Nanoparticle-based Drilling Fluids with Improved ...

University of Calgary

PRISM: University of Calgary's Digital Repository

Graduate Studies The Vault: Electronic Theses and Dissertations

2013-09-16

Nanoparticle-based Drilling Fluids with Improved

Characteristics

Zakaria, Mohammad Ferdous

Zakaria, M. F. (2013). Nanoparticle-based Drilling Fluids with Improved Characteristics

(Unpublished doctoral thesis). University of Calgary, Calgary, AB. doi:10.11575/PRISM/27055

http://hdl.handle.net/11023/977

doctoral thesis

University of Calgary graduate students retain copyright ownership and moral rights for their

thesis. You may use this material in any way that is permitted by the Copyright Act or through

licensing that has been assigned to the document. For uses that are not allowable under

copyright legislation or licensing, you are required to seek permission.

Downloaded from PRISM: https://prism.ucalgary.ca

UNIVERSITY OF CALGARY

Nanoparticle-based Drilling Fluids with Improved Characteristics

by

Mohammad Ferdous Zakaria

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF CHEMICAL AND PETROLEUM ENGINEERING

CALGARY, ALBERTA

SEPTEMBER, 2013

© Mohammad Ferdous Zakaria 2013

ii

Abstract

The success of well-drilling operations is heavily dependent on the drilling fluid. Drilling

fluids cool down and lubricate the drill bit, remove cuttings, prevent formation damage,

suspend cuttings and also cake off the permeable formation, thus retarding the passage

of fluid into the formation. During the drilling through induced and natural fractures, huge

drilling fluid losses lead to the higher operational expenses. That is why, it is vital to

design the drilling fluid, so that it may minimize the mud invasion in to formation and

prevent lost circulation. Typical micro or macro sized lost circulation materials (LCM)

show limited success, especially in formations dominated by micro and nano pores, due

to their relatively large sizes. The objective of this thesis was to investigate the

performance improvement by the usage of NPs (nanoparticles) as lost circulation

additives in the drilling fluid. In the current work, a new class of nanoparticles (NPs)

based lost circulation materials has been developed. Two different approaches of NPs

formation, and addition, to water based and invert-emulsion drilling fluid have been

tested. All NPs were prepared in-house either within the invert-emulsion drilling fluid; in-

situ, or within an aqueous phase; ex-situ, which was eventually blended with the drilling

fluid. The laboratory measurements included measuring mud weight, pH, lubricity

viscosity, gel strength, standard API LTLP filter test and high temperature and high

pressure (HTHP) test. In this work we evaluated fluid loss performance of a wide range

of NPs preferably selected from metal hydroxides, e.g. iron hydroxide, metal

carbonates, e.g. calcium carbonate and metal sulfate and sulfide e.g barium sulphate

and ferrous sulfide respectively.

The use of improved NP-based invert emulsion drilling fluid showed an excellent fluid

loss control, rheological properties together with a good lubricity profile. This thesis

reports an experimental and theoretical study on filtration properties of invert emulsion

drilling fluids under static conditions. Under API standard filtration test at LTLP and

HTHP, more than 70% reduction in fluid loss was achieved in the presence of 1-5 wt%

NPs.

iii

The results have also shown that the filter cake developed during the NP-based drilling

fluid filtration was thin (thickness less than 1 mm), which implies high potential for

reducing the differential pressure sticking problem, formation damage and torque and

drag problems while drilling. Moreover, at the level of NPs added, no impact on drilling

fluid apparent viscosity, and the fluid maintained its stability for more than 4 weeks.

Other NPs prepared by in-situ and ex-situ method also showed an excellent fluid loss

control. Results of the modeling showed that NP-based drilling fluid didn’t follow the

Darcy equation at the initiation of filtration and therefore the initial region was found flat

and nanoparticles reduced the premeability instantly. It was also shown that

nanoparticles transport in filtration was predominantly influenced by the Brownian

diffusion. Compare with the drilling fluid alone and drilling fluid with LCM, increasing

shear rate did not increase the same extent of shear stress in case of NP-base fluid

(both ex-situ and in-situ prepared), which can be attributed to the fact that smaller

particles were dispersed more effectively than the larger bulk particles and provided

bridging between clay particles due to their larger surface area. Tailor made NPs with

specific characteristics is thus expected to play a promising role in solving the

circulation loss and other technical challenges faced with commercial drilling fluid during

oil and gas drilling operation.

iv

Acknowledgements

All kinds of praise and all thanks belong to ALLAH, the One, the Lord of the Universe,

the Creator, the Most Gracious and the Most Merciful.

I would like to express my deepest sense of appreciation, gratitude and

indebtedness to my respected supervisor Dr. Maen Husein and co-supervisor Dr. Geir

Hareland for the opportunity of being part of their research team. Thanks Prof. Husein

and Prof.Hareland, it has been a great honor for me to be associated with your team. I

greatly appreciate your continuous support, excellent supervision and encouragement

throughout this work. I would also like to thank my defense committee members Dr.

Roberto Aguilera, Dr. Brij Maini, Dr. Ronald J. Spencer and Dr. T. Nguyen for their time

and providing their constructive criticism of my work.

It is also a pleasure for me to express again my sincere appreciation and

profound regards to Dr. Maen Husein and Dr. Geir Hareland for providing the guidelines

to efficiently conduct all the laboratory experiments, constructive suggestions and

criticism throughout the period of research work. I am also thankful to Dr. Husein in the

final preparation of the manuscript. I am grateful to Ms.Patricia Teichrob for editing my

thesis and all kinds of support during my research works.

I would also like to extend my sincere thanks to the current and past members of

the Nanotechnology for Energy & Environment (NTEE) research team; Salman Al-

khaldi, Belal Abu Tarboush, Ahmad Al-As'ad, Alex Borisov, Nashaat Nassar, Zied Ouled

Ameur, Amr Abdelrazek Elgeuoshy Meghawry Abdrabo and everyone in the Real-Time

Drilling Engineering Research Group for their support, brilliant ideas and

encouragement.

I wish to thank all the staff of the Chemical Engineering department for their

valuable support. Special acknowledgement to Bernie Then and Ms.Paige Deitsch for

providing a convenient entourage to conduct the laboratory experiments.

I would like to thank team members of nFluids Inc; David Edmonds and Jeremy

Krol for their constructive criticism and support in our current research. And I also would

like to extend my appreciation to my colleagues at Ineos Oligomers and my family friend

v

Dr.Soumaine Dehkissia for their unconditional support in various ways which have

inspired me this far.

This research was financially supported by a grant from the Natural Science and

Engineering Research Council of Canada (NSERC), Talisman Energy Inc and Pason

Systems. This support is gratefully acknowledged. Finally, my acknowledgments go to

Queen Elizabeth II Graduate (Doctoral) Scholarships for financial support.

