low permeability filter cake that allow minimum filtrate and ultra- fine colloids invasionCHAPTER 4 LITERATURE REVIEW Ever since Van Everdingen"! & Hurst” introduced the idea of a skin factor to the Petroleum industry in 1953, the primary research focus has been on evaluation and minimizing the formation impairment. The impact and importance of this skin factor on the economical production of crude oil seriously engag: the attention of many researchers in the last five decades. In these efforts, the effects of mud characteristic, Reservoir behavior and drilling & workover operation conditions on the formation damage have been studied with physical, analogue, analytical and numerical simulation models. To achieve high production rate and avoid gas & water coning problem, horizontal wells getting preference over their counterpart vertical wells. But most of the horizontal wells are completed open hole and are more likely to face this 36a7 damaged zone formed during drilling operation around wellbore where a cased vertical well overcome it by perforation. in this regard, the influence of skin on the performance of horizontal well for various reservoir, production and drilling ‘operation conditions have been studied widely in recent years. These research studies investigated many parameters involved during drilling and production conditions by experimental, analytical and numerical techniques. A brief review of literature is presented below focusing the formation damage research studies induced during drilling operation. Abram (1977)", emphasized the addition of bridging material to the drilling mud to minimize solids invasion and formation impairment. He defined two rules for selecting the size and concentration of the bridging additives. (i) The median particle size of the bridging additives should be equal to or slightly ar than 1/3 the median pore size of the formation. (ii) The concentration of the bridging solids must be at least 5% by volume of the solids in the final mud mixture. Laboratory flow test on two different types of rock system were conducted (i) High permeability (4-6 Darcy) unconsolidated Brezas River sand, tests were performed on a radial model packed with above sand under 50 psi differential38 pressure and mud (water based) rate of 2-3 fumin for 20 minutes. (ii) 5-50 md Dolomoite cores in linear model. Experimental results with competent muds showed that impairment caused by invasion of mud particles occurred to a depth of less than 1 inch, Studied system has no significant effect of this impairment on productivity and injectivity because it was easily penetrated by perforation. Gruesbeck & Collin (1982)"", described the results of an experimental study performed to determine the factors affecting entrainment and re-deposition of naturally occurring fine particles in porous media, leading to abnormal decline in productivity of producing wells. Authors proposed a phenomenological theory of entrainment and di ition to determine local laws of deposition and entrainment. The central concept of this theory was representation of both particle and pore size distribution by partitioning the porous medium at any cross section into parallel plugging and non-plugging pathways. The results showed a minimum interstitial fluid velocity for fine entrainment is required, which depends on properties of the porous medium and contained39 fiuids. Transition from single- to two-phase fiow can result in a reduced criticat velocity and enhance entrainment. Peden, Arthur, & Margarita (1984), provided an overview of an experimental study on the dynamic and static filtration characteristic of drilling mud. Effects of fluid loss controlling factors such as annular velocity, fluid temperature, hole angle, shear rate and effect of the filter media were investigated. Experimental setup was made up of a mud circulating system across the face of the core through annular space. Filter medium consisted of four type of core plugs, Aerolith 5, Aerolith 10, Diapor (synthetic mixed silicate material) and Elgein (naturally occurred sandstone of size ¢-inch x 3-inch for dynamic and 1- inch x 1-inch for static runs). Three fluid systems were used. A 9.7 ppg Calcium chloride brine and two water based Gypsum-lignosulphonate & KCI Polymer muds. The results showed that the ability of mud to limit spurt loss and filtration was improved by addition of bridging particles such as CaCOs, gypsum, Barites. Equilibrium dynamic fluid loss rate increased with increasing annular velocity.40 Regarding the effect of temperature and hole angle, they observed that temperature influenced dynamic fluid loss through parameters such as shear stress, filter medium and fluid undergoing filtration. For angled hole, fluid loss is greater for the high side than for the low side of the hole due to difference in cake thickness and composition of the cake obtained. They also found that in static filtration, filtration volume and spurt loss is proportional to pressure applied while increase in concentration of barite as the weighting material result in a reduction in fluid loss, an increase in cake thickness and cake permeability. Vaussard & Martin (1986), carried out an experimental research study on the filtration of drilling fluids under borehole condition and on its main consequences such as differential pressure sticking, rate of penetration and damage to formation. Laboratory tests were performed in a special apparatus with three types of mud (i) bentonite freshwater mud (ii) fresh water polymer mud (iii) inverted olf emulsion mud, Experimental equipment consisted of (j) dynamic filter press to allow filtration measurement while controlling the shear rate against the filtration surface using a cone-and-plate assembly. (ii) filtration & sticking simulator, to simulate the filtration process under differential sticking. (ii) drilling Simulator allows rotary drilling of drilling samples using drill bits to demonstrate filtration during drilling operation under the drilling bits.a They concluded that the dynamic filtration is independent of permeability. Filtration and cake formation depend on two antagonistic phenomena; deposit of solids and erosion. An efficient cake is one where solids have a high colloidal nature. It is overall physico-chemical equilibrium that allows good filtration control under borehole condition. R.F. Krueger (1986)"", in his study regarding impact of formation damage on well productivity indicated that almost every operation in the field is a potential source of damage to well productivity. The remedy of formation damage is usually difficult and costly, the basic approach should be to prevent damage. His investigation revealed that the loss in well productivity varies from operation to operation and reasons are usually associated with either the transport of fine solids, chemical reactions or a combination. Salinity changes and shear by moving pore water may trigger the release of weakly bonded clays, mica, feldspar and other minerals from pore walls. The released particles will then flow with the water, bridging at pore constrictions. Salinity shock is also 2 source of fine release in water-sensitive formations when there is a large difference in salinity between the invading water and the formation water. He also observed that the permeability of core flooded with potassium based filtrate are less subject to damage by fresh mud. Potassium brine fluids42 are more stable and not washed out during the drilling operation. Damage from mud solids is strongly dependent on pore size distribution of the formation, particle size distribution in drilling fluid and wellbore overpressure. He concluded that in order to obtain maximum well productivity from the initial completion and to maintain it during the subsequent routine production operations requires a fundamental knowledge of the causes of formation damage, the ability to anticipate problems in advance and diagnose them correctly (if they occur), advance planning and effective supervision. Control of formation damage also requires emphasis on quality control of chemicals, fluids and treating and operating conditions. Schaible, Akpan & Ayoub (1986)*%, presented a field case study of a highly formation damage offshore oil well. Damege was susceptibly took place during the completion operation. Authors investigation was directed toward the techniques employed to identify the presence of the formation damage, evaluate its possible sources and formulate an acid system, which would effectively treat the formation damage. Formation impairment was identified by means of transient pressure test, Nodal analysis and Production logging. The integrated results of these techniques showed a skin of +209.a X-ray & SEM analyzed the clay swelling and fine migration as a main source of formation damage. Experimental runs were also carried out with representative samples and used completion brine, to confirm the basic cause of damage and to develop an effective treatment system. After performing the acid treatment, authors got a skin damage of +15.8 (by transient test) and increase of 280 % of initial permeability to oil for acid treatment. Claux Marx and S. Rahman (1987)°", introduced a new formation damage evaluation method, caused by drilling fluid. This method was especially designed to evaluate the formation damage in reservoir having pressure lower than hydrostatic pressure and of special interest for EOR and underground gas storage projects. Authors used pressure difference, temperature, annular velocity and contamination time as the influencing variables to evaluate the damage. Hassler cell type core holder that could take 1-inch diameter and 10-inch long core was used. Experiments were conducted under overburden pressure of 2100 psi, differential pressure of <_1400 psi, annular velocity of 2 m/s, temperature of < 150°C and contamination time upto 5 hours.ra They found out that the differential pressure doesn't lead to severe formation damage but only in first 2 inch of core section due to invasion of mud particles (@ T=70°C, V;=0.8 m/s, t=1 hour). They also observed that above 158°F, temperature is very influencing factor in formation damage and damage ratio decreases sharply much more in low permeability formation, probably due to degradation of polymers in the mud. Higher the annular velocity, more filtrate would get into formation and more high formation damage. Especially for lower permeability formation, contamination period is highly influencing period and damage ratio and sectional damage ratio decreases with the period. Marx et al, recommended that the evaluation of the formation damage caused by drilling muds must be carried out under conditions that closely simulate the actual borehole conditions. $.D, Joshi (1987)®" , reviewed the state-of-art of horizontal well and drainhole technology and provided guidelines for the initial evaluation of horizontal well and drainhole drilling prospect. Horizontal well characteristic and applications such as in thin, naturally fractured, tight, gas & water conning problem reservoirs were discussed.45 He observed that for a given length of F rizontal well, relative productivity impairment over an unstimulated vertical well is higher in a thin reservoir than in a thick reservoir. The influence of well eccentricity on well productivity is very small. He concluded that the technology to drill and complete horizontal well is available. Methods to forecast oil production from horizontal wells and drainholes have been developed. S.D. Joshi (1988), with the help of analytical solution and correlation of numerical model, summarized production forecasting methods for horizontal well. Methods were suggested for thin (upto 50 meters) single phase flow, solution gas drive, naturally fractured and bottom water drive reservoirs. The study included comparison of productivity for vertical stimulated wells with horizontal wells. He found that horizontal wells are effective in thin reservoir with high vertical permeability. Horizontal well facilitates many high initial production rate, without premature water conning than those from vertical well, although its rate decreases rapidly than that for vertical well because of decline of oil column quickly. Horizontal well takes longer time for water to breakthrough than vertical well and cumulative oil production from horizontal well is higher than that from a vertical well46 He concluded that in reservoirs with top gas and bottom water a horizontal well would be more appropriate to minimize coning problems. $.D. Joshi (1988)™", came up with an mathematical equation to calculate the productivity of the horizontal well using potential-fluid theory. The equation can also be used to determine the influence of anisotropy, height, well drainage area and well eccentricity on productivity. Laboratory experiments with an electrical analog were also conducted, which showed good agreement with the theoretical equation indicating it accuracy. In this study, Joshi also compared vertical-, slant-, horizontal well productivity indices under similar condition and found that for a 100 ft thick reservoir, horizontal well productivity is two to five times greater than vertical-, or slant well productivity depending on anisotropy ratio and if the thickness is above 400 ft, vertical and slant perform better if anisotropy ratio < 1, The results indicated that horizontal wells are suitable for reservoirs that are thin, show high vertical permeability or exhibit gas and water-coning problems. The developed equation is useful in initial evaluation of a horizontal well drilling proposal. A slant well with a 60-70 degree angle to vertical through the pay zone produces oil at a rate of 2-3 time faster than a vertical well undera7 similar pressure draw down condition. A minimum slant of 35 degree is required to increase oil production at least 1.2 time that of a vertical well. Kenneth E. Porter (1989), presented an overview of formation damage induced during various operations. He reported that the best wells are often the ones which expose the formation to the mud system for less amount of time, no matter what kind of fluid is being used. Na-compounds should be avoided because it disperses the natural clays in the reservoir rock and destabilizes the non-reactive fraction of the reservoir clays. He mentioned a "Reverse Funnel Effect" phenomena that it is easier to fiow a fluid into the formation, where more openings are available for the fluid to pass through as it travels farther into the formation but it is very difficult to get this same material to fiow back into the well bore from the reservoir. It is easier to get slid Jaden mud filtrate to flow into the formation than it is to get it out. These two effects can tremendously alter