And last but not the least, I profoundly acknowledge gratefulness to my beloved

parents and wife who have provided constant encouragement.

vi

Dedication

This works is dedicated to:

My lovely supportive parents, brothers and sisters

My beloved wife, Asma Sharmin

And my wonderful daughter Subah Maknun

With love and appreciation

vii

Table of Contents

Abstract ................................................................................................................. ii Acknowledgements .............................................................................................. iv Dedication ............................................................................................................ vi Table of Contents ................................................................................................ vii List of Tables ......................................................................................................... x List of Figures ...................................................................................................... xii List of Symbols, Abbreviations and Nomenclature .............................................. xv

CHAPTER ONE: INTRODUCTION .......................................................................1 1.1 Problem statement and significance of the research ...................................1 1.2 Research Objectives....................................................................................4 1.3 Organization of the Thesis ...........................................................................7

CHAPTER TWO: LITERATURE REVIEW ............................................................9 2.1 Introduction ..................................................................................................9 2.2 Drilling fluid Classification ..........................................................................10 2.3 Functions of Drilling Fluids .........................................................................11 2.4 Drilling fluid related challenges ..................................................................12 2.5 Clay Chemistry used in drilling fluids .........................................................16 2.6 Nanoparticles .............................................................................................20

2.6.1 Nanoparticle synthesis.....................................................................24 2.7 Nanoparticle-based drilling fluids ...............................................................26 2.8 General characteristics of drilling fluid filtration ..........................................36 2.9 Filtration mechanism..................................................................................42

CHAPTER THREE: EXPERIMENTAL METHODS..............................................48 3.1 Drilling Fluid Samples ................................................................................48 3.2 NPs and NP-based drilling fluid formation .................................................49

3.2.1 Ex-situ preparation of NPs ...............................................................50 3.2.1.1 Fe(OH)3 NPs .........................................................................51 3.2.1.2 CaCO3 NPs ...........................................................................51 3.2.1.3 FeS NPs ................................................................................52 3.2.1.4 BaSO4 NPs ............................................................................52

3.2.2 In-situ preparation ............................................................................53 3.2.2.1 Fe(OH)3 NPs .........................................................................53 3.2.2.2 CaCO3 NPs ..........................................................................54 3.2.2.3 FeS NPs ................................................................................55 3.2.2.4 BaSO4 NPs ............................................................................56

3.3 Characterization methods and techniques .................................................56 3.3.1 Particle characterization ..................................................................56 3.3.2 Toxicity evaluation ...........................................................................57 3.3.4 Emulsified water droplet measurement............................................58 3.3.5 Drilling fluid characterization ............................................................58

viii

CHAPTER FOUR: RESULTS AND DISCUSSION ..............................................63 4.1 Fe(OH)3 Nanoparticles (NPs) Characterization .........................................63

4.1.1 X-ray diffraction analysis..................................................................64 4.1.2 Water droplet size distribution .........................................................66 4.1.3 Size distribution of ex-situ prepared Fe(OH)3 NPs ..........................67 4.1.4 Determination of particle size of in-situ prepared Fe(OH)3 NPs .......68

4.2 Drilling fluid Characterization .....................................................................70 4.2.1 Stability of NP-based Fluid ..............................................................70 4.2.2 LTLP Filtration .................................................................................71

4.2.2.1 Commercial NPs ....................................................................72 4.2.2.2 In-house prepared Fe(OH)3 NPs ..........................................73

4.2.3 Filtrate Characterization...................................................................78 4.2.4 HTHP Filtration ................................................................................79 4.2.5 Effect of high shear on fluid loss control ..........................................83 4.2.6 Effect of presence of organophillic clays on fluid loss ......................85 4.2.7 Effect of Oil: Water ratio on fluid loss ...............................................86 4.2.8 Rheology behavior of NP-based fluid ..............................................87 4.2.9 Drilling fluid density and pH .............................................................93 4.2.10 Drilling fluid lubricity .......................................................................93 4.2.11 Preparation and performance evaluation of Fe(OH)3 NPs in invert

emulsion drilling fluids provided by different suppliers .......................97 4.2.12 Performance of Fe(OH)3 NPs in water based mud (WBM) ..........100 4.2.13 Toxicity evaluation Fe(OH)3 samples ........................................... 102

4.3 CaCO3 Nanoparticles (NPs) Characterization ........................................ 103 4.3.1 X-ray diffraction analysis................................................................ 103 4.3.2 Size distribution of ex-situ prepared CaCO3 ..................................104 4.3.3 Determination of particle size of in-situ prepared CaCO3 .............. 106 4.3.4 LTLP Filtration of in-house prepared CaCO3 NPs .......................... 112 4.3.5 HTHP Filtration of in-house prepared CaCO3 NPs ........................ 116 4.3.6 Drilling fluid density and pH ........................................................... 117 4.3.7 Rheology behavior of NP-based fluid ............................................ 118

4.4 Invert emulsion drilling fluid API fluid loss characterization using other NPs ......................................................................................................... 121

4.5 Summary of the API fluid loss study of different NPs in Invert emulsion.. 122

CHAPTER FIVE: MODELLING ......................................................................... 126 5.1 LTLP API filtration model using Darcy’s law ............................................ 126 5.2 NP based fluid transport using Stoke-Einstein equation .......................... 140 5.3 Rheology model of NP-based fluid ......................................................... 145

CHAPTER SIX: CONCLUSIONS‎, CONTRIBUTIONS TO KNOWLEDGE‎ AND RECOMMENDATIONS ............................................................................. 149

6.1 Conclusions ............................................................................................. 149 6.2 Original contributions to knowledge ......................................................... 154 6.3 Recommendations for future research..................................................... 154

ix

REFERENCES .................................................................................................. 157

APPENDIX A: CLASSIFICATION OF LOST-CIRCULATION ZONES…………172 APPENDIX B: LOST CIRCULATION MATERIALS SIZE SELECTION METHODS …………………………………………………………..173 APPENDIX C: DIFFUSION COEFFICIENT AND PECLET NUMBER…………174