the reservoir up to 5-10 ft from the wellbore. He often observed in the field that filter cake is composed largely of drilling cuttings 90% +, with polymers, Bentonite and fiuid loss additives making up the fluid coating on the core face. He agreed that by using the right polymers, one should be able to create a kind of sand control by gluing the sand grains48 together. This is more beneficial for horizontal wells where the entry velocities for fluid coming into the wellbore are dramatically reduced. He recommended the use of stabilized chemicals in the drilling fluid all the time. Synthetic brines of same cations and anions as formation water must be used for permeability determination to examine the situation as close to reality as possible. S. Rahman and Marx (1991), conducted an experimental study to evaluate the formation damage caused by drilling fluids and cement slurry. They found that drilling fluids weighted by ground calcium carbonate caused about 40% in permeability deduction over 20 cm length, where else bentonite polymer drilling fluid impaired about 75% only in first 5 cm. On the basis of these results they concluded that bentonite polymer mud is more beneficial for cased hole is because reduction in 75% permeability can easily be overcome by perforating work. Experimental setup consisted of a dynamic flow loop, which contained 2 core holder that could take 25 cm long and 2.5 cm diameter cores. The differential pressure of 36 bars, temperature of 70°C and annular velocity of 0.8 m/s were maintained during all experiments. Two types of core sample of 5-15 md and 1500-2500 md permeability and 17-20 % and 20-24% porosity49 respectively were used in the study. Four drilling fluids, Bentonite suspension, Bentonite suspension weighted by barite, Bentonite suspension weighted by ground calcium carbonate and ground calcium carbonate suspension were used. The result showed that the Bentonite suspension caused less permeability impairment than other drilling fluid. The reason behind it is that the Bentonite suspension contains sufficient amount of bridging particles of sizes less than the average pore diameter, which formed a thick layer of very low permeable internal filter cake. Renard & Dupuy (1991)! , provided a basis for comparing the flow efficiencies of vertical and horizontal wells. Analytical expression was derived assuming steady-state flow of an incompressible fluid through an anisotropic medium with no flow condition at the top and bottom of horizontal boundaries. The comparison considered an altered zone of the same radius and reduced permeability around the vertical and horizontal wellbores. Results of the analytical expression indicated that the skin damage is less detrimental to horizontal wells than to vertical wells upto a relatively large permeability ratio value. Lost production in a horizontal well is more than a vertical well but flow efficiency is always greater for horizontal wells.50 Authors also observed that anisotropy ratio magnifies the influence of the skin damage of horizontal wells and increasing the length slightly reduced the influence of skin on flow efficiency. Jiao & Sharma (1992)"! , conducted a series of drilling mud invasion experiments on Berea sandstone cores to measure the extent and depth of formation damage by mud filtrates and particles. In this investigation, a specially designed dynamic filtration cell was used to study permeability impairment under dynamic flow conditions. Driling muds were circulated across the faces of the core and dynamic fittration rates were measured. The extents of formation damage were measured by calculating the retum permeabilities, Seven types of Water Based Mud were used to study the effect of particle concentration, particle size and mud filtration properties on permeability impairment. Experimental setup consisted of a standard hassler sleeve core holder with five pressure taps along the length of the core holder. Berea cores of 8 inch in length and tinch in diameter were used. Muds were circulated at the rate ofat 0.044 dm%s with overbalanced pressures of 90-95 psi for 10 hrs.. The confining pressure of 1000 psi was also maintained. From the study they concluded that the higher particle concentration in muds, results in the formation of stable particle bridge that prevent particle invasion. The quality of the extemal mud cake is a function of the particle size, concentration and the mud additives. If the saft concentration in the mud filtrate is less than the critical salt concentration (that required to prevent fine migration) then formation damage by fine release and migration can be quite significant. They also observed that the depth of mud particle invasion is a function of mud composition, Formation damage is permanent, once the mud particles have invaded the rock, it is difficult to remove them by back flow. They recommended the need to formulate such mud that can rapidly form a low permeable, high strength external mud cake. Ghofrani, Alaboudi & Sengupta (1992)"", described the results of a research project about the extent of damage caused by clay based and clay free inhibitive fluids in two consolidation sandstone formations. In this work, authors studied the role of K, Mg or Ca, clay & solid bridging material (chalk) on dynamic filtration characteristics and rheological properties of Drilling fluid. They also82 investigated the effects of increasing temperature and differential pressure on formation damage. The experimental setup contained a closed loop circulation system. Hassler cell was used as core holder that could accommodate 1-inch diameter and 10-in long core. The experimental conditions were, temperature 90°C, circulating pressure upto 10Mpa (685 psi), confining pressure 15 Mpa (1000 psi), fluid velocity along the face of the core upto 3 m/s and circulating time thr. ‘They observed that the presence of clay and pore bridging materials are the main factors influencing the extent and value of damage in high permeability sandstone. In low permeability sandstone, polymer molecules blocking the comparatively narrow pore channels mainly cause the damage. It was also found that the addition of pore bridging materials (chalk) brings down the value of damage significantly. Peng & Peden (1992), conducted an experimental and theoretical study on filtration properties of water based drilling fluids under dynamic and static condition. They came up with a dynamic filtration equation using a mass balance method based on the cake filtration theory prevailing in chemical engineering industry.53 The results of dynamic filtration equation came out in substantial agreement with the experimental investigation carried out as a function of mud types, solid concentration, pressures and sheers rate. The experimental setup was similar to that used by Vassard 4, Two mud systems were used, —seawater/KCl/Polymer and_-—sFresh Water/Gypsum/Ferrochromo Lignosulphonate mud. All static filtration tests were performed for two hours and dynamic filtration tests were conducted for 8 hours on Clashach sandstone. ‘They concluded that the filter cake formed by drilling fluids is compressible and compressibilities decrease with increasing solids concentration. Jioa and Sharma (1993), used a specially designed dynamic filtration cell to study the dynamic filtration of 30/70 water/oil invert emulsion muds with different additives. Experimental setup included @ core holder having ports along the length and one-end piece has two ports to circulate the mud. 30/70 water /oil invert muds with different additives were circulated across the Berea sandstone (105.32 md permeability & 22.3 % porosity) of 8-in in length and tinch in dia54 under 1000 psi confining pressure. Permeabiliies before and after the mud circulation were determined as a criterion for mud performance. During their study, they found that the use of water-wet solids result in very poor-quality extemal mud cakes and high fluid loss rates. Better external mud cakes can be obtained by adding organophilic clay to the mud. Higher solid concentration provides less spurt loss and equilibrium filtration rates. They also observed that the dynamic filtration rate is much higher than API static filtration rate. Addition of filtrationtoss control additives impede solid invasion and intemal mud cake formation due to reduction in spurt loss. Shear resistivity of invert-emulsion mud cake is less than water-based mud cake. Jiao & Sharma (1994)""|, proposed 2 simple model that predict a power ship between the filtration rate and shear stress at the cake surface. Cake growth model were developed by performing a force balance on the particles at the cake surface and two mechanism of particles release, sliding and rolling were evaluated. Parameters such as the shear stress on the cake surface, the instantaneous filtration flux and the cake permeability were incorporated in the model.55 The results of the model were found to be consistent with the experimental data at different filtration times at various suspension flow rates using three different suspensions. Experimental setup was consisted of crossfiow filtration cell that could take 8-inch long and 1-inch diameter core plugs. Berea sandstone of brine permeability of 103.4 md and 21.4% porosity were used as filter medium under 1000 psi confining pressure and 100 psi overbalance pressure. Three suspensions (i) Bentonite + water (ji) Bentonite + 2% NaC! & (il) Bentonite + 2% NaCl + 3% Lignosulphonate were used during study. Laboratory results showed that cake growth is a function of shear stress, higher shear rates result in thin cake ultimately produced high equilibrium filtration rates. Hydrodynamic forces tending to deposit the particles on the cake surface is proportional to the fluid loss rate while the shear stress tending to entrain the particle in the flow stream is independent of fluid loss rate. As the cake thickness increases, fluid loss rate decreases, smaller and smaller particles will be deposited on the cake surface. Mechanism of cake growth gives rise to a heterogeneous cake with both large and small particles at the bottom and only small particles in the upper portion of the cake. They concluded that thickness of a mud cake is a function of rheology, mud shear rate and core permeability. A minimum overbalanced pressure is56 required in order to form a filter cake during dynamic filtration process. This pressure requirement increases with decreasing core permeability. Initially mud cake is inhomogeneous with smaller and smaller particles being deposited as filtration process proceeds. An equilibrium cake thickness is achieved when no particles small enough to be deposited are available in the suspension. Frick & Economides (1993), presented a conceptual and mathematical description of the damage to a horizontal well. An analytical expression incorporating skin effect and anisotropy ratio was developed. They mentioned that vertical to horizontal permeability anisotropy, the time of exposure to drilling, completion and workover fluids have great influence on the shape and distribution of damage around a horizontal well. During drilling, mud filtrates penetration would generate a truncated cone with the larger base near to vertical section of the well. The bases of this concentration would be radial in the case of permeability isotropy and elliptical in all cases. They also observed that the damage around the horizontal well can reduce it productivity index significantly. A matrix stimulation approach using coll tubing with changing withdrawal rates have a profound beneficial effect. A partial stimulation approach found as an optimum and economical solution.7 Tovar, Azar, Sharma & Pilehvari (1994)"", conducted an experimental study to investigate the mud solids and filtration invasion in reservoir rock while using drilling muds. The main objectives of the study were to evaluate and select the most compatible drilling fluids on different reservoir rocks and to delineate the pore blocking mechanism. On the basis of filter cake properties and the laboratory test data they proposed a permeability reduction mechanism in the reservoir rocks. During their study, they observed that higher the initial rock permeability, higher and much deeper were the impairment in the rock and vice versa. The drilling mud particles penetrated well into interstices of the formation and causing significant reduction in absolute permeability of the rock. They also mentioned at the location and extent of damage varies with other parameters such as mud annular velocity, differential pressure. and characteristic of the rock as well as the characteristic and composition of the mud filtrate and solids that flow with it under dynamic conditions. Experimental setup consisted of a dynamic flow loop, with a core holder having five ports along the length. Muds were circulated @ 120-240 ft/min across Berea sandstone, limestone and dolomite cores of 1-in diameter and 3-in in length under temperature of 150°F, 1000 psi confining pressure, 100-200 psi overbalance pressure and 50 psi back pressure. Different types of oil based mud and water based muds were used in the study.Ismail, Peden & Arshad (1994)"* , studied the effect of solids concentration and filter media on fluid loss and permeability recovery of the cores in an experimental investigation. KCl-Poimer muds of different barite concentrations and different types of core were used to perform the test. Scanning Electron microscopy (SEM) was used to identify the solids impairment at the core face. The result showed that the increase in permeability recovery during backflow is obtained with increasing solids concentration. Poor permeability recovery and high spurt loss was observed with mud containing polymer and no solid particles. ‘They also observed that the filtration and permeability restoration is not only the function of size, shape and solids concentration in the mud but also affected by pore size distribution and core characteristics. They recommended that properly sized, shaped and concentration of solids must be added to the mud to reduce formation damage and improve recovery. Better permeability recovery can be expected if the pore size of formation is smaller than the drilling mud particles because there is no internal bridge to block the fluid fiow. The lower the initial permeability the higher the retum permeability. Experimental setup contained a core holder that could take a core plug of 14-inch in diameter and 4-inch in length. The mud was injected into core under59 overbalanced pressure of 300 psi for 2 hours. Two synthetic cores, Diapor and Aerolith, having permeability of 10-40 & 1800-2300 md and porosity of 55-59% & 44-48% respectively and one naturally occurring Clashach core were used with