x

List of Tables

Table 2.1: Comparative cost analysis study of NP-based drilling mud ................33 Table 3.1: Compositions of the invert emulsion and water based muds employed in this work .........................................................................48 Table 4.1: API LTLP loss of drilling fluid in the presence and abscense of 1 wt% commercial Fe2O3 NPs. NPs were thoroughly mixed with the invert emulsion drilling fluid. No LCMs added to both samples ....73 Table 4.2: Comparative study of API LTLP fluid loss of drilling fluids with 1.6 wt% conventional Gilsonite LCM, and 1 wt% in-situ and ex-situ prepared NPs ....................................................................................74 Table 4.3: API LTLP fluid loss comparing drilling fluid and drilling fluid with Gilsonite LCM as base cases with drilling fluid samples containing the in-house prepared Fe(OH)3 NPs only ..........................................78 Table 4.4: ICP results of the filtrate collected following API LTLP to determine the Ca and Fe content .......................................................................79 Table 4.5: HTHP filtration property of different drilling fluid samples ..................81 Table 4.6: Effect of temperature and pressure on mud cake thickness ..............82 Table 4.7: HTHP Fluid loss of different drilling fluid samples using engineered NPs only .........................................................................82 Table 4.8: Effect of shearing effect on LTLP fluid loss control ............................84 Table 4.9: Effect of organophillic clays on LTLP fluid loss control ......................85 Table 4.10:Effect of Oil: Water ratio on Fluid loss Control when using LCM .......87 Table 4.11:Effect of Oil: Water ratio on Fluid loss Control when using NPs ........87 Table 4.12:Density and pH measurements of drilling fluid samples with LCM and Fe(OH)3 NPs ......................................................................93 Table 4.13:Co-efficient of friction (CoF) of drilling mud samples .........................94 Table 4.14:Coefficient of friction (CoF) and % torque reduction in the presence and absence of NaCl salt in the invert emulsion drilling fluid ........................................................................................96 Table 4.15:Effect of ex situ and in situ prepared Fe(OH)3 NPs on the performance of three different invert emulsion samples of drilling fluids provided by three different suppliers. Concentration of NPs 1 wt%, composition of invert emulsion: (90:10) oil:water (v/v) ...........98 Table 4.16:API LTLP WBM fluid loss with and without NPs .............................. 100 Table 4.17:Microtox bioassay of Fe(OH)3 NPs .................................................. 102 Table 4.18:API LTLP fluid loss comparing invert emulsion drilling fluid as base cases with invert emulsion drilling fluid samples containing the in-house prepared CaCO3 NPs using reaction-2 (R2) ................................................................................ 114 Table 4.19:API LTLP fluid loss comparing water based drilling fluid as base cases with water based drilling fluid samples containing the in-house prepared CaCO3 NPs by reaction-2 (R2) .................... 114

xi

Table 4.20:API LTLP fluid loss comparing invert emulsion drilling fluid as base cases with invert emulsion drilling fluid samples containing the in-house prepared CaCO3 NPs using reaction-5 (R5) ............... 115 Table 4.21:HTHP fluid loss comparing invert emulsion drilling fluid as base cases with invert emulsion drilling fluid samples containing the in-house prepared CaCO3 NPs using reaction-2 (R2) ..................... 117 Table 4.22:HTHP fluid loss comparing invert emulsion drilling fluid as base cases with invert emulsion drilling fluid samples containing the in-house prepared CaCO3 NPs using reaction-5 (R5) ............... 117 Table 4.23:Density and pH measurements of drilling fluid samples with CaCO3 NPs .............................................................................. 118 Table 4.24:Co-efficient of friction of invert emulsion drilling fluid samples......... 121 Table 4.25:API LTLP Fluid loss using BaSO4 and FeS NPs based invert emulsion .......................................................................................... 122 Table 5.1: Permeability reduction of mud cake in the presence and absence of NPs. (R2) and (R5) refer to the raction used to prepare the CaCO3 NPs per Section 3.2 ......................................... 133 Table 5.2: Experimental and Bingham Plastic viscosity and Yield point ............ 148 Table B.1: Lost circulation materials selection methods .................................... 173 Table C.1: Effect of particle sizes of DF (dp =2-200 μm in DF) on Peclet number which is a control sample of Fe(OH)3 NP-based fluid ......... 174 Table C.2: Effect of Fe(OH)3 NPs size in DF ranges from 0.001-0.3 μm (1-300 nm) on Peclet number .......................................................... 174 Table C.3:Effect of particle sizes of DF (dp =2-200 μm in DF) on Peclet number which is a control sample of CaCO3 NP-based fluid ........... 175 Table C.4:Effect of CaCO3 NPs size in DF ranges from 0.001-0.3 μm (1-300 nm) on Peclet number ........................................................... 175

xii

List of Figures

Figure 1.1: Schematics showing a) fluid loss control using NPs along with LCM, mechanism of pore throat blocking by b) plugging and sealing, c) bridging and mud flow restriction by NPs only, and d) fluid loss using conventional LCM ..................................................3 Figure 2.1: Drilling mud circulation down the drill pipe ......................................... 9 Figure 2.2: Drilled fines and fluid particles invasion into the formation ...............14 Figure 2.3: Drilling fluid loss into the formation ...................................................15 Figure 2.4: Basic units of clay minerals and the silica and alumina sheets ........ 16 Figure 2.5: Schematic representation of Montmorillonite clay (bentonite) Structure ............................................................................................17 Figure 2.6: Schematic representation of the fixed and diffused double layer near a clay surface ...........................................................................19 Figure 2.7: Particle size scale ............................................................................ 20 Figure 2.8: Pore throat sizes in rocks ................................................................. 21 Figure 2.9: Surface area to volume ratio of same volume of materials............... 21 Figure 2.10: Schematic representation of NPs in invert emulsion fluid ............... 25 Figure 2.11: A characteristic filtration plot of drilling fluid during drilling .............38 Figure 2.12: Three Types of Filtration Curves .................................................... 39 Figure 2.13: Bridging effects with varying particles diameter in pore throat ....... 42 Figure 2.14: Overview of Filtration mechanisms ................................................45 Figure 2.15: Effect of particle diameter on collision probability ........................... 46 Figure 2.16: NPs plugging probability during drilling ..........................................47 Figure 3.1: Schematic representation of the ex-situ method for NP-based drilling fluid preparation ...................................................................50 Figure 3.2: Schematic representation of in-situ NP-based drilling fluid ..............53 Figure 3.3: Schematic of In-situ prepared CaCO3 NPs-based drilling fluid using CO2 ........................................................................................55 Figure 3.4: Drilling fluid loss apparatus for a) LTLP and b) HTHP tests ............ 59 Figure 3.5: Fann Model 35A viscometer for measuring viscosity .......................60 Figure 3.6: OFITE drilling fluid lubricity tester .....................................................62 Figure 4.1: X-ray diffraction pattern for the ex-situ prepared iron-based NPs. ...64 Figure 4.2: X-ray diffraction pattern of ex-situ prepared Fe(OH)3 NPs collected on the filter paper ..............................................................65 Figure 4.3: Particle size distribution histogram of water droplet obtained from a water-in-oil emulsion by dispersing water into base-oil with the aid of primary emulsifier ....................................................66 Figure 4.4 : TEM photographs and corresponding particle size distribution histograms of ex-situ prepared Fe(OH)3 NPs in the range between a) 1-120 nm and b) 1-30 nm ............................................................ 68 Figure 4.5: SEM Images at 48x magnification of mud cakes following API LTLP filtration tests a) without NPs, b) with in-situ NPs (90/10 oil/water invert emulsion mud and 1 wt% Fe(OH)3 NPs) .................69

xiii

Figure 4.6: Elements contained in mud cake a) without NPs, b) with Fe(OH)3

NPs as per EDX analysis .................................................................70 Figure 4.7: Photos comparing NP-based and original invert emulsion drilling fluids (Invert emulsion (90 vol. oil/10 vol. water); 1 wt% Fe(OH)3 in-situ prepared NPs).......................................................................71 Figure 4.8: Mud cake of drilling fluid with commercial NPs and without NPs ... 73 Figure 4.9: Mud Cakes with thickness of a) DF only, b) DF+LCM, c) DF +LCM with 1 wt % ex-situ NPs, and d) DF+LCM with 1 wt % in-situ NPs ...........................................................................77 Figure 4.10: Mud Cakes of a) DF only, b) DF+LCM, c) DF with 1wt % ex-situ NPs and d) DF with 1wt % in-situ NPs ........................................... 78 Figure 4.11: Filter cakes obtained following API HTHP tests on invert emulsion drilling fluids with and without Fe(OH)3 NPs and Gilsonite LCMs ...82 Figure 4.12: Mud cakes obtained following API HTHP tests on invert emulsion drilling fluids with in-house prepared NPs only ............................... 83 Figure 4.13: Quality of unblended and blended mud cake ................................. 85 Figure 4.14: Rheological behavior of drilling fluid containing a) LCM together with in-house prepared 1 wt% Fe(OH)3 NPs, b) 1 wt% Fe(OH)3 NPs no LCMs. ................................................................................89 Figure 4.15: Gel strength behavior of drilling fluid a) with LCM and NPs together ex-situ and in- situ method b) in the absence of LCM, with NPs only ex-situ and insitu method ........................................91 Figure 4.16: Shelf life of drilling fluid samples in terms of rheology behavior ......92 Figure 4.17: Aging effect of drilling fluid samples in terms of gel strength Behavior ......................................................................................... 92 Figure 4.18: Apparent viscosity at 600 rpm of 3 invert emulsion drilling fluids provided by different supplies in the presence and absence of 1 wt% Fe(OH)3 NPs. Composition of invert emulsion: (90:10) oil: water (v/v) ........................................................................................99 Figure 4.19: X-ray diffraction pattern of ex-situ prepared CaCO3 NPs starting from the aqueous precursor salts .................................................. 104 Figure 4.20 : TEM photographs of ex-situ CaCO3 NPs at two different magnifications ............................................................................. 105 Figure 4.21 : Particle size distributions of ex-situ prepared CaCO3 NPs ........... 105 Figure 4.22 : SEM images of mud cake a&b) without NPs ;c&d) in-situ CaCO3 NPs (R2); and e&f) in-situ CaCO3 NPs (R5) ................... 107 Figure 4.23 : Elements containing mud cake a) without NPs,b) with In-situ NPs (R2) and c) with in-situ NPs (R5) from EDX data ...... 109 Figure 4.24 : Available pore openings (nm) in mud cake of DF without NPs............................................................................................... 110 Figure 4.25: Particle size distribution of in-situ CaCO3 NPs, prepared by reactions (R2) and (R5), in the mud cake .................................... 110

xiv

Figure 4.27: Rheological behavior of invert emulsion drilling in presence and absence of 4 wt% in-house prepared CaCO3 NPs. No LCMs added ........................................................................................... 119 Figure 4.28: Gel strength behavior of invert emulsion drilling fluid in presence and absence of 4 wt% in-house prepared CaCO3 NPs. No LCMs added ........................................................................... 119 Figure 4.29: LPLT fluid loss behavior of different ex-situ NPs in invert emulsion drilling fluid ..................................................................... 123 Figure 4.30: LPLT fluid loss behavior of different in-situ NPs in invert emulsion drilling fluid .................................................................... 123 Figure 4.31: Different NPs and NPs-containing drilling fluid stability Evaluation ..................................................................................... 125 Figure 4.32: NPs-containing drilling fluid filter cakes (thickness <1 mm) ........... 125 Figure 5.1: Filtrate volume variation with square root of time in the presence and absence of in-situ and ex-situ prepared Fe(OH)3 NPs ................................................................................. 131 Figure 5.2: Filtrate volume variation with square root of time in the presence and absence of in-situ and ex-situ prepared CaCO3 NPs ............ 131 Figure 5.3: Mud cake permeability variation with time in the presence and absence of in-situ and ex-situ prepared Fe(OH)3 NPs ................... 132 Figure 5.4: Mud cake permeability variation with time in the presence and absence of in-situ and ex-situ prepared CaCO3 NPs.................... 133 Figure 5.5: Comparison of permeate (filtrate) flux with time in the presence and absence of in-situ and ex-situ prepared Fe(OH)3 NPs ........... 136 Figure 5.6: Comparison of permeate (filtrate) flux with time in the presence and absence of in-situ and ex-situ prepared CaCO3 NPs .............. 136 Figure 5.7: Variation of Mud cake thickness with time for NP-based fluid ....... 140 Figure 5.8: Effect of particle sizes of DF (dp =2-200 μm in DF) on Peclet Number.......................................................................................... 144 Figure 5.9: Effect of Fe(OH)3 and CaCO3 NPs size in DF ranges from 0.001-0.3 μm (1- 300 nm) on Peclet number ................................. 144 Figure 5.10: Bingham Plastic model for Fe(OH)3 NP-based drilling fluid .......... 146 Figure 5.11: Bingham Plastic model for CaCO3 NP-based drilling fluid (R5) and (R2) refer to the reaction used to prepare the particles as detailed in section 3.2 ............................................................. 146

xv

List of Symbols, Abbreviations and Nomenclature

Symbol

Definition

Definition

A cross sectional area of the filter cake under static filtration,cm2

Ar Hamaker constant which has values generally in order of Jouls

NPC Concentration of nanoparticles,molarity

BMD Brownian diffusion coefficient, cm2/sec

NPD diffusion co-efficient of nanoparticles,cm2/sec

Evdw Van der Waals potential energy, Jouls

J total flux, moles/cm2-s

DiffusionJ diffusion flux of NPs, moles/cm2-s

advectionJ advection flux, moles/cm2-s

M Molarity, (mol/L)

N rotor speed (rpm)

Pe Peclet number

Qc volumes of the filter (cm3) cake at a given time, (cm3/min)

Qf volume of filtrate in (cm3) at a given time,(cm3/min)

R radius of the particle,cm

S filter effective surface area = 62.06 cm2

T absolute temperature,K

U characteristic velocity of flow,cm/sec

Vt sedimentation velocity of particles in dilute suspension,cm/sec

V volume of permeate or filtrate in mL

Yp yield point (lbf/100ft2)

ac collector radius or collector characteristics length,cm

ap particle radius,cm

dg diameters of the grains,cm

dp diameters of the particles,cm

dNP nanoparticles diameter,cm

h distance between the particles (nm)

hmc thickness of the mud cake at a given time, cm

k permeability in darcies

Bk Boltzman constant= 1.38X10-23 J/K

xvi

r ratio between the volume of the filter cake at a given time to the volume of

the fluid filtrated in the filter press t time in sec

∆P differential pressure in atmospheres ,(atm)

Shear stress,Pa

Yield stress,Pa

Shear rate ,sec-1

µ dynamic viscosity of the liquid,cP

density of the particles,g/cm3

density of liquid, g/cm3

collision probability

φ porosity of medium

μp plastic viscosity (cP)

viscometer dial reading (o)