five types of KCl-Polymer mud having different concentrations of barite. Beatty, Hebner, Hiscock and Bennion (1995) , carried out laboratory and field case studies to evaluate and minimize formation damage in Horizontal wells. They mentioned that although damage due to drilling mud cannot be eliminated but it can be minimized by selection of the optimum drilling fluid with the help of proper experimental techniques. They recommended that proper simulation of the drilling mud invasion into the formation and parameters such as fluid rheology, solids content, size distribution, overbalanced pressure, formation permeability and wettability should be taken into consideration to provide a good basis for proper mud selection, The experimental setup consisted of a core holder having two ports at one end to circulate the mud across the core face of 1.5 in diameter. Experiments were conducted under 715 psig overburden pressure and 37°C temperature. They concluded that the formation damage induced by drilling mud has a propensity to cause severe productivity reduction in horizontal wells. Since formation damage is highly reservoir specific, simulation of the driling mud60 invasion process on representative core material can assist in the selection of optimum mud system.. Jack. D. Lynn (1995)*", reported a new test apparatus and procedure to evaluate permeability damage caused by drilling fluid in the near well bore region of the reservoir. The apparatus was effective at comparative testing and capable to circulate slurries containing large micro-size bridging solid without allowing settling of the particles. Several damage effects such as solids invasion and fines mobilization through excessive production rates can be reproduced by the help of this device. Experiments were conducted under 165°F temperature, 1500 psig confining stress, 400-psig pore pressure and 200 psig overbalanced pressure. He concluded that the higher permeability samples show more damage with a given mud species because the higher permeability samples would have 2 large average pore entry radius and allow greeter invasion of solids and a greater depth of invasion of solids. Among the five drilling fluids, oil based mud provided the best return permeability than water based mud,61 Gruber & Adair (1995), in a research study indicated that the previous leak-off test methods are oversimplified and don't consider the real wellbore cleanup procedures. They came up with a new comprehensive test procedure that used incrementally increasing pressure across the sample cores to measure the regain permeability. They found out that no fluid flow would occur until a minimum “threshold differential pressure" is achieved. The threshold differential pressure varies between different mud systems and rock permeability. So regain permeability is a function of increasing pressure drop and volumetric through put. Experimental setup consisted of a hydrostatic core holder with one end having two ports to circulate the mud and one port at other end to collect filtrate. Two types of mud were used (i) a standard Gel-chem mud (ii) water-based mud containing an oil-soluble resin of density 7,030 kg/m3 and 1005 kgim3 voly. The muds were circulated across the core (1.5 in dia and 2.0 in length) for 2 hrs at the rate of 2.0 mi/minute under 2770 psia overburden pressure, 1380 psia pore pressure, 493 psia overbalance pressure and 46°C temperature. Authors determined the permeabilities before and after the damage of the core to measure the regain permeability. They observed that the regain permeability depends on the imposed pressure drop across the core as well as62 the cumulative volumetric throughput. For high permeability samples, threshold pressures are low while for low permeability samples the pressures are high. They concluded that the regain flow capacity is a function of differential pressure, volumetric throughput, rock properties, drilling fluid properties and drilling parameters. Li, Rosenberg, Argillier, Durrieu and Montes (1995)! » performed static and dynamic filtration experiments to correlate the filtration behavior of water based muds with structural properties of the cake. The cake was analyzed by cryo-scanning electron microscopy. Authors used suspension of bentonite as @ base fluid and studied the influence of addition of monvalent salt and water-soluble polymers (Polyanionic cellulose polymer (PAC) & Sulphonated terpolymer of acrylaminde (STP)). Experimental setup consisted of an API Static and Dynamic filtration apparatus. Filtration tests were conducted on filter papers and rock slices under a pressure drop of 7 bars. It was found that the dynamic filtration leads to a cake texture characterized by a more regular network and smaller pore size. The effect of63 shear rate decreases with the addition of polymers because of high degree of dispersion and establishment of connection among particles. They also observed that the addition of salt NaCl lead to an increase filtrate volume due to screening of the electrostatic repulsion between the clay sheets with NaC! that induces some aggregate of the clay particles & result in a more permeable cake. Conversely, in presence of STP Polymer, fitrate volumes are much smaller due to build up of less permeable cake. Fraser, Reid, Williamson & Enriquez (1995) , in a laboratory study investigated the mechanism and nature by which filter cakes develop against sandstone rock face with different type of drit-in fluids. The study demonstrated by dynamic fluid loss measurement, imaging of dried and wet filter cakes using SEM & Environmental SEM respectively. Authors used sized salt and carbonate-weighted polymer mud and un- weighted mixed metal hydroxide (MMH) in their investigation and found MMH fluid cake less damaging to formation, less permeable and easily removable by wash fluids or simply by application of back pressure. Experimental setup contained a dynamic filtration apparatus that could accommodate 25.4 mm (2.54 cm) diameter and 30 mm (3 cm) long core. Staticand dynamic tests were conducted for 1 hr under applied pressure of 200 psia. Clashach sandstone, Birchover sandstone and Ketton limestone cores were used in the study. Longeron, Argillier & Audibert (1995)*" , developed an integrated experimental approach to study the mechanisms of mud cake formation, mud filtrate invasion and formation damage induced by drilling muds in oil bearing formation. This approach included (i) Characteristic of mud properties (li) Filtration experiments through paper filters and mud cake observation using Cryo-SEM (iii) Filtration tests (static and dynamic), Experimental setup included two mud leak-off apparatus to perform filtration tests on 40-cm long and 5-cm diameter core samples. The first apparatus consisted of a tri-axial core holder sell made in aluminum to monitor in-situ saturation by mean of X-ray CT scanning. The maximum working pressure and temperature were 300 psi & 100°C respectively and mud was circulated for 20 hours. Second mud leak off apparatus was a dynamic filtration core holder cell equipped with five pressure taps along the length of the core holder. The core sample was placed under a confining pressure of 1000 psi, 80°C temperature and 150 psi overbalanced pressure. For second apparatus two polymeric based muds were used, first was salted formula with 30 gmv/L KC! and other non-salted FormulaThey concluded that the spurt loss & cumulative volumes of filtrate are the function of oil viscosity and overbalanced pressure. Higher the oil viscosity, lower the spurt and filtrate volume. Dynamic filtration gives higher invasion rate and earlier filtrate breakthrough than static filtration under same overbalanced pressure. Hystersis of saturation, due to trapping of aqueous phase, induce additional damage due to adverse permeability effect. Ryan, Browne & Burnham (1995), in a joint industrial project, studied the parameters which control mud damage and its clean up with breaker systems to develop a more systematic approach to fluids selection and the design of horizontal completion operations. Experimental work was carried out on both small and large scale. Small scale laboratory test was conducted to quantify mud damage & test breaker effectiveness while large scale flow loop testing was used to evaluate overall clean up technique including displacement efficiency and damage to pre-packed screen completion. Small scale testing was carried out in a specially engineered piece of test apparatus. Clashach core plugs of 2.36-inch in diameter and 3-inch in length were used. Six mud systems were used in dynamic filtration runs circulated over the core face with shear rate of 270s" and 209 psi overbalanced pressure for 468 hours. Large-scale flow loop consisted of a designed wellbore simulation model 40-m long with 8-1/2 inch hole. Synthetic core made up of resin coated sand blends were used. Static filtration tests were conducted for 16 hours The results showed that oil based muds have lower breakthrough pressure and damage level than water based mud. Complete extemal filter cake removal is not necessary to produce oil through the formation. Shaw & Chee (1996), in a laboratory study, evaluated the formation damage mechanism and quantified the damage effect caused by the newly formulated drilling mud system. Results of their experiment showed that the most important parameters affecting the regain permeability are, (i) mud composition, (i) core permeability (ji) overbalanced pressure (iv) filter cake rigidity (v) pressure drawdown during backfiow (vi) mud particles Vs pore-throat size distribution. They concluded that the bentonite and cellulose based bridging agents appear to cause high degree of damage than ground calcium carbonate. Repeated removal of existing mud cake can cause much deeper invasion and much severe damage. The invasion of mud particles is retained to the first few inches but filtrate invades the whole core.or Experimental setup consisted of Hastelloy C core holder with a dual port end to circulate the mud across the face of core (diameter of 1.5 inch). Experiments were conducted at 60°C temperature, 2000 psia confining pressure, 300 psia overbalance and 1000 psi pore pressure. Authors also determined the cake displacement pressure CDP, which is threshold pressure, required to remove the cake and initiate the flow. Lower CDP favors horizontal well due to low draw down. The regain permeability is inversely proportional to both CDP and formation permeability. Yan, Jiang, Fan and Su (1996)*"), reported a case study of formation damage and damage control for horizontal wells drilled in the Shengli and Dagang sandstone formations. They observed that formation damage is an important factor influencing productivity among other factors such as reservoir thickness, horizontal well length, anisotropy ratio, etc. They also found that the effect of formation damage on productivity rate is constantly much higher for a horizontal well than for a vertical well. They concluded that an ideal drilling mud for damage control should be compatible with the rocks and fluids in the reservoir, have a fine quality of filter cake, low fluid loss, properly sized bridging particle according to the average pore size of formation, good rheology properties and acceptable cost.68 Bennion, Thomas & Biety (1996)" parameters which should be evaluated to design an effective drilling program and presented a special core analytical technique used to optimize drilling fluid and drilling process design. They concluded that shallow damage is more significant in open hole horizontal wells because fluid has to flow through the zone of impaired permeability during production in comparison to the perforate cased hole vertical wells, Analytically, they also observed that severity of damage in horizontal wells is significantly increased as the ratio of vertical to horizontal permeability degrades and also to a lesser extent as formation thickness increases. They suggested under-balanced drilling as partial solution if properly executed They recommended that important reservoir parameters, drilling fluid system and special core analysis test should 52 considered and conducted prior to design a drilling program of an open hole horizontal well. Burnett & Hodge (1996), came up with a new laboratory test method to illustrate the effect of drilling solids contained in drill-in fluids on formation damage and well screen plugging in a openhols horizontal well.69 A 2-feet long radial flow horizontal wellbore model incorporating dynamic filter cake deposition on natural and synthetic core material was used to study the formation damage. Formation damage tests with drill solids were performed for both water-based and synthetic based drill-in fluids. The experimental setup consisted of a core holder that can accommodate 2-ft long and 4-in diameter core. The apparatus was enclosed in a fibreglass- heating mantle and could be heated up to 350°F temperature. Fluids were circulated at the pressure of 1500 psi, through 0.5-inch diameter tubing passing through the drilling out core Authors used return permeability as @ criterion for formation damage. They concluded that formation damage occurs due to surface damage at the formation face, not by particulate invasion because of filter cake. Filter cake containing drill solids causes productivity impairment and plugging of sand control screens. The concentration and reactivity of drill solids highly influence the wellbore damage. Ghofrani, Zhang and Bosch (1996)"” , defined a new damage ratio to evaluate the core damage using different drilling fluids. it was demonstrated that new defined damage ratio is an expedient value for rock with original70 permeabilities, which allows to determine 2 reliable extent of the damaged zone of the sample core. In the experimental study, they found out that the extent of the damaged zone depends upon the contact time between the drilling fluid & the core and the stability conditions of the drilling fluid used. Colloidally unstable drilling fluids result in higher permeability filter cake, causing more formation damage due to entrance of very fine solids in the formation after the built-up of filter cake or vice versa. Laboratory setup contained a dynamic loop for mud circulation across a 25-cm long core in a Hassler type core holder. The mud was circulated with a velocity of 0.6 m/s for 0.5 - 5 hrs under 90°C temperature, 3.5 Mpa (500 psia) overbalanced pressure and 5Mpa (725 psi) confining pressure. Bentheimer sand with 20-25% porosity and 1-10 um? (D) permeability, 30-20um-pore size and 1-% clay were used in the experiments. Three types of water based mud were used with different composition of polymers and salts. Hatcher, Chen, S.Rehman & Hogg (1996)*, reported the results of a series of modified aqueous core floods conducted to investigate the potential damaging fluid/rock and fluidfluid interaction after static period in a high clay content and dual porosity system of Glauconitic sandstone.1 During the floods, flow was periodically interrupted to allow equilibrium between the fluids in the macroscopic and microscope systems. Flow rate sensitivity, pH & Petrographic characteristic of core and effluent were studied They concluded that a long saturation period should be given to invaded fluid and rock/fluid of core to allow for full equilibration before conducting the corefioods. No significant permeability impairment arising from delayed fluid/rock and fluid fluid interaction using simple brine. The principle damage mechanism is pore plugging by both expandable and non-expandable Glauconitic clay fines dislodged by the swelling of Glauconitic -smacitite interlayer. Leerlooijier, Kuijvenhoven & Francis (1996) , emphasized the development of a truly solids free drilling “uid system based on high density brines to control filtration. The polymer systems have been identified, those permit acceptable fluid loss without solid particles addition. The micelles formed, have the dual functionality of solids to plug the pores and polymers to reduce cake permeability. A multi-core dynamic fitration equipment has been designed to study static & dynamic filtration (including spurt loss condition) and assess their effect on retum permeability up to 4 core samples at a time. Continuos mud thickness could be measured by modification of the equipment.72 ‘They observed that the major permeability reduction is occurred in the first ‘5mm of the core, which can not be removed by the reversal of the flow. PH of Xanthan solution has a pronounced effect on fluid loss as fluid loss increases with decrease in pH values. Yan, Jiang & Wu (1997) ®, provided a model for the quantitative evaluation of the formation damage caused by drilling and completion fluid in horizontal wells. They found that when anisotropy ratio B < 3.8, the effect of formation damage on the flow efficiencies of horizontal well is less detrimental but if it is large when the anisotropy ratio B > 3.5. They reported from the calculation that even though the flow efficiency of horizonta! well is greater than that of vertical well, the formation damage can cause more reduction in oil production in horizontal well than in a vertical well. The value of anisotropy ratio is very important for a horizontal well because the productivity depends considerably on the vertical permeability of formation, In a horizontal well, the skin factors at each direction and each point of the horizontal section are not the same because of permeability anisotropy and different contact time.73 The fiow efficiency for a horizontal weil is greater than that for a vertical well at a given skin factor when the anisotropy ratio is relatively small (B=1.0-1.4). It becomes close to that of vertical well with @ reduction in vertical permeability (8 increased to 3.5) and it will be even less than that for a vertical well, if the vertical permeability further reduces (B equal to 10.0) The results indicated that the production rate of horizontal is greater than vertical well imespective to anisotropy ratio and decreases gradually with increase in skin. However the production rate of a horizontal well, reduces faster than that of a vertical well, especially for the formation having large anisotropy ratio due to large drainage area of horizontal well. Evaluation results show the formation damage is dictated by skin factor. The effect of skin factor is more significant when the horizontal well length is short and the anisotropy ratio of permeability is large. In contrast the effect is relatively small with a long horizontal well length and a small anisotropy ratio of permeability. Saleh, Rustam, El-Rabaa & Islam (1997)"!, came up with an innovative and unique horizontal wellbore model, designed to simulate realistic radial flow conditions of horizontal wells to evaluate the formation damage caused by drilling fluids.7 The main purpose of their study was to address, (i) Flow profile assessment in a open-hole completed horizontal wellbore before and after mud damage (li) Pressure damage characteristic (skin factor, skin depth and shape of skin damage region). (il) Pressure draw down requirements for open hole well clean up from mud cake and unplugging of fines. Experimental setup contained a core holder that could accommodate 4-ft long and 3-inch diameter cylinder core. Eight outlets with on/off valves were provided along the core holder to inject and produce fluids. The core was subjected to a confining pressure of 400 psi. Bentonite water based mud was circulated for 6-8 hrs with an annular velocity of 1.5 m/s and 50 psi overbalanced pressure. Mud filtrates were collected as a function of time. They concluded that (i) longer mud circulation creates greater damage in Berea cores. (ii) More damage is occurred in area of maximum overbalanced pressure. (iii) Draw down pressure required for initiating well cleanup is in the range of 5-20 psi. (iv) Low permeability cores exhibit higher filtration losses than those of high permeability.75 Sarbar, Lynn & Algam (1997)*" , carried out a lab study to assess the potential formation damage due to Jilh-water injection into the Unayzah formation. The study showed that the injection of Jilh-water result in rapid loss of injectivity and core-plug damage. Tests were conducted in a flow loop apparatus under overburden pressure of 4500 psi, pore pressure of 1500 psi, temperature of 67°C, and injection pressure of 0.2-5 mi/min. They concluded that the permeability damage is caused by the mechanical mobilization of fines due to high-velocity flow and chemical damage from clay swelling induced by low salinity brine. Jack . D. Lynn (1998) "®!, conducted a laboratory study to evaluate the relative damaging performance of five XC-Polymer based drilling fluids in high permeability Hawtah formation. The fluid formulations were based on minimum 40% KCI brine and also contained sized metamorphic carbonate (marble) for filter cake bridging. The study was carried out in a closed loop dynamic apparatus to determine the regain permeability. Experiments were conducted for 30 minutes76 at 180°F temperature, confining stress of 1000-psig and pore pressure of 400 psig. He concluded that driling fluid with high percentage of XC-Polymer and sized metamorphic carbonate gives the best retum permeability on Hawtah formation cores. Addition of 10 ppb of simulated drill solids improves the performance of the drilling fluids. Longeron, Alfenore & Poux Guillaume (1998), in a three year joint industry project, studied the impact of various parameters such as initial core permeability, fluid saturation (oil and gas), temperature and shear rate on formation damage during overbalanced drilling and completion operation. The aims of project were (i) to establish a methodology for evaluating filtration properties and permeability damage obtained from different drilling mud formulations under representative bore hole conditions. (ii) To quantify the impact of various parameters on return permeability after mud invasion (ii) to assess the additional damage due to the use of the heavy brine as completion fluid after drilling operation. Experimental runs on 40-cm long and 5-cm diameter core were conducted with shear rates of 50 & 200s" and temperature of 22 & 80°C in a dynamic flow7 equipment and dynamic filter press respectively. Eight typical drilling fluid formulations, including water based and invert synthetic oll based muds were used to perform static and dynamic filtration tests on outcrop sandstone core samples. The performance of each mud was evaluated on the basis of fluid losses during mud invasion and return permeability measurement after mud ‘exposure. Results showed that the water-based muds impair the formulation permeability significantly than the invert oil based mud due to invasion of filtrate generated from water based muds. It was also observed that oil based muds attribute a favorable relative permeability effect due to a reduction of connate water saturation during mud filtrate invasion resulting in greater oil return permeability than initial oil permeability. He concluded that water based mud give higher “spurt ‘ losses and filtration rate than the OBM and POBM for a given time of mud exposure. Dynamic filtration rates are always greater than static ones. Azizi, Chen & S. Rahman (1998), carried out an experimental study to manage the wellbore instability and formation damage by improved drilling mud. The objective of this study was to experimentally investigate the mud caking property of different drilling mud systems in a low permeability gas formation.78 ‘They suggested that it is ideal to have a drilling mud which can quickly form a thin, tight and low permeable filter cake on the well bore face to prevent any further mud fitration invasion into the formation. The formation of tight cake will resist pore pressure to build and reduce the chances of formation collapse. Experimental setup consisted of a humidified air permeameter to saturate the core to brine and determine initial permeability to air. A dynamic filtration apparatus was used to circulate the mud across the face of core (4 in long) at reservoir condition. The mud was circulated at a velocity of 0.5 m/s under overbalanced pressure of 100 psi and temperature of 60°C for 4 hrs. The cores were also in contact for 10 hours under static condition. Experiments were conducted by using four different muds (i) Glycol mud (ji) ester mud (ji) high temperature clay inhibitive HTC! (iv) Modified version of HTCI. They concluded that the ester mud has the ability to form the tightest internal mud cake and produces low dynamic fittration volume without spurt loss.CHAPTER 5 OBJECTIVES & APPROACH OF THE STUDY Well productivity is critically important if the 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 to open hole completions places additional emphasis on damage avoidance. Near wellbore permeability impairment from drilling and completing fluids can have a substantial, yet potentially avoidable, impact on well productivity. Horizontal well advantages such as productivity increase, recovery improvement and less gas and water conning problems could be jeopardized by formation damage due to mud invasion during drilling, completion and/or workover operations. 7980 The literature survey presents the extensive work done in evaluating and eliminating the permeability impairment due to drilling fluids. Many important parameters such as mud compositions, core permeability, overbalanced pressure, pore throat size distribution etc were studied by running leak off tests for short period of times. But no experimental work has been done which can really compare the effects of different overbalanced pressures and long mud contamination times on the formation damage i.e. regain permeability, mud invasion depth and skin factor. Furthermore, there is no reliable non-destructive method for determination of invaded mud depth. Conventionally, depth of mud invasion is determined by measuring the permeability at different section of mud flooded core plugs or by Scanning-Electron Microscopy (SEM). Hence the objective of the present work is to better understand the influences of overbalanced pressures and contamination times on the formation impairment during drilling operation with an innovative ultrasonic mud invasion depth measurement method. To achieve these objectives, a leak-off apparatus is constructed to simulate the mud circulation process across the formation face during drilling operation, Berea sandstone core samples are subjected to different overbalanced pressures for various mud contamination times in the leak-off apparatus under actual reservoir condition. Regain permeability & Skin factorsat are calculated through experimental data obtained by leak-off apparatus and mud invasion depths are determined by a new ultrasonic method. Data collected from these experiments provide the basis for determining the optimum overbalanced pressure in term of regain permeabilities, skin factors and mud invasion depths.CHAPTER 6 EXPERIMENTAL SET-UP & PROCEDURE Leak-off experimental setup is designed to simulate the drilling fluid circulation process at the formation face in the welbore under bottom hole conditions. 6.1 Apparatus 6.1.1 Core Holder A Hassler type core holder (figure 6.1) is used in the experiment. This Stainless Steel core holder can accommodate 1-ft long and 2-inch diameter core. The core is mounted inside the rubber sleeve and subjected to overburden (confining) pressure. One end piece of the core holder is fabricated to have 8283 JOPjOH B09 adfy sajssey :1"9 einBiy peay Mol UL pnW, ano PAW. J — exeo PHA Bupy saoedg, Janoars seqana, Pare, wag pu84 two ports and referred as ‘Injection end’. These ports are used to circulate the drilling fluids across the face of the core and also to inject oil & brine as well. Other end piece, which is known as ‘Production end’, has only one port to collect the filtrate/oil/brine, pumped from the Injection end. 61.2 Spacer A stainless steel spacer of 10-mm in length is placed between the core face and Injection end to allow the mud to circulate and form cake on the core face. 6.1.3 Oven The whole setup except pump and backpressure regulators is confined in an electric oven to operate at reservoir temperatures up to 300°C. 6.1.4 Transfer Cell ‘Transfer cells are employed to deliver mud/oil/brine to the core holder at a specific flow rate and pressure, which is dictated by a piston pump, connected to them. 6.1.5. Piston Pump ‘A Beckmann piston pump is connected to the transfer cell to deliver the fluids (mud/oil/brine) at desired flow rate and pressure. A mixture of kerosene85 and silicon oil is used as transferring fluid in the pump. The pump can deliver the fluid at a maximum flow rate of 10 ec/min and pressure of 10000 psi. 6.1.6 Back Pressure Regulator (BPR) Back Pressure Regulators (BPR), at each end, are installed to control and maintain the desired pressure in the system by mean of pressurized nitrogen gas. Regulator at injection end is used to maintain the overbalanced pressure across the core while other BPR retains pore pressure inside the core. 6.1.7. Differential Pressure Transducer and Chart Recorder Validyne DP303 differential pressure transducer is mounted over the core holder for measurement of differential pressures across the core. These differential pressures are displayed on analogical digitor. The electrical signals sent by the transducer are also transferred to a chart recorder where pressure difference across the core is recorded. 