ϴ600 dial readings at 600 rpm

ϴ300 dial readings at 300 rpm

Abbreviation

Abbreviation

AADE The American Association of Drilling Engineers

AEUB ALBERTA ENERGY AND UTILITIES BOARD

API American Petroleum Institute

DF Drilling Fluids

EDX Energy dispersive X-ray

HTHP High Temperature and High Pressure

IF-WS2 Inorganic fullerene-like disulfide tungsten

LCMs Lost circulation materials

LC50 Lethal concentration used as an indicator of the toxicity of a compound

LTLP Low Temperature and Low Pressure

NPs Nanoparticles

OBM Oil based mud

SEM

Scanning electron microscopy

TEM

Transmission Electron microscopy

WBM Water based mud

XRD

X-ray diffraction analysis

nm nanometer

1

Chapter One: Introduction

1.1 Problem statement and significance of the research

The success of any well-drilling operation depends on many factors and one of the most

important is the drilling fluid. Drilling fluids, a.k.a. drilling mud, are circulated from the

surface into the drill string and subsequently introduced to the bottom of the borehole as

fluid spray out of drill bit nozzles and back to surface via the annulus between the drill

string and the well hole. Drilling fluids cool down and lubricate the drill bit, remove

cuttings from the hole, prevent formation damage, suspend cuttings and weighting

materials when circulation is stopped, and cake off the permeable formation by

retarding the passage of fluid into the formation (ASME, 2005). However drilling

operations face great technical challenges with drilling fluid loss being the most notable

of them. Drilling fluid loss is defined as the partial or complete loss of fluid during drilling.

Loss of fluid, in turn, impacts the cost of drilling. The cost of the drilling fluid system

often represents one of the single peak capital expenditure in drilling a new well and can

bump up swiftly when drilling deep holes, complex formations or in remote locations

(Abdo and Haneef, 2010). According to a recent in-house estimate, fluid losses during

drilling costs the industry around $800 million per year. Regardless of the real number

of the economic impact in this segment, it represents a very large portion of the total

non-productive expense for drilling a well and therefore fluid loss/circulation loss issues

have intensified than past. Provided that the overall economics prove to be favorable, a

more efficient route needs to be addressed during drilling by eliminating losses of fluid

or at least controlling them to the extent that drilling can continue uninterrupted (Fraser

et al., 2003). Therefore, drilling fluids are typically formulated with loss circulation

materials (LCMs). The primary function of LCM is to plug the zone of loss in the

formation, away from the borehole face so that subsequent operation will not suffer

additional fluids losses. LCM forms a barrier which limits the amount of drilling fluid

penetrating the formation and prevents loss (Chenevert and Sharma, 2009). Most of the

new lost circulation materials have been developed in the past 10 years (McLean et al.,

2

2010). However, using these existing lost circulation materials are not found so effective

to serve their primary goals of curing fluid loss. Current experience shows that it is often

impossible to reduce fluid loss successfully with these micro and macro type fluid loss

additives due to their physio-chemical and mechanical characteristics, e.g. size, surface

charge, solvation and mechanical resistance etc., thus raising the economic

consequences of non-productive drilling time (Chenevert and Sharma, 2009; Fraser et

al., 2003). For example, LCM with diameters in the range of 0.1-100 µm may play an

important role when the cause of fluid loss occurs in 0.1 µm-1 mm porous formation. In

practice, however, the size of pore opening in shales that may cause fluid loss varies in

the range of 10 nm-0.1 µm. Therefore, nanoparticles (NPs) as a loss circulation material

could fulfill the specific requirements by virtue of their size domain, hydrodynamic

properties and interaction potential with the formation (Amanullah et al., 2011; Srivatsa,

2010; Abdo and Haneef, 2010). Alternatively, NPs can help bridging empty gaps

between macro LCMs, and therefore, providing an effective seal to formation with larger

pore throat size. The plugging of pore throats by the use of nanoparticles is a new

approach for controlling fluid penetration into shales and could significantly reduce

wellbore instability problems (Sensoy,2009). Pore space is defined as a collection of

channels through which fluid can flow. The effective width of such a channel varies

along its length. Pore bodies are wide portions and pore openings or pore throats are

the relatively narrow portions that separate these pore bodies (Nimmo,2004).

NPs thus could be a promising option for the development of drilling fluids to provide the

effective sealing, filling and cementing properties resulting in the reduction of porosity,

permeability of the wellbore formations and thereby prevent the loss of fluid. This is not

viewed as formation damage, since these particles can be used during the drilling

operation far from reservoir formation. These particles are ultrafine in nature and

possess very high specific surface area of interactions. By adding small quantities of

NPs in drilling fluid ensuring mixing at the molecular level, wrapping and

interpenetrating network structures to achieve this new class fluid. By forming a thin, low

permeability filter cake which seals pores and other openings in the formations

3

penetrated by the drill bit as shown in Figure 1.1, NP-based drilling fluid could also

prevent unwanted influxes of formation fluids into the borehole from permeable rocks

penetrated during drilling. Kanj et al. (2009) suggested that small particles of high

concentrations might bridge across the pore throat. Again smaller particles aggregate

around larger ones to fill the tinier spaces and hence effectively plug the pore opening

spaces. In water-based drilling fluids, NPs of mixed metal hydroxides (MMH) have

already been used to replace polymers as viscosity modifying agents (Agarwal et al.,

2009). NPs of MMH work as a bridging material, which promotes aggregation between

the platelets of bentonite/montmorillonite clay to form a gel structure. Particle size and

surface characteristics of NPs can also be easily manipulated in water-in-oil emulsions

in a similar fashion to those formed in (w/o) microemulsions (Husein and Nassar,

2007a&b).

Figure 1.1: Schematics showing a) fluid loss control using NPs along with LCM, mechanism of pore throat blocking by b) plugging and sealing, c) bridging and mud

flow restriction by NPs only, and d) fluid loss using conventional LCM.

a) b)

c)

d)

No/ Partial fluid loss using NPs

Mu

d

flo

w

- --

-

-------

-------

Fluid loss without using NPs Mu

d

flo

w

- -----

---

------

------

------

--

------

------

------

--

------

------

------

--

-

-

-

-

-

-

-

-

-

Legend

LCM

Nanoparticles

4

In light of the aforementioned functional properties of NPs, our approach consisted of

developing tailor made NP-based drilling fluid which would best interact with the rest of

the drilling fluid components as well as the formation and reduce fluid loss during

drilling, meanwhile optimize the functionality of the drilling fluid over a wide range of

conditions; including temperature, pressure, drilling environment and formation.

Moreover, the proposed NP-based fluid would reduce the total solids and/or chemical

additives to the typical drilling fluid leading to an overall lower fluid cost (Amanullah et

al., 2011; Abdo and Haneef, 2010; Mokhatab et al., 2006).

1.2 Research Objectives

This research investigates the role of in-house prepared dispersed NPs in reducing

drilling fluid loss and the impact their existence might have on drilling fluid

characteristics; including viscosity, density, pH and lubricity. The hypothesis is that

dispersed NPs in drilling fluids are better able to conceal pores as a result of a fine

balance between particle dispersion and deposition onto micro and nanopores. As

shown in Figure1.1, the NPs will selectively deposit over fine pores or will conceal gaps

between already deposited clay particles. Such research will be the key to unlocking the

problems of inter channel pore clogging of formation (keeps away the migration of

drilled fines entering the pores), reduce fluid loss and improve the productivity of the

wells. In order to meet the challenge of improving the properties of drilling fluid, this

research has been undertaken with several parallel developments of NPs. These well-

dispersed NPs employed in fluid formulation are unique and have high surface energy,

which can readily attach with other additives and create a barrier to lower the fluid loss

in an efficient manner. All this needs to be achieved without introducing fundamental

property change in the drilling fluid. The overall goals of this research are therefore, to

develop a method for in-house preparation of NPs, which can be easily mixed and

stabilized in water as well as invert emulsion based drilling fluids, and to evaluate the

performance of the final product. Low aromatic hydrotreated oil was selected, since

such a base oil makes the invert emulsion fluid more environmentally friendly.

Accordingly, the objectives of this research are summarized as follows:

5

1) In-house synthesis and characterization of NPs and the preparation of stable NP-

based drilling fluids.

2) Study the impact of the presence of the NPs on drilling fluid loss using API low-

temperature-low-pressure (LTLP) test as well as the high-temperature-high-pressure

(HTHP) test. Both oil-based and water-based drilling fluids were tested.

3) Detail the effect of NPs on drilling fluid properties; including viscosity, density, pH

and lubricity.

4) Investigate how the behavior of NPs in the drilling fluid is affected by other

components of the drilling mud.

5) Investigate the possibility of eliminating other loss circulation material (LCM)

additives as a result of NPs addition, which may lead to an overall drilling fluid price

reduction.

The project has been divided into four main phases:

Phase one: In-house, in-situ and ex-situ, preparation of the dispersed Fe(OH)3(s) NPs

in the drilling fluid and their characterization:

1. NPs of Fe(OH)3(s) were successfully prepared in-house. Two schemes were

used. Ex-situ scheme, where the NPs were prepared by aqueous reactions of the

precursor salts and the product NPs mixed with the drilling fluid. In-situ scheme,

where the aqueous precursors were directly added to the drilling fluids, and the

NPs nucleated within the drilling fluid.

2. Characterization of the ex-situ prepared particles, which included particle

identification using X-ray diffraction (XRD), and determination of particle size

distribution using transmission electron microscopy (TEM).

3. Characterization of the in-situ prepared particles followed their collection on the

filter cake; including particle identification using energy-dispersive X-ray

spectroscopy (EDX), and determination of particle size distribution using

scanning electron microscope (SEM).

6

4. Study the fluid loss of NP-based drilling fluid following API low temperature and

low pressure (LTLP) and high temperature and high pressure (HTHP) filter press

protocols. Also, determining the thickness of the resultant filter cake, since thin

filter cake prevents stuck pipe during drilling.

5. Characterize the NP-based drilling fluid in terms of its viscosity, density, pH and

lubricity.

Phase two: In-house, in-situ and ex-situ, preparation of CaCO3(s) NPs in the drilling

fluid and their characterization:

1. NPs of CaCO3(s) were prepared in drilling fluid. Three schemes were used in this

case. Ex-situ scheme, where the NPs were prepared by aqueous reactions of the

precursor salts and the product NPs mixed with the drilling fluid. Two schemes of

in-situ preparation of the CaCO3(s) NPs were adopted. In one scheme the

aqueous precursor salts were directly added to the drilling fluid, and the NPs

nucleated in the drilling fluid, while in the other scheme an aqueous calcium salt

was added to the drilling fluid followed by CO2(g) bubbling. This scheme of in-situ

preparation of the CaCO3(s) particles helps creating the particles in the drilling

fluid while in the formation, and therefore, prevents any changes to the nature of

particles during drilling.

2. Characterization of the ex-situ prepared particles included particle identification

using X-ray diffraction (XRD), and particle size determination using transmission

electron microscopy (TEM).

3. Characterization of the in-situ prepared particles followed after their collection on

the filter cake; including particle identification using energy-dispersive X-ray

spectroscopy (EDX), and determination of particle size distribution using

scanning electron microscope (SEM).

4. Study the fluid loss of NP-based drilling fluid following API low pressure and low

temperature (LTLP) and high pressure and high temperature (HTHP) filter press

protocols. Also, determining the thickness of the resultant filter cake, since thin

filter cake prevents stuck pipe during drilling.

7

5. Characterize the NP-based drilling fluid in terms of its viscosity, density, pH and

lubricity.

Phase three: In-house, in-situ and ex-situ, preparation of BaSO4(s) and FeS(s) NPs in

the drilling fluid by two methods and characterize their LTLP fluid loss property only.

1. NPs of BaSO4(s) and FeS(s) were successfully prepared in-house. Two schemes

were used. Ex-situ scheme, where the NPs were prepared by aqueous reactions

of the precursor salts and the product NPs mixed with the drilling fluid. In-situ

scheme, where the aqueous precursors were directly added to the drilling fluids,

and the NPs nucleated within the drilling fluid.

2. Study only the fluid loss of NP-based drilling fluid following API low pressure and

low temperature (LTLP) protocol.

Phase four: This phase involves developing a mathematical model to describe fluid

loss and cake growth using NPs as a lost circulation material.

1.3 Organization of the Thesis

This thesis is organized into six chapters. The first chapter presents a brief scope of the

research and its significance. General overview of lost circulation material used in

drilling fluid and challenges faced while drilling is introduced. Introduction of NPs as new

lost circulation materials and its potential application in fluid loss reduction are

explained.

Chapter two presents an extensive literature review on drilling fluids, nanoparticles

(NPs), NP-based drilling fluid, filtration mechanism and governing equation used in

filtration process.

In Chapter three, experimental methods used for in-house NPs preparation (ex-situ and

in-situ) is explained together with methods used to characterize the NPs and the NP-

based drilling fluids.

8

Results obtained from the experimental works are discussed and analyzed in detail in

Chapter four. NP-based fluids are compared with the base fluid in terms of fluid loss at

LTLP and HTHP conditions, density, viscosity and lubricity. TEM and XRD analyses

reveal the ex-situ prepared NPs characterization, whereas SEM images of mud cake

unveiled the characteristics of in-situ prepared NPs.

Chapter five deals with modeling of NP-based fluid filtration performance through

porous media (API filter paper) at LTLP condition. The cake thickness growth model at

30-min time period are proposed. Also Bingham plastic model are used to describe the

rheological behavior of NP based fluid.

Chapter six presents the conclusion drawn from the work, original contributions to

knowledge and recommendation for future research to extend this study.

9

Chapter Two: Literature Review

This chapter reviews the drilling fluid general functions and their related challenges, clay

chemistry, nanoparticles properties and previous experimental studies of drilling fluid

properties using lost circulation materials and nanoparticles with particular reference to

those which are directly relevant to the subject under investigation.

2.1 Introduction

Drilling fluids are composed of a number of liquids and gaseous fluids and mixtures of

fluids and solids (Vasii,2008). A drilling fluid is typically used in a drilling operation in

which that fluid is circulated or pumped from the surface, down the drill string and is

subsequently introduced to the bottom of the bore hole as it squirts out of nozzles on

the drill bit and back to the surface via the annulus as shown in Figure 2.1.