6.1.8 Fraction Collector An ISCO fraction collector is used to collect the producing fluids in 10cc test tubes. The Fraction collector can be set to move according to time. In this way the volume collected in a certain time can be recorded,86 6.1.9 Miscellaneous Items Tubings, fittings, shut-off valves, pressure gauges, Ostwald’s Viscometer, hydrometers, stopwatch, flask, beaker, test tubes, mud collector, pressure multiplier. ‘Schematic of Leak-off apparatus is shown in Figure 6.2. 6.2 Pre-Experimental Procedures 6.2.1 Fluids 6.2.1.1 Oi! Arabian medium crude oil of 33.5° API is used during the experiments. Paraffinic compounds are removed by mean of an oil filter of 1 micron before using it. 6.2.1.2 Brine 3.5% (35000 ppm) KCI brine is prepared to saturate cores. This brine is selected to avoid chances of clay swelling and clay migration and to keep the salt concentration above Critical Salt Concentration (> 20000 ppm).a7 snyeuedde jjo-yeo7 s1weudg Jo syeUIeYydg :7"9 eunB}4 TO ung vais yuer en Joysoli09 evenly yeinBow, ‘aunssete yoeg el iol {T+ seonpsues, lenuoueyic ounssera Past -38 Preparation of Brine: 35 gm KCl is dissolved in one liter of distill water in a mixer for 1-2 hours. Brine, then filtered by no. 50 filter paper and evacuated in a desiccator to remove air bubbles for 1-2 hours. 6.2.1.3. Drilling fluids Water based polymeric drilling fluid is used during the experiments. The composition of the drilling fluid is given in table 6.1. Preparation of Mud: Mud is prepared in a tank mounted with a mixer. The ingredients are mixed for a specified time period in recommended order. Table 6.2 shows the sequence of ingredient and mixing time. 6.2.1.3.1 Determination of Drilling Fluid Properties To determine the characteristic of the Drilling fluid, following properties are determined. Fann Viscometer is mainly empioyed to determine the viscosity at different shear rates. + Apparent Viscosity — S500 May29 4 Plastic Viscosity Hp = Feo P00 © Yield Point YP = Osq9 — Hp ¢ ~— PowerLaw Boo ~ 8, n=3.32log, ( 600 :) 7 309 ~ 95 ¢ — pHof Drilling fluids pH of drilling fluid is determined by pH-meter. ¢ Fluid Loss the fluid loss properties of the Fluid loss apparatus is used to mea drilling fluid. Fluid is filtered through a no. 50 filter paper at 100-psi pressure differential for 30 minutes. The collected filtrate volume show the fluid loss characteristic of mud in a specified time. The drilling fluid properties are shown in Table 6.3Table 6.1: Water Based Polymeric Drilling Fluid Composition Drilling Fluid Water, bbi T0894 KCI, ppb 40.0 XC-Polymer, ppb _ —] 1S Disprac, BbT 05 | Dextrid, ppb ] 6.0 KOH, ppb 1 03 ‘CaCOs (fine), ppb | 50Table 6.2: Mixing times for Drilling Fluid Components. of Mixing Time, min Composition per Liter | Water, cc ~ aaa | KCI, gm 5 133 XC-Polymer, gm — (Xanthan Gum) 15 5.0 (Thickener & suspending agent) é Disprac, gm a (Filtration control agent) 15 17 Dexirid,gm itration contro! agent) i 20 KOH, gm 5 TO CaCOs (fine), gm 15 16.7Table 6.3: Water Based Polymeric Drilling Fluid Properties | DrillingFluid Density, ppg 94 Fluid Loss, per 30 67 PH O77 = ‘Apparent Viscosity, cp 315 Plastic Viscosity, cp 192.0 Yield Point, op 25 Gel Strength (10 sec) Ib/100sqft 55 Gel Strength (10 min)b/00sqt [| = OO Power Law Index, n 0.577 9293 6.2.2 Core preparation Berea sandstone cores are reduced to required size. Their dry weights, dimensions, gas porosities and gas permeabilities are determined. Then cores are evacuated for 8 hours in a saturator before soaked with prepared brine for 12 hours under 2500 psi pressure as per API recommended procedure), 6.2.2.1 Determination of Core Properties Following properties of Berea sandstone are determined on each sample before running the Leak-off test. + Gas Porosity: Helium gas porosimeter is used to determine the gas porosity of Berea sandstone core. + Liquid Porosity: Liquid Porosity of core is determined by saturation method. Dry and wet core weights are used. . Gas permeability is measured by passing gas (air) at constant differential pressure and determining the flow rate across the core.The following measurements have been performed on one sample as a preliminary test measurement. + Pore Size Distribution: Centrifugal method is employed to determine the pore size distribution. Brine collected during centrifugation and speed of centrifuge are used to determine the Capillary pressure. Pore size distribution is measured by mean of Capillary pressure and surface tension. Pore size distribution is presented in figure 6.3. Procedure for pore size measurement is discussed in appendix-A. ¢ Particle Size Distribution: Image analyzer is used to determine the particle size distribution in the drilling fluid. Distribution and particle image are shown in figure 6.4a&b. Scanning Electron Microscopy are also employed to demonstrate the size and shape of the bridging particies, which is illustrated in figure 6.4c.95 e109 ojduies valeg 105 uoynqisig ezIg a1od :¢°9 enBiy uos9]ur ‘9215 840¢ a JOA 810d $0 %sasfjeuy afew Aq sejoqieg BuiBpiig yo uopnquysiq ezis :ep'9 einBiy os9}uI ‘ezis ejoed bree wee ekUz dee 3 = ee ee i eBejussied9798 Figure 6.40: Particle Size Distribution of Bridging Particles by Scanning iron Microscopy (SEM).99 6.3 Experimental Procedure Experimental procedure for leak-off test consists of following steps 6.3.1. Absolute Permeability Brine is used to determine the absolute permeability of core under reservoir conditions. Fully brine saturated Berea cores are mounted inside the core holder and flooded with brine at a constant flow rate until steady state is achieved. Differential pressures at different fiow rates are measured by mean of pressure transducer. Absolute permeability is determined by using differential pressures and flow rates through regression analysis and Darcy's equation 6.3.2 Sw & Effective Permeability After flooding with brine, filtered crude oil is passed through the core at very slow rate to displaced brine in order to get irreducible water saturation Sw; Initially 5 pore volumes of displaced oil and brine is collected in a graduated flask to get Sw by phase separation overnight. Oil permeability is also measured by fiowing crude oil through the core and measuring the flow rates and differential pressures at steady state condition. This effective oil permeability is considered as initial core permeability before damage or mud circulation.100 6.3.3 Circulation of Mud Drilling fluid is circulated across the oil and water saturated core at @ constant flow rate of 1.0 cc/min at different overbalanced pressures, maintained by Back Pressure Regulator installed at drilling fluid outlet. Every time fresh mud is provided to core through the transfer cell end used mud is collected in 2 mud tank. As the circulation starts, drilling fluid begins to form mud cake on the face of the core and filtrate also commences invading the core. The invaded filtrate pushes the oil and effluent is collected from production end in a fraction collector at desired time intervals. Filtrate has to pass through a 1000-psi backpressure regulator before accumulating in the fraction collector. Drilling fluid is circulated for different time periods and at different overbalanced pressures. Now the core is considered as damaged with invaded filtrates and mud particles. 6.3.4 Return Oil Permeability Oil permeability of damaged core is determined by flowing oil in opposite direction at constant flow rate until steady state is achieved. This oil permeability is referred as Return Oil Permeability after the damage.101 By comparing before and after damage permeabilities, one can conclude about the compatibility and performance of drilling fuid to the formation rock. Measured values are presented in Appendix-B. 6.3.5 Apparatus Cleaning Process All lines and openings are cleaned with Carbon tetrachloride CCl, before use for next experiment.CHAPTER 7 ULTRASONIC INSPECTION AND ANALYSIS OF THE DEPTH OF MUD INVASION 7.1 Introduction Ultrasonic testing has established itself as non-destructive method for probing the surfaces and interior of solids, both to locate defects and measure material properties. Some useful parameters in the context of ultrasonic probing include the ultrasonic wave velocity through the specimen, attenuation of ultrasonic wave (decrease in the wave amplitude) and frequency spectrum of the wave transmitted through the specimen. More specifically, any variation in the velocity of the wave passing different sections of the specimen could be related to the inhomogeneties, anisotropy or material defect in the specimen. Similar considerations apply to the variation in the amplitude of the wave passing through different sections of the specimen. Both these parameters are easy to 102103 measure experimentally. Analysis of frequency spectrum on the other hand, although equally interesting, is considerably more complicated because of Fourier transformations involved there. We used the velocity variation of ultrasonic waves in the analysis of mud invasion in the Berea rock sample. Conventional transducers were used as source as well as detectors of ultrasound. 7.2 Objective & Approach of the Study The main objective of this new method is to map out the depth of mud invasion in rock samples from oil wells. Ultrasonic techniques are used for measuring the above mentioned depth and thickness of mud invaded rock samples. This involves the determination of sound velocity through different sections of mud flooded rock sample where densities are different in the sense that pores in the rock are filled with mud. In this study, compressional wave velocities in Berea rock samples initially saturated with oil (SO) and irreducible water (Swin) are measured, and subsequently after the mud invasion. We consider cylindrical rock sample taken from Berea formation. The difference in compressional velocities between these conditions should assist in determining the mud invasion depth along the length of the rock.7.3 Ultrasonics Ultrasonics covers the study of pressure waves that are of the same nature as sound wave but that have frequencies above the audible limit (typically > 20kHz). In solids there are two main modes of wave propagation, viz longitudinal and shear, which depend on the directions of particle movement. 7.3.1 Longitudinal / Compressional Waves In these waves, the vibrations or displacements of the particles of the medium (rock sample in our case) take place in the direction of the propagation of the waves. In other words, the particles of the medium move backwards and forwards along a line in which the wave is travelling. In fact, the individual particles of the medium oscillate about their equilibrium position without there being any net movement of the medium as a whole. The longitudinal wave relevant to the study of ultrasonic is sometime called compressional waves, because the imaginary layers in the materials of propagation are subjected to alternate compressional stresses by the waves. 7.3.2. Shear/ Transverse Waves Another form of propagation is that of transverse waves, for which the vibrations of the particles in the material occur at right angles to the direction of the wave motion. The imaginary layers are subjected to alternating shear105 stresses and the resulting strains are directed parallel to the surfaces. Shear waves can generally be passed through solids only because liquids and gases don't usually support shear stresses. Both the compressive and the shear velocity are found to vary from one rock type to another, because the velocity of wave propagation is likely to be influenced by the constituent minerals and their grain size, composition, pore size distribution etc. The compressive wave velocity is usually computed from the ratio between wave path length (equal to diameter in our case) of specimen and the mean transit time. 7.4 Influence Of Petro-physical Characteristics On The Compressional Wave Velocity. 7.4.1. Control of Mineralogy on Velocity The compressive velocity values are found to vary from one rock type to another probably because of variation in mineralogy and grain size. Since different minerals exhibit different wave velocity ™, the velocity of wave propagation in a rock is directly influenced by the respective velocities of the constituent minerals. Birch © contended that the average velocity of a rock material could be evaluated from modal percentages and various mineral velocities provided all other parameters are assumed constant. A rock material predominantly106 composed of minerals that exhibit high wave velocities, for example, homblende and quartz, would accordingly display a reasonably high velocity while abundance of low velocity minerals would be reflected through an overall low rock velocity. Tsidzi ! investigated the effect of mineral content on the rock velocity in a variety of samples. He found a general increase in mean dry velocity with increasing quartz content whereas decrease in velocity with increasing mica content. Ramana and Venkatanarayana ©7, having studied schistise rocks of predominately hornblende and quartz composition, reported an increase in velocity with increasing homblende content and decrease in velocity with increasing quartz content. The latter observation, however, appears to contradict the study conducted by Tsidzi ©. But this may not necessarily be the case as explained below. For 2 property such as average wave velocity through the medium that often is characterized by a wide array of competing factors, it is possible to have a situation wherein the effect of one factor may totally overshadow that of another. The rocks studied by Ramana and Venkatanarayana ®7) had porosities varying between 2% and 12% that were large enough to subdue the effect of a parameter such as mineral content. On the other hand, in the study of Tsidzi I, porosities were generally low and within relatively narrow range (0.2-1.5%). However, this was associated with wide variation in their degree of foliation107 (weakly fo highly foliated) which significantly affect the mean velocity values and could possibly suppress any infiuence of mineral content. So, mineralogy may not necessarily have sensitive control of wave velocity in rocks especially if large variations in numerous other rock characteristics also occur. If other characteristics such as pore sizes do not significantly vary, the general assertion is valid that preponderance of low velocity minerals in a particular rock would impart to it a reasonably low velocity while a rock composed of high velocity minerals would have an overall high velocity, especially when the rock is relatively fresh and non-porous "), But it has to be remembered that compressional wave velocity also depends upon other petro-physical parameters such as grain size, porosity and density in addition to mineralogy 7.4.2. Control of Grain size & Microstructure on Velocity. With regard to grain size, fine-grained rocks e.g. Slates, phyllites, and homfels, tend to exhibit higher source wave velocities than the medium grained rocks (quartzites, schists and gneisses) while the relative coarseness of the gneisses as compared to the schists is reflected in their lower velocity values. This dependence of wave velocity on mineral grain size can be related to the influence of microcracks on the velocities of elastic wave in the rock material 871, Taking the length of a microcrack to be approximately equal to the size of a108 mineral grain © one would expect a reduction in velocity of wave propagation through a rock material consisting of large mineral grains because longer microcracks would tend to retard the waves to a greater extent than the shorter ones, provided all the other parameters including microcrack width remain constant. This intuitive explanation hes been experimentally confirmed by Anderson ®l, who indeed, observed that rock materials with abundantly high aspect ratio (width-to-length ratio) microcracks exhibited higher wave velocities than those with low aspect ratio microcracks. The actual size of the source of ultrasonic waves, the receiver or detector, and the phase front (cross sectional area through which the wave is propagating) also comes into consideration over here. In general, this phase front is expected to be much bigger than the size of the microcracks or pores. Thus parts of the wave can pass through the surrounding region and the wave reaching the detector is actually an averaged transmission characteristic. 7.4.3 Influence of Density on Velocity There is a linear functional relation between the dry compressive velocity and mean bulk density as reported by Tsidzi "=, despite the wide scatter of results reflected in the low coefficient of correlation (r=0.47) shown in figure- 7.4a. Such a poor relation between velocity and density has also been reported by Ramana and Venkatanarayana ©" for schistose rocks, thus suggesting that irrespective of the functional relation that may have been established between6000 Figure 7.1a : Plot of mean velocity versus mean bulk density for dry samples. 109110 sonic velocity and density of rocks, rock density could not be reliably employed in elucidating variations in velocity, Contribution of other rock properties such as porosity and anisotropy fabric would have to be considered at the same time. 7.4.4 Control of Porosity on Velocity. A parameter of no less importance influencing the propagation characteristic of elastic waves through a rock material is porosity. Wave velocity generally decreases with increase in porosities and variation strongly obeying the inverse power law. 7,45. Directional Variation of Velocity. Effects of anisotropic fabric on wave velocity in rocks have seriously engaged the attention of several research workers in the last three decades or so wherein velocity anisotropy has been ascribed to preferred orientation of mineral grains and/or micro-cavities in rock materials or to discontinuities in rock masses. For rock materials that do not show obvious intrinsic anisotropies, oriented microcavities, especially microcracks, tend to exercise a much greater control on the velocity anisotropy than mineral grains. Maximum velocities occur when specimens are oriented parallel to foliation whereas minimum values occur in perpendicularly oriented specimens but111 occasionally in specimen with oblique foliation orientation. These velocity changes may be attributed to the variational effects of wave energy loss due mainly to wave reflection, the least energy loss occurring in B=0° specimen (where B is orientation angle). In addition, the amount of wave retardation, hence decreases in velocity, bears a direct functional relation with the number of critically oriented reflecting surfaces. 7.4.6. Effect of Water Saturation on Velocity. A significant change in wave velocity occurs when the rock material is saturated with water, the change being controlled largely by porosity and grain size. Velocity measurements are quite different for "dry" or 100% saturated samples. Figure 7.1b gives an example of velocity measurements versus pressure for dry, partially and fully saturated Massillon sandstone. For P waves at partial saturation, the velocity is lower than the “dry” velocity, which is lower than the velocity at 100% saturation. The qualitative explanation is simple: for P waves and for spherical pores, the only effect due to the introduction of water into the dry sample is a density increase leading to decrease in velocity. By contrast, when complete saturation is reached, the pores become more difficult to compress, increasing the velocity. For granite, the experiments of Nur and Murphy "I (figure 7.2) show that the "dry' velocity is lower than the velocity at partial saturation, which is lower112 than the velocity at 100% saturation. In this case, the porosity is too low for the density effect to be greater than the matrix stiffening effect (figure 7.3). In the foliated rocks "*, while the velocity change decreases in fine- grained rocks, it tends to increase in coarse-grained rocks. Conversely, Water saturation tends to increase velocity anisotropy in fine-grained rocks while decreasing it in coarse-grained varieties. Water saturation also enhances the correlation between the degree of foliation and velocity anisotropy. The type of saturating fluid also influences the acoustic wave velocity, if only because of the compressibility of the fluid. The variation in velocity is given in the figure 7.4. Furthermore, as clearly observable in this figure, the fluid viscosity (and hence its temperature) influences the velocities, even more than the specific type of saturating fluid.ea \ eect st Dey Fully and partially st ° 10 20 30 Effective pressure (MPa) Figure 7.1b : Velocity/water saturated relationship in Massillon sandstone vs, confining pressure (resonant bar)(after Nur et al., 1980). 11314 ‘Sieera White granite Positioners | Vg (1762 -2088 Had Ts 13kHe Vg (1077 1157 Hat ee ra 2 Water stueation (8) °o 2% 4 6 8 100 eo Wate saturation (8) a Extensions! Velocity (km/s) 2 2 Mason sndstone Masson sndetone 18 18 ca orn = fia ‘vg 1599-997 Hel 13 vg 1573-647 He) ome 0—onooocoone} Eg / ’ \ vg 085-659 Wg (85-985 He 08 | os L . fees 2 | . 2 sos #8 wo 1 2 30 4 0 60 70 60 90 100 Woter saturation (%) loa sawetion (%) i Figure 7.2 : Velocity/water saturated relationship for different materials (resonant bar). Curve (b) after Bourbie and Zinszner, 1985. Curve (a) and (c) after Murphy, 1982.Figure 7.3 Air saturation 32) 100 $0 ° —S—— pe = 35 MPa Compressional velocity Woter saturation (%) “Characteristic’ behavior of velocity as a function of water saturation for consolidated sediments and a confining pressure of 35 Mpa (after Gregory, 1976). 15118 somo | | 100% ot 7] peor qa geese || oe g oe be ayn dues : pyres oe © 25% © 135°C i #15090 3 40 500 «1 620 020 a9 80 Diffecencia’orezsuce (MPs) Figure 7.4 : Influence of temperature on P-wave velocity in Venezuelan sands for different saturating fluids (ultrasonic measurements) after Tosaya et al., 1985)"7 7.4.7 Effect of Temperature on Velocity. Velocity varies slightly with temperature. For example, the velocity in distilled water increases by 7% as temperature rises from 10 to 100°C, (Kaye and Laby "")), and the velocity in quartz decreases by 0.4% as the temperature varies by 100°C (Carmichael 7), This shows that temperature exerts a very slight intrinsic effect on velocity. For low porosity rocks such as granite and gabbro, the decrease in velocity with temperature is nearly always less than 5% for a temperature increase about 100°C. For dry and water-saturated sandstones, velocity decreases with rising temperature. Velocity variation observed in sandstones may be explained by the different thermal expansion constants between the different constituents of the solid, which include clays and quartz. 7.4.8. Effect of Saturating Fiuid Viscosity on Velocity. Investigations on velocity variations as a function of saturating fluid viscosity have been conducted by heating the sample using different fluids. Velocity variations in solids as a function of temperature are negligible and since the variation in velocity of the fluid as @ function of temperature is small in comparison with that of the viscosity, raising the temperature is equivalent to observing velocity variation with viscosity. For glycerol, for example, from -77 to +100°C, the viscosity decreases by ten orders of magnitude, while its118 compressibility only varies by a factor of 3 =". in granite, limestone and in the sandstone samples, a rise in velocity is observed with increasing viscosity for saturating fluids such as glycerol and oil. in the case of water, the velocity is independent of temperature to within measurement accuracy, and the viscosity of water is virtually invariable in the interval mentioned above (in comparison with that of oil). The higher the saturation of the medium with viscous liquid, the more important the viscosity effects. 7.5 Experimental Setup & Procedure The experimental setup contains a Panametric Pulser-Receiver (Model 5072 PR), connected to a 500 MHz Oscilioscope (Hewlett Packard) through a Panametric Pre-Amplifier. Two Panametric Transducers model V403 (2.25 / 0.5 x 4.0 inch) are used to inject and receive ultrasonic waves generated by the Pulser-Receiver. In order to get a good contact between transducers and rock surface, refined petroleum jelly is used as couplant. The Oscilloscope is interfaced with a PC to transfer the data (records of the generated waveforms) for records and subsequent analysis. Fig 7.5 gives a schematic diagram. Basically, ultrasound pulses generated by coaxially positioned Pulser - Receiver are passed through the specimen and the time taken for the ultrasonic pulses to travel from the transmitter to the receiver is read out on the oscilloscope.119 Cylinder shaped rock samples of 5.08 om diameter but having different lengths, were investigated by sending ultresonic waves, along the diameter at different positions scanning the entire length with a 0.5-cm resolution (Fig 7.6). The transit times of the waves travelling from the source to the receiver along the diameter were recorded on the Oscilloscope and stored in the computer. Signal averaging was used to eliminate spurious and random noise. For each data point on the ultrasonic waveform, 64 samples were averaged. Velocities of ultrasonic waves were determined simply by calculating the ratio of the specimen thickness (diameter) and the transit time. A total of 11 samples of varied lengths were investigated. The lengths of cores were selected from 4 to 10 inches to investigate the effects of mud circulation time and overbalanced pressure. The velocities of longitudinal wave were measured at upto 47 different positions along the length of the rock, depending on the length. The velocities along the length of the rock were plotted in simple graphs to obtain unambiguous information about the mud-invaded portion of the sample.120 =| Eb ss ; ——1 — | ee | ee oo! 3 4H e s io Receiver Transmitter ‘Transducer Transducer Figure 7.5 : Schematic of Ultrasonic mud depth measurement method.121 Rock Sample eh Transmitter —_ lc hk Figure 7.6: Scanning position of122 For each core, the compressional wave velocities were determined under three different conditions as explained below, 1 Dry Sample The sample is completely dry and pores are presumably filled with air. This gives the reference velocities under dry conditions. Oil Saturated Sample Velocities at this stage are considered as baseline velocity. The dry samples were completely saturated with brine and then displaced by crude oil to get irreducible water Sw, and initial oil in place Soi (as desoribed in chapter 6). Oil saturated sample contained only Sw; and Soi. Mud Damaged Sample At this stage, core would have been damaged with drilling fluid while circulating across the face of core for different overbalanced pressure and contamination time in leak-off apparatus.123 7.6 IDEAL DELINEATION OF VELOCITY DISTRIBUTION Figure 7.7a presents an ideal delineation of velocity distribution for the above mentioned three stages (Dry, Oil & Swi Saturated, and “Mud Invaded” samples), used to determine the ‘mud invasion’ depth. The mud-laden region is shown on the right side of the figure and referred as ‘Dense End’ whereas the ‘Lean End’ with no mud appears on the left side. As the ‘mud’ starts to flow across the face of the core as shown in figures 7.7 b, solid particles and filtrate from the mud begin to invade into the core. After some time, one can expect a Belly shape fiow. Depending on the pressure difference and time, it is expected that a certain portion of the core will be fully saturated with mud solids & filtrate. Beyond that region there will be a portion with linearly decreasing concentration of mud solids and filtrate, and further along the length of the core a undamaged virgin portion may be expected. Line 1 in figure 7.7a presents the velocity distribution of dry sample along the core length. The mean dry velocity is supposed to be lower than oil saturated and damaged sample velocities. In this case, sound waves follow a zig-zag path (like snake) instead of direct path through the rock matrix in order to reach the other end as shown in figure 7.8a. Due to this zig-zag path, the actual path length of wave increases and it takes more time which ultimately appears as a decrease in velocity.126 Line 2 represents the velocity distribution of initial oil- and irreducible water- saturated sample. The velocity distribution under these conditions acts as baseline velocity. In general, the mean velocity, in this case, should be higher than the dry sample case because now sound waves can follow straight path (like arrow) due to the fact that pores are fill: with liquid as presented in figure 7.8b. The oil saturated core sample, discussed above, is then allowed to be invaded by solids (particulates) and filtrate of the driliing fluid that is circulated at the face of core. The velocity distribution along the length of the mud—invaded sample is shown by line 3 in the figure 7.7. Solids particles will plug the pore throats and adjust themselves there. Since solid particles exhibit higher wave velocity than liquid, so sound wave will travel faster. Thus, one can get higher mean velocity for mud damaged sample than the previous two cases. ‘The piateau at the beginning of the dense end reflects that in essence, the pores are completely filled with invaded solid particles. Beyond the point-A (7.7a), a decrease in the velocity indicates a decrease in particles and filtrate concentration along the length until the base velocity (oil saturated sample velocity) is reached at the point B, beyond this point the core is still virgin. So we can say that in the above case the mud has invaded upto a depth equal to the distance between the dense end and the point B.Velocity, m/s Dry Sample, Line 1 Lean End Dense End Distance, cm Figure 7.7a: Ideal Delineation of Velocity Distribution126 7 Concentration ! Mud Solids and eee filtrate Invaded Zone Figure 7.7b: Invasion of Mud Solids & Filtrate in the Cylindrical Core.127 Mathematically, the velocity profile for the above three stages at a particular point is given by; For Dry Sample: L[Vo=1,[V, For Oil & Swi Saturated Sample: L[Vos =L,[Vq+]y[V+L,/¥, For Mud Invaded Sample: L/Vu =L,/V, +L, /V,+L,1¥, Where, V5sVoss¥. = Average sound velocity through Dry, Oil & Swi saturated, and Mud invaded samples respectively. V,V.Ves¥.sV, = Sound velocity through Grain, Oil, Water, Air, and Invaded solids respectively. L,,L,,Ly,L,,L, = Part length occupied by the Grain, Oil, Water, Air, and invaded solids respectively in the direction of propagation of the sound wave direction. L_ = Total length of the sample.128 O Rock Matrix Pores with Soi & Swi © Mud Particles Fig 7.8 a: Dry Sample Sound wave follows zig-cag path, e 8 Fig 7.86 §| Oi Saturated Sample. > ‘Sound wave follows 5 straight path s » g Fig 7.8¢ =| Mud Invaded Sample. 