Figure 2.1: Drilling mud circulation down the drill pipe (courtesy of Payson Petroleum, reprinted by permission).

10

Large pumps are used to circulate the mud on a drilling rig. They pick up the mud from

the mud tank and force it into and down the drill string and to the bit. Typical pressure at

the exit of these pumps can be as high as 7,500 psi (52,000 kPa) (Dyke and

Baker,1998). At the bit the mud jet out of the bit nozzles to move cuttings away from the

bit. The mud then moves back up the hole to the surface. The mud picks up cutting

made by the bit and carries them as it returns to the surface. The mud and cuttings

return to the surface in the annulus between the outside of the drill string and the inside

hole. At the surface, the mud and cuttings leave the well through a side outlet with a

pipe called the mud return line. At the end of the flow line, mud and cuttings fall on to a

vibrating screen (or sieve) named as shale shakers which is the device on the rig for

removing drilled solids from the mud. A wire-cloth screen vibrates while the drilling fluid

flows on top of it. The liquid phase of the mud and solids smaller than the 200 wire

mesh (< 74 μm) pass through the screen and go back to the pits while larger solids are

retained on the screen and eventually discarded (ASME, 2005; AADE,1999;

Chilingarian and Vorabutr, 1983).

2.2 Drilling fluid Classification

Drilling fluids are typically classified according to their base material into water-based

muds and oil-based muds. In water-based muds (WBM), water is the continuous phase

and solid particles are suspended in water or brine. Oil-based muds (OBM) are exactly

the opposite. Oil is the continuous phase and solid particles are suspended in oil, water

or brine is emulsified in the oil by surfactants (ASME, 2005; Srivatsa, 2010). Oil based

drilling fluids have definite advantages when compared to water based fluids. These

include maintaining stable rheology and filtration control for extended periods of time

and increased lubricity. In addition, oil base drilling fluids can be used to drill through

most troublesome shale formations due to their inherent inhibitive nature and

temperature stability (Mas et al.,1999). The filtrate from a water based mud may cause

clays in the formation to swell and disperse, which can cause severe damage to well

productivity. Many instances are on record where a formation of proved productivity has

been exposed to water or water based mud and consequently production was greatly

11

decreased or in some cases completely lost (Sharma and Jiao, 1992; Tovar et al., 1994;

Argiller et al., 1999). A study has shown that drilling fluid loss costs the oil and gas

industry over $800 million per year (Fraser et al., 2003). An attempt has therefore been

made to develop an invert emulsion drilling fluid (water-in-oil) which would mitigate the

problem. Oil alone does not have the ability to form a filter cake on the wall of the bore

hole but mud additives are used to restrict the loss of fluid into permeable formations.

The filter cake or sheath is water impermeable and substantially oil impermeable so that

virtually none of the fluid base oil or the water in the fluid is lost into the formation. Even

though the filtrate is small amount of oil, fluid which may penetrate the filter cake does

not substantially affect formation permeability (Baker 1995; 2006). Therefore this oil

based system has been directed towards modification by obtaining satisfactory

suspending particles and forming thin filter cake characteristics. These have resulted in

the development of the emulsification of water or water based mud in the oil. The use of

invert emulsion oil mud has greatly increased over the past few years due to the

demands of drilling deeper and more difficult wells.

2.3 Functions of Drilling Fluids

A properly designed and maintained drilling fluid system performs the essential

functions. A drilling fluid is used to carry out the following functions (ASME, 2005;

Chilingarian and Vorabutr, 1983):

a. Removal of Cuttings. Drilled cuttings are removed that results in a cleaner hole.

The ability of a mud to carry cuttings to the surface depends partly on the

characteristics of the mud and partly on the circulating rate in the annulus. When

the pump capacity is too low to provide adequate annular velocity for cuttings

removal, increasing the mud viscosity particularly the yield point may result in a

cleaner hole.

b. Suspension of Cuttings. Good drilling fluids have thixotropic properties that

caused the solids particles, being carried to the surface, to be held in suspension

when circulation is stopped.

12

c. Control Formation Pressure. It is a very important function of drilling fluid

because it is the first line of defense against possible blowouts.

d. Caking off Permeable Formations. A good drilling fluid provides filtration

properties that retard the passage of fluid into the formation. In many cases it

may be necessary to add fluid loss control additives to reduce the fluid loss.

Ideally the muds form a thin tough filter cake across the permeable formations.

This keeps the hole in stable condition. It also minimizes the quantities of mud

and filtrate entering the formation.

e. Cooling and Lubrication. During drilling operations, both the drill string and the bit

develop heat through friction. Drilling mud helps to cool the drill string and also

provides lubrication by reducing friction between drill string and borehole walls.

Thus the lubricity of the mud is important. The cooling function depends upon the

thermal conductivity of the mud.

f. Reduce Formation Damage. Formation damage is very much tied to the filtration

properties of the mud. Damage from filtrate invasion depends on the quantity of

filtrate entering the formation.

g. Minimize Corrosion. In water based mud corrosion is controlled by alkalinity or by

addition of corrosion inhibitors. It has been found that in muds containing oil as

the continuous phase, little or no corrosion occurs.

2.4 Drilling fluid related challenges

Many drilling problems are due to conditions or situations that occur after drilling begins

and for which the drilling fluid was not designed. Zamora et al. (2000) discussed 10

mud-related concerns. Failure to adequately address these concerns can lead to

excessive well costs, unscheduled trouble time, unnecessary high-risk activities, and

poor performance. Some of these problems can be solved by adding materials to the

drilling fluids to adjust their properties. The top 5 mud related problems are found

directly relevant to the subject under investigation and described as follows:

13

a. Borehole instability. Borehole instability is common problems in shale section.

Any formation can collapse if the mud weight is not appropriate to control it. To

minimize its borehole instability, proper mud characteristics (mud viscosity, drag

and torque reduction and fluid loss) are important.

b. Stuck pipe. During drilling oil and gas wells, drill string consisting of pipes and

collars are used to drill the formation. Filtrates invade permeable zones and filter

cakes are deposited on the wall of holes. A portion of the drill string is then

embedded in the mud cake on the walls of the borehole. When the drill string is

no longer free to move up, down or rotate, the drill pipe is supposed to stuck.

This problem is generally caused by the drill pipe sticking to the mud cake on the

wall of the wellbore due to filtrate loss in the wall of the well and the formation of

a thick filter cake or due to the cuttings backing into the wellbore as drilling fluid

circulation is stopped. The pull force to free the pipe is a function to the

differential pressure, co-efficient of friction and the total contact area of the pipe

on the hole wall. The co-efficient of friction (CoF) is one of the important functions

of drilling fluid. An oil-based drilling fluid has co-efficient of friction (CoF) of 0.10

or less (metal to metal) (Chang et al.,2011). In comparison, water has a CoF of

0.34 and the CoF of water-base drilling fluids typically ranges between 0.2 and

0.5 (Chang et al., 2011). It is known that presence of ordinary materials in drilling

mud can cause increased viscosity and mud weight (Dickerson and Rayborn,

1992). This high mud weight can cause damage to sub-surface formations,

plugging of production zones, hole erosion, decreased penetration rate, pipe

failures, stuck pipe and lost circulation (Amoco, 1996; BHI, 1998; Reid et al.,

2000; Njobuenwu and Nna, 2005). So in order to decrease probability of stuck

pipe it is necessary to design new materials which do not increase viscosity and

mud weight (Paiaman and Al-Anazi, 2008). To minimize differential sticking,

maintaining proper mud characteristics (fluid loss, mud density, lubricity, low solid

in mud) is very important.