3] Sound wave follows g straight path s 8 Figure 7.8: Velocity Distribution at Different StagesCHAPTER 8 EXPERIMENTAL RESULTS & DISCUSSION In this chapter the experimental results are presented, analyzed and discussed. These results are based on the laboratory work that was conducted to evaluate the effects of the following factors on formation damage. * Overbalanced Pressure © Contamination Time Parameters like regain permeability, mud invasion depth and skin were obtained for each of the above influencing factors. Comparisons of different overbalanced pressures and contamination times are made in order to evaluate the impact of these factors on the formation impairment by varying only one of the factors while keeping all other parameters constant. In particular, the 129130 temperature of the sample, confining pressure, pore pressure, differential pressure, mud flow rate and mud composition were kept constant during all ‘experiments. However, while investigating the influence of contamination time, overbalanced pressure was kept constant. Similarly, in order to study the influence of overbalanced pressure, the contamination time was kept fixed. ‘An innovative, non-destructive ultrasonic method is employed to determine the depth of mud invasion. Details of the method have already been discussed in chapter 7. ‘A summary of the experiments conducted with variable parameters is shown in Table 8-1. Table 8-2 presents the values of the parameters that were kept constant during all the experiments. We wiill first present the results of the ultrasonic mapping of the mud invasion.134 8.1 Results of Ultrasonic Method The depth of mud invasion in the sample core for different overbalanced pressures and contamination times was measured by means of an ultrasonic method. The basic idea is to get a velocity profile along the length of the core. Figures 8-1 to 8-11 present the velocity profile for each experiment conducted for different overbalanced pressures and contamination times. In these figures, velocity distribution at three stages; dry, oil and water saturated (at So; & Sw) and mud-laden sample are illustrated. The velocity profile for oil saturated sample acts as base line velocity distribution. As expected, the mud-laden portion of the sample core showed higher velocities. In general, these velocities decrease as we move along the core length. The portion beyond the point where this intersects the baseline velocity is considered as virgin zone. It is clear that all the figures 8-1 to 8-11 essentially follow the same patterns i.e first there is a plateau and then a linear decrease in the velocity. The plateau shows that this particular zone is completely filled with mud solids and filtrate while linear decrease in velocity is due to decrease in solid and filtrate concentration (incomplete invasion). The remaining portion of the sample has not been reached by the mud solids and is considered as the virgin zone.132 A summary of the mud invaded depths measured by the ultrasonic method for different core samples is given in Table 8-3. Complete data related to velocity distribution for all cases are presented in Appendix-C.133 Table 8.1: Summary of the Experiments conducted with variable parameters, 7 W12 1400 4 2 I-15 100 ae 3 J14 100 24 4 I-16 100 48 5 SAT 200 24 6 I-13 300 4 7 7 300 12 8 STA 300 24 9 JB 700 4 10 J9 700 12 1 S10 700) 24Table 8.2: Values of constant parameters. Temperature, °F 160 Confining Pressure, psi 2500 Pore Pressure, psi 7000 Differential Pressure, psi 7500 Mud Flow rate, co/min 10 Mud Compositions ‘constant135 ‘Table 8.3: Summary of the mud invaded depths measured by ultrasonic testing method. Mud Invasion Depth, cm @ 4 7.805 - 6.335 6.83 12 97 - 7.945 9.24 24 12.9 11.3 8.31 11.94 48 16 - : .136 say p = ow, UoHeULIEIUOD isd 90} = emsseig paouejeqierg Zep eduses 405 ayo ANI0aA 1° OANBIA wo ‘souesiq pug wee eee 9 s ’ ¢ z 1 he ran070N Fr oO A201 poresmes 10 Yeon ‘Aooj9n pobewed sy s/w ‘Ay}90190137 Say Z} = eu] UOMBUIWIEWUOD Isd QQ} = amnsseg paoujeqiong ‘Ayo, pobeweg soy a aoe ooze | gi-r ejduseg 40} e1yolg Ayooj9q : z'g OunB | puz esuog wo ‘aouessiq. pug uve oo # 4 oh HO tO OF 6 8 2 9 GF Ee OO = — —- + oor i eer ete eee ere —_ Ogre (92° x) Ayo019A Arq) unoW | | | 0 | f come | = ‘Ayo010n 3 pateimes 10 ween once & 3 od | Pe138 | Siu y2 = SUL, UOeULE;U0D Isd 00} = enssoig poouejeqiang pi-f ejduies 104 eyord Ayoojen :¢°g anBiy tuo ‘eoueysig pu esueg puz ue] 92 v2 2 ze 12 02 Gt OL LI OF GE HL EL ULE OG) eee cece eo a 7] see eve Gert) Apo}on Aq uwope ozoz ‘Arsojen poseweg sony —————————————————————E—e oe139 say gy = einssaig UoHeUTWIE;UCD Isd 00} = einssaig pasuejeqieng OP ejduies 405 opyord AWOOI9A 2p'g OANBIg pug esueg uro ‘eoueysiq pug ueoy Rew UHMH ANE e Loo roe E (10 Arpoyn ha w0on ‘2070 Awe Arojen pobeweg seyy ove ore 0052 = 3 8 oase & 2 oosz @ 0992140 Say pz = eWLLUOReUWE;UOD Isd 00z = eansseig pesuejequeng Li-f ajdwies 104 ajyoig Ayo0]@A :¢°g eunBi4 pug esueq oc ob ag Zt of St rl ce Zt uo OF 6 @ £9 G Fe ee wo ‘aouessiq. poe ° (sex sxiAio01%9 fig ueor ‘Aj20I0n pabeweg soy osee i ‘spur ‘AD0}0,141 sunoy p = ewilL UOReUIWEIUOD Isd 90g = eunsseig peouesqienQ €b-f eidwes 10} ejyorg AWOOJeA :9°g eunBiy pug esueg wo ‘eouesiq pug veo (et x) Auoojen fig ueope ‘Auojen pobeweg s0Hy ose ose g 8 ooze 7 sju ‘A0}01492 say Z}, = eully UoReUjUE;UOD isd 90¢ = eunsseig poouejequeng Lf ojdues 104 ajyorg AYOO[EA #2°g O4NBiy 308 pg weer pugs wo ‘aoue}sig o 6 8 £ 89 8 + © 2 + 2 oor y 0sKe 22 TTT TT Fei taen ea ww | ease eines ose vonens | cose j owe Asojan pabowed sayy spi ‘Ay2018A,143 Say pz = SUE, UOeUIWIEZUOD, isd 90g = eunsseig paoueleqieng Vast ojdwies 10) ayorg Aj90]8, :g°g enBi4 on 6 a Z 9 s ’ © @ ' pug esueq wo ‘eoue}siq pug uee7 o 20100 poreamies 10 weer ‘Ayooren poBewed yy ossz 082 s/w ‘ANOO}OA,144 say p = oul, UoMeUNUETUOD isd 001 = ensserg psouejeqieng Sf o[dutes 40) elyorg AysojeA : 6°g enBig urs ‘eourysiq | puz esueg pug uee7 on 6 8 Z 9 $ ’ e z ° + 0082 a cea er wipoen tog veer osre cose = Aporen 3 pavemies 10 UesH ose | 3 o 092 ‘Arooren pabewed sayy see145 siy Zp = oly UOEUIWEIUCD isd 002 = einsseg peouejeqieng Gf a[duies 105 eYo1d AYOOIeA :01'8 OANBIY puz esueg wis ‘eoue}sig. puy ueoy os 6 8 Z 9 s ’ © z , 0 (62'1 3) Aopen fig weon Aooren paveunies 10 veo Aroojen poBewea soy + os9z os9e osez og see sjuu ‘AD010A,146 sil pz = ull) UOEUTWIEUOD Isd 902 = einsseig paouejeqiend Ob-f ofduies 404 eyorg ApOo}aA :L1°g OANBIY pug esueq wo ‘eoue}sig pug ueey o 6 @ £ 9 $ # € 2 £ O KF @& Bee = S222 Fez es GEL Mhiooren Ka ues Ayooren paiesmes 10 weon ‘Ayojen pobeweg sey seo su92 jus ‘Ay}00)00 sete sue1467 8.2 Overbalanced Pressure (AP) One of the main objectives of this experimental study is to evaluate the effects of overbalanced pressure (AP) on the formation impairment. Experiments were cartied out using four different AP values of 100, 200, 300, 700 psi for this purpose. 8.2.1 Regain Permeability The effect of AP on the regain permeability, which is defined as mud damage permeability divided by initial permeability, is evaluated for different mud circulation times. Figures 8-12 to 8-16, present the effect of AP on the formation permeability. A general trend exhibiting a decreasing regain permeability with increasing AP is observed. Figure 8-14 illustrates the regain permeability for different AP maintained for 24 hrs during mud circulation. Decreases in regain permeability of 75.6, 57.7, 32.6 & 26% were observed at AP of 100, 200, 300, & 700 psi respectively. One possible reason for this decrease is that, with the increase in AP, more force acts on the mud solid particles to enter into pores while circulating across the formation. Accordingly, Micro-fine particles with mud filtrate migrate deeper inside the formation. On the other hand, somewhat large sized fine particles (> 4/3 of pore size) start sticking to the pore throats, thereby148 008 “sunoy p seye Apiqeeuied ujeBoy uo einssaig peouejedieng Jo j90 :ZL-g enB|4 Jsd ‘ounssaie pooueyeqiono 002 009 00g 00r 00 00z 00 sayy seu, % ‘Aupiqvouiog uyeBox109 uyeBoy uo aunsseld peouejeqieng Jo 39043 :¢1-g eanBi4 Isd ‘oinssaid paauejequaAQ 009 “sunoy Z| Joye Ayiqeeuied 00s oor 008 o0z oor suy Zp = oul on 4 g 8 % ‘Aypiqeeuued ujeboy on oor150 “sinoy pz Joye Aypqeewed uyeBoy Uo eunssoig peouRjeqierg Jo 19eH3 :p1-g aNBLy sd ‘ounssaid peauejeqiaAQ. 002 009 00g oor o0€ 002 oot say ye = OLE pot} ao ge g 8 % ‘Aypiqeeuusd uyjeboy151 “sinoy gp 1ene {sd ‘oinssaid pesuejeqiaAQ eoUed ujeBey Uo ainsseig peouEseqiend Jo 390H5 :¢1-8 eANBL4 oz 009 00g oor oe 002 oor 0 +0 on ee -- - was ~S oz s oe ‘S4y gp = OWL oa oF og 09 ou 08 06 + 004 % Ayygqeeuueg wyebey152 peUed ureBey uo ainssalg pedt BAO JO DENA :9)-8 eunBiy Isd ‘eunsselg paouejequeng one 009 0s oor ove ove oot en ee = SUy Bp = OWL it ~A See NL — s1y y2 = own mee se. ~ ~ am SA suet sou, \ ~ ~ ~ ~ \ x ~ x —— ¥ ~ ~ NN ~e ~ N ~ ~AL N s1y p= oun se. \ se SK ~s.L UTS --5 on oe op os on % ‘Aupqeauued ujeboy153 blocking the pore constrictions/throats more firmly. During backflow, these solid particles remain there in the pore throats and ultimately decrease the return permeability. At low AP, the solids donot stick themselves so firmly because less force acts on them and they flow out relatively easily compared with the behavior at higher AP during back flow. 8.2.2 Invasion Depth The influence of AP on the invasion depths for different contamination time is presented in figures 8-17 to 8-20. One would expect an increase in invasion depth with AP in the sense that higher pressure would force mud particles and filtrate to penetrate deeper and deeper into the formation. Contrary to this expectation, however, the result of our experimental work indicates a decrease in the invasion depths with increase in AP up to a certain value i.e. 300-psi. Beyond this value, invasion depth starts to increase with AP in line with expectation. Figure 8-19, shows the experimental results on invasion depths as a function of AP (which is maintained for 24 hrs). it is observed that invasion depth is considerably higher i.e. 12.9-cm for AP=100-psi than for AP=200 & 300 psi AP which are 11.3-cm & 8.31-cm respectively. Further increase in AP tends to increase the invasion depth reaching 11.97-cm for 700-psi. It can be inferred from this observation that AP of 300-psi acts as a critical pressure which154 produces minimum invasion depth. Figure 8-20 compares the effects of AP on invasion depths when it is maintained for different lengths of 4, 12 & 24 hrs. Evidently, all curves follow the same trend. The unexpected observation that lower APs can produce higher invasion depth, is very significant. In an attempt to check if this has been documented in the literature before, we have reviewed some relevant published papers, and some interesting observations are given below. Back in 1959, Monaghan et al. "!, in a laboratory study relating to formation damage in sands containing clays, mentioned that at high pressure gradients and flow rates, randomly dispersed particles apparently tend to interfere with each other as they approach pore constrictions and finally bridge. However, at low pressure gradient and flow rates, the particles are in more gentle movement and may either (i) gradually align themselves so that one by one they can work their way through the constrictions without bridging or (ji) are displaced in the water envelop to non-blocking position out of the main flow stream or in the interstitial water layer. Similarly, Krueger & Vogel (1967) conducted an experimental study in order to determine the effect of pressure drawdown on clean up of clay or silt- blocked sandstone. They reported that permeability damage from drilling mud or water can be substantially avoided through use of an initial low-pressure gradient155 followed by gradually increasing gradient during clean-up by reverse oil flow (figure # 8-21). They concluded that higher-pressure gradient and rates apparently cause particle bridges. A good recovery of permeability obtained through use of low rate clean up of sandstone core after water or filtrate invasion. Low flow rate permit flow of individual particles through pores without interference from their neighbors. Gruebeck & Collin", in an experimental work, also support the ideas of Monaghan and Krueger. They concluded that high flow rate (high AP) cause more entrainment of particles and results in decline in permeability than at low flow rate. In another paper, Krueger ' stated that continuing research on formation damage during drilling lead to an unexpected observation that emphasized the importance of particle movement: better recovery of permeability to oll was obtained during backflow of oil in a mud damage core at low pressure drop and therefore at low rate than at high pressure drop.156 ‘uojeinouyo sanoy p 48ye UO|SeAu] pnw Jo uideg uo einsserd pesueleqieng Jo Jey :21-8 einBi4 Isd ‘eunssolg pesuejequesg ove 009 008 oor ove ove os 0 ~S | sayy «oun — wis “jdeg uojseauy151 ‘uoneinoujo sinoy Z}. Joye UoIseAUt pnpy so uydeg uo oinssesg peouejequeng Jo Joey :eL-g ein6i4 Isd ‘esnsselg pauelequang oo 009 008 oor oe ove ot ae ‘say 24 = ou, s a eh oh sh wo ‘indeq uojseauy158 “UOHJE[NDIO SINOY PZ Joye UOISBAU pniq Jo yjdeg uo einsseig pesuEjeqiend Jo 39843 :61-8 eNBLS Isd ‘insseig peaueleqiaaQ oe oe 008 008 oor o0€ 02 oor ¢ 9 Z | ei | a = “7 ‘\ & - 6 ae \ $ “7 \ a9 “7 \ 3 uc % “eS Pod suy ye = owns yA ag159 008 “uolsenuy pny Jo uydeq uo einsseig peouejeqieng jo 19143 :0z-8 OuNBI Isd ‘eunssaig paouejeqieng 00L 009 00S oor 00 eS een say y = ous - -----e 3 eo _——e~ na oie N & e-7 uc s su zes out 2 “7 3 “7 = 2 “7 3160 E00 = 20 ag 2 ks oe eg 8 Sie Pare & Sz z > 40) s|3 = (8c . Bla =| 3 [32 6 B 38] 3 132 i 20) gle 23) B18 é oleae esi e 12 So OWL GASE WaTER Base. DRILLING FLUID DRILLING FLUID DRILLING FLUID TYPE Figure 8-21: Effect of BackFlow Pressure on Permeability Recovery of 12-inch long Core Exposed to Circulating Drilling Fluid.161 In 1997, Saleh at el." also discussed the above ideas of Gruebeck and Krueger in the introduction of their paper. Basically they confirmed the observations forwarded by Krueger. In particular, they mentioned that at low flow rates, dispersed fines can gradually align themselves to migrate individually through the pore constrictions or can become immobilized in the interstitial water envelope at the pore wall. At high flow rates, the randomly distributed fines interfere with each other and bridge in a “brush heap’ effect. All the above mentioned authors reported that low AP yields better permeability recovery mechanism based on their work during cleanup process of the formation after it was invaded by mud or water. The same mechanism applies during drilling operations. At low AP, micro-fine particles that are in suspended form in drilling fluid can move gradually through the mud cake & pore and can cause deeper invasion. While at high AP, particles move randomly, interfere with each other and bridge at pore constrictions. Another point that should be explained is the increase in invasion depth with AP beyond 300 psi. It is believed that at low AP e.g. 100 psi, the mud solid particles gather themselves loosely at the pore constrictions. These loosely blocking particles allow filtrate and micro-fine particles to pass through them and in this way deep migration occurs as shown in figure 8-22a.162 As AP increases, these particles start to form relatively tight bridges at the pore throats and begin to act as “one way check valves’, which allow less filtrate and micro-fines through them and to move into the formation as seen through decrease of invasion depth when AP increased to 300-psi. At 300-psi AP, particles appear to form a perfect check valves as they developed the tightest bridge at the pore constrictions which is presented in figure 8-22b. Further increase in AP results in breaking of these tight particle bridges. Either these particles break into small sizes or pore constrictions/throats get enlarged by entrainment of some grains from the pore throat walls, which help filtrate and solids to migrate until they again stick at other constrictions. Figure 8- 2c illustrates the mechanism described above. The above mechanism is also discussed by Gruebeck "7 in his study relating to entrainment and deposition of fine particles in porous media. He reported that while the smaller pathways are blocked as a solid laden fluid flows in a porous media, the flow is diverted to larger pathways and more fine particles reach the effluent. But eventually the pressure differential across the blocked pathway is so great that some plugs (bridges) are broken. These result in a spurt of fines in the effluent but then new deposits start growing,163 (© Mud Solid Particles © Micro-fine Mud Solid Particles p> Drilling Mud & Mud Filtate Micro-fine Particles & Fitrates Figure # 8-22.a : Bridging mechanisin of the mud particles at the pore throat for 4P=100 psi. Drilling Fluid Micro-fine Particles & Filtrates Figure # 8-22.b : Bridging mechanism of the mud particles at the pore throat for AP=300 psi. Dritting . Fluid Micro-fine Particles & Filtrates Figure # 8-22.c : Bridging mechanism of the mud particles at the pore throat for P2700 psi.164 8.2.3. Skin Variation Skin factor for each of the above described cases is calculated by Hawkin's equation © where damage permeability is the most influencing factor. Our experimental results show a decrease in return permeability with increasing AP while the invasion depth decreases first and then starts to increase when AP increases. The variations in skin factor with AP are presented in figures 8-23 to 8- 26. It is apparent that skin factors increase with increase in AP primarily due to decrease in regain permeability. Figure 8-25 shows a case where AP is maintained for 24 hours and the skin factor is increased by 115%, whereas regain permeability decreased by about 23.6% for an increase in AP from 100 to 200-psi. Similarly, an increase in skin of 438 & 750% was observed with the decreasing of regain permeability of 57 & 66 % for AP of 300 and 700-psi respectively. Figure 8-26, shows a general trend of increase in skin factor with AP for different time periods.165 SANOY f 10} peurejuiew eanssaig peouRjeqieng YUM UOHeURA UNIS : €7-8 e2NBLy 18d “eunssoid peouejeqseno 008 0S2 O0L 099 009 085 005 Or OOF OSE ODE Ost OO OS OOF os Oo ee ooo ‘sayy = ou, on ups186 ‘Sunoy ZI} 405 pouleyuyew ainsseig peouejeqieAg YIM UOHeHEA UPS : pZ-g euNBI Jsd “sinssaig peouejeqiang 008 0S£ O0L 059 009 OSS 00S OSr OOF OSE OE ose O0Z Ost OOF Os -----a . say ZL = ou et oe ups167 sanoy pz 404 peuejusew ainsseig peouejeqieag WIM UOeLEA UPIS : $Z-g eanBLy Isd -aunsseig paouejeqioag 008 0S 00L 059 009 OSS 00S Os Or OSE ONE OsZ O0e Ost OOF Os168 008 ose einsseig peouRjeqieng YpM UOHEUEA UIs : 92-8 ANB Jsd ‘insseig paouejeqieno 00£ 089 009 08S 00S OSr OOF 05 ODE Osz OO Ost OOF os 0 ers = ==a a“-7 an ot - - - _-7 uo / nm - / supeoum : a s a o ie = a ‘Say Zh = OUULL Sw a _-7 ee ups169 8.3 Contamination Time During driling operation, contamination time or mud circulation time depends upon the rate of penetration and the length of the well to be drilled. Especially for horizontal wells, contamination time is a very critical factor because the horizontal section of the well has to be drilled through the producing zone. Keeping in view its importance, the influence of contamination time on regain permeability, invasion depth and skin is also investigated during leak-off tests. 8.3.1 Regain Permeability Regain permeability decreases with contamination time. Figures 8-27 to 8- 30, present the regain permeability profile as a function of time for different AP. Figure 8-27, shows a decrease in regain permeability with time as AP is kept constant at 100-psi. Initial permeability decreased to 86.6% after 4hrs. Similarly, after 12, 24 & 48 hours, it declined to 83.9, 75.6 & 35.2% respectively. It is apparent that with the increase in contamination time, mud solids and filtrate migrate deeper and deeper into the formation and gradually impair initial permeability as the time passes. In figure 8-30, all three curves follow the same trend of decreasing regain permeability with increasing contamination time.170 0g “einsseud peourjeqieng Isd go} sepun uyeBoy yj Uo aus UoMeUjWEUOD Jo EN 222-8 euNB) or ge say ‘outs se ob on oL og oor % ‘Aupqvauuad uyjeboy7m “ainsseud peouejeqieng isd gog Jepun Ayjiqeeued ujeBoy oyy Uo aul} UoeUWeWUOD Jo DENA +gz-g ONL ‘say ‘oul, 08 sr ov se of sz oz ca o 8 tsd 006 = dF on of oF % ‘Ayiqeouueg ujeBox72 “eunsseid paouejequeag isd 902 sepun Ayyiqeeuued uleBey aug uo aul, UoeWIURIUOD Jo DEY :6z-8 NBL say ‘outs, OO ace epee 0s 5 ~~ a oz Tae ~ isd 002 = a7 —— oc ~N sy oF 0s } 09 | oe 9 06 % ‘Myjiqeouued ujeGoy173 “Ayqeouued ujeBoy oyy uo aut) uoneuueyUCg 40 98H :0€-8 BNBI say ‘uly os gh hE SH $ 0 o To 184 002 = av on Srlttt = Zotz tl oz “Sire = wrote I 7 vsd oot = av = oe oe > x Y or ~ ~ s \ | ~ % os s - \ mS \ ~ ‘ 09 wd ~ x \ tsd 00b = av a a Vb oe ~e ~~ oo . 08 | Tae Dif 06 % ‘Auiqeouueg weBoy174 8.3.2 Invasion Depth Results of experimental study indicate from examination of figures 8-31 to 8-34 that the invasion depth increases with the contamination time. Figure 8-31, presents mud invasion depth versus contamination time for a AP of 100 psi. It should be observed that the invasion depth is not constant with time. After 4 hours of mud circulation, an invasion depth of 7.8 cm was measured, which raised to 9.7 cm after 12 hours with an increment of 3% per hour. After 24 & 48 hours, mud front moved to 12.9 & 16 cm with 2.75 & 1% per hour increment respectively. The above decline in the rate of mud movement is dictated by the filtration rate through the core, which decreases with increase in contamination time. 8.3.3 Skin Variation Both permeability impairment and invasion depth increase with the contamination time, causes skin factor to rise. Skin variation profiles with contamination time for different AP (100, 300, 700 psi) are presented in figure 8-35. From these profiles, one can easily infer that the increase in skin factor is more influenced by permeability damage than invasion depth. As discussed before, with increase in contamination time and AP (which cause more damage), skin jumps to higher values as shown in figure 8-36. ‘At AP=100 psi (which is a common field practice), the skin profile follows a linear trend as shown in figure 8-36. The equation of this linear line is given by175 Skin = 0.1126 x Time ~ 0.8087 (8.1) The above skin equation can be used to predict skin variation with time. If the horizontal section of a well takes 20 days to drill, one can predict a skin value of 55 at the toe after 20 days as shown in figure 8-37. 8.3.3.1 Skin Profile Along The Well Length Contamination time can also be related to the length of the well by Rate of Penetration (ROP). In this manner, skin profile can be obtained for the entire length of the well. Since, Length ROP Time _ Length ROP If the ROP remains constant throughout the drilling operation, equation 8.1 will be, Lengthofthewell Skin = 0.1126- —0.8067 (8.2)176 Equation 8.2 can be used to predict skin factor along the length of the well. Suppose that ROP of a well is 5ft/hr, equation 8.2 will give same skin value of 55 at the well length of 2500 ft, as shown in figure 8-38.m7 (isd 001 = eunssesg peouejequeag) “ell, UONRUIWEJUOD YBIM a1YO1g YIdeq UOLseAY PIN : Le-g esnByy say ‘oui og Sr or se oe se oz st or s wo ‘ade uosseauy178 (isd ogg = esnsseig peouejeqieng) “Oull UOREUIWEWUOD UI elyoud yjdeq uojseaul pny : ze-8 asnBig suy ‘oui sr oF oe se oe on on s oe ene e eo 18d 006 = 7" on a n oF a oz. wo ‘yydeg uojseauy179 (isd 002 = eunsseig peoueyequeag) souls UOReUIWEIUOD YM elyold UJdeq UOIseAU} PnIW : ¢e-8 eUNBIg say ‘outs os s% or se oe sz oz su o s t o z ’ 9 = s ae 8 “7 es an 3 oT “> cd 3B - ue 18d 00L=¢7 9 3 oz180 eu} UOIJEUIWIE]UOD UWM ajyO1d YIdaq UO;seAU pn : pE-g ENGI suy ‘oui ee a e228 Bia =-T252 a 7a - te Wdo0=dP ye Te wr - ---7 isd oot = dP wo “indeg uo}seauy181 ainsseid PeoUuejequeAQ queleyiq 4epun outs UoPUIWE;UOD YIM [Youd UPI : Se-g @NBLY say ‘ous a es | ~*~ a oo of “7 2 e 7 - - / - isd 00b= a P_ t Z 2 2 Zi 7 - . Z 2 - isd 00e= dP a 2 _- -7~ = - 18d 00L= 47 ups182 sp “einsSeid PSOUR|EqIEAG jsd 00} Jopun ew uopeUWE;LOD Y}IM o]Yo1d UY 298-8 ENB say ‘oul, op se oe @ oz os on s o 18d 00= 47° tze90= 49080- oztL0=4 uopejodenx3 adh nt urs183 poled papuayxg 40} einssaig paoueleqierQ Isd 90} 2epun out! UoHeUwIE OD YYM B1YOId UPIS +ZE-8 eANBLS ‘say ‘oul, ocr or OSE ESCH o 70 o oz or or 2 18d 00=d7" oo ¥ 5 + 09 0 vae0= 08 (£908°0- x9Z1b'0 =A 06 uogejodenxs odAy se9ur7184 4yBup] [1am ey) Buoye e]yosd UpIS : ge-8 EANBIy 4 ‘yjbueT cost 008 008 ° isd 0012 47 bzag0 = 24 1908°0-9ZLV0= 4 uoperodenxa odKt se0ury upsCHAPTER 9 CONCLUSIONS & RECOMMENDATIONS 9.1 Conclusions An experimental study has been conducted to investigate the influence of Overbalanced Pressure and Mud Contamination Time on the formation impairment during drilling operation. An innovative non-destructive ultrasonic method was employed to determine the mud invasion depth. The study was conducted using a dynamic leak-off apparatus. Berea sandstone core plugs were used as core samples during the experiments. The fluid used included 3.5% KCI brine, medium arabian crude oil (Specific gravity=0.862; Viscosity=11.19 cp @ 24°C) and water based polymer drilling fluid (which is being used in most sandstone reservoirs of the Saudi Arabia). 185186 Based on the presented experimental results and analyses of the regain permeability, mud invasion depth and skin factor, the following conclusions are made; 1. Mud invasion depth decreases with the increase in overbalanced pressure down to a minimum mud invasion depth, obtained for a critical value of AP observed to be approximately 300 psi. Beyond that, invasion depth starts increasing with AP. 2. Anew ultrasonic method has been developed to determine the mud invasion depth, The performance of this innovative technique is impressive as it simplifies the measurements and saves time without destroying the samples. 3. During the mud invasion process, the mud penetration increases with the contamination time. 4, Regain permeability decreases with the increase in AP & contamination time. AP and contamination time should be kept low in order to get maximum production by avoiding excess formation damage, 5. Skin factor increases with the AP and the contamination time.187 6. Based on the observed experimental data, an example of application of the result to horizontal wells is presented. it will illustrate how the method can predict skin profile with the contamination time and along the horizontal well length. 9.2 Recommendations 4. A decrease in invasion depth is reported for AP < 300 psi. Beyond this pressure, invasion depth increases with the overbalanced pressure. in this manner AP= 300 psi acts as a critical pressure which produced minimum invasion depth. A more detailed study is recommended which can exactly pin point this critical pressure, 2. In the present study, ultrasounds were generated by Pulser and a velocity profile along the core length was obtained with the resolution of 0.5om. To get more precise reading with a resolution ~ 1.0 mm and to avoid couplant error factor, LASER generated ultrasounds alongwith laser-based detection of ultrasound should be used in future work. 3. Variations in sound velocities & attenuation of the waveform while moving through the core sample, can also be related with other petro-physical properties of the reservoir rocks and fluids. These should be investigated.11] [2] (3) (41 CHAPTER 10 REFERENCES S.T. Saleh, R. Rustam, W. EL-Rabaa, & M. R. Islam "Formation damage study with a horizontal wellbore model". Journal of Petroleum Science and Engineering 17 (1997). pp 87-99. T.P.Frick, M.J. Economides , "Horizontal Well Damage Characterization and Removal". SPE-21795, SPE Production & Facilities, Feb 1993. D. Brant Bennion, F. Brent Thomas, Ronald F. Bietz, "Formation Damage and Horizontal Wells - A Productivity Killer?", SPE-37138, 2" International Three-day Conference on Horizontal Well Technology, Calgary, Alberta Canada, 18-20 November, 1996. Van Everdingen, AF. "The Skin effect and its influence on the productive capacity of a well”, Trans., AIME, Vol. 198, pp.171, 1953. 188