14

c. Salt section hole enlargement. Salt sections may be eroded by the drilling fluid,

which causes hole enlargement. To avoid this problem, salt saturated mud

system is prepared for drilling through the salt bed.

d. Formation damage. Formation damage is generally a reduction in permeability

near the wellbore with porosity reduction. This represents a positive skin effect.

Almost every field operation is a potential source of damage to well productivity.

Diagnosis of formation damage problems has led to the conclusion that formation

damage is usually associated with either the movement and bridging of fine

solids in the producing horizons and the penetration of drilling fluid particles into

formation which cause pore plugging of the porous media. The fine solids may be

introduced from wellbore fluids or generated in situ by the interaction of invading

fluids with rock minerals or formation fluids. There are a number of ways that

drilling fluid filtrate might interact with the formation to cause permeability

damage. Some of these have been investigated in published papers (Al-Hitti et

al.,2005; Zamura et al.,2000; Chilingarian and Vorabutr,1983). Figure 2.2 shows

the blockage of the reservoir-rock pore spaces caused by the fine solids in the

mud filtrate or solids dislodged by the filtrate within the rock matrix.

Figure 2.2: Drilled fines and fluid particles invasion into the formation

(Zwager, 2007).

15

In addition to pore throat blockage, the formation damage also can be happened due to

clay-particle swelling or dispersion, scale and precipitation, emulsion blockage and

water blockage (Peng, 1990).

e. Lost circulation. Lost circulation means the uncontrolled flow of substantial

amount of drilling mud to an encountered formation. This can be a partial lost,

some returns to surface or a complete loss with no returns to the surface. Lost

circulation can occur in several types of formation, including highly permeable

formations, fractured formations and cavernous zones (Chilingarian and

Vorabutr, 1983). Different lost circulation zones are described in Appendix A.

Lost circulation occurs when hydrostatic pressure of mud exceeds the breaking

strength of the formation and that creates cracks along which the fluid will flow.

Fluid will flow in large fracture greater than 100 microns. In practice, the size of

pore opening of shales that can cause lost circulation is in the range of 10 nm-0.1

microns (Sensoy et al.,2009). Overbalance pressures in excess of about 7000

kPa (1000 psi) are generally considered to be severe and may cause serious

losses of filtrate and associated solids to the formation (Bennion et al.,1997).

Figure 2.3 shows the fluid loss during drilling.

Figure 2.3: Drilling fluid loss into the formation (courtesy of nFluids Inc, reprinted by permission).

16

Lost circulation materials can be added to mud to bridge or deposit a material where

the drilling fluid being lost to the formation. To minimize lost circulation, proper mud

design (usage of suitable lost circulation materials, maintan proper mud weight,

maintain adequate hole cleaning) is necessary. In addition, the appropriate size

distribution of bridging materials to create an effective sealing of impermeable filter

cake which deposits very rapidly on the face of the formation and thereby inhibiting

continual losses of drilled fluid (Al-Hitti et al., 2005).

2.5 Clay Chemistry used in drilling fluids

Clays are naturally occurring materials mainly composed of hydrous aluminum silicates

and typically formed over long periods of time by chemical weathering of rocks that

contain silicate (Deriszadeh, 2012). The colloid chemistry of clays to drilling fluid design

is of value at the present and will continue to be so in the future (Browning and

Perricone,1963). Clays are usually microscopic in size (typically < 2 µm) and also occur

as submicroscopic particles. Pauling (1930) studied the crystalline structure of clays. It

is also pointed out by Grim (1953) that most of the clay minerals have two structural

units that are the building blocks of their atomic lattices. Silica tetrahedron is the first

unit and alumina octahedral coordination which is the second unit as shown in Figure

2.4. These basic sheets are stacked together to form different clay minerals.

Figure 2.4: Basic units of clay minerals and the silica and alumina sheets (Mitchell and Soga, 2005).

17

Clays are the major constituents of shale and due to their special characteristics they

play a crucial role in the mechanical and chemical properties of shale (Santamarina et

al., 2002). Clays are also the major constituent in drilling fluid and provide the fluid a

distinctive character. The properties exhibited by a particular drilling fluid largely depend

on the origin and characteristics of the clay component present in the fluid (Maduka,

2010). Three types of clays are used in drilling fluid formulation: montmorillonite

(smectite), kaolinites and illites. Among them montmorillonite, which is also known as

bentonite, is the most commonly used clay because of its superior ability to swell

uniformly in fresh water upon shear application resulting in a more homogeneous clay-

water mixture (Chilingarian and Vorabutr, 1983). Montmorillonite clay has the formula

[(A)0.3 (Al1.3,Mg0.7) (Si4)O10.(OH)2.xH2O] where A is an exchangeable cation, K+, Na+, or

0.5 Ca2+. Figure 2.5 displays the structure of Montmorillonite clay minerals according to

Grim (1962). According to literature, (Santamarina et al., 2002; Mitchell and Soga,

2005), bentonite has a high specific surface area of 800 m2/g. The diameter of bentonite

platelets could vary between 2000 to 20000 A° with a thickness of about 10 A° bentonite

platelets are bonded together by weak Van der Waals bonds. Therefore it may allow

water to enter the space between the platelets (Deriszadeh, 2012).

Figure 2.5: Schematic representation of Montmorillonite clay (bentonite) structure

(Grim,1962).

18

Browning and Perricone (1963) showed the effect of pH on clay surface. pH values

increment from 10.5 to 11 in the absence of deflocculates rapidly increased the system

resistance to shear with time indicating an increased tendency of surface area of the

clay suspension. The more stable clays system can be introduced by maintaining

proper pH. Chilingarian (1952) found that hydroxyl ion (OH-) adsorb on bentonite clay

platelets increased the total negative charges on the clay sheets causing repulsion of

nearby platelets (dispersed state). Figure 2.6 represents the ion exchange behavior of

clay surface. A double layer occurs when counter ions are attracted to the charged

surface of a particle and form a layer. The ions are held in place by Coulombic forces.

A diffuse layer occurs when the electrostatic attraction between the ions in solution and

the colloidal surface is counteracted by diffusion where the charge on the particle is

neutralized by a swarm of ions. The net negative charge of the clay surface has the

capacity to attract cations or positive charge molecules. With cation exchange may

result in a well dispersed mud system and simply as a result of neutralization of

negative charges on clay platelets. In such cases, plastic viscosity with decreasing

internal friction and gel strength reduction may occur. Highly charged cations can

impact greater attraction of clay platelets allowing lower fluid loss and form a filter cake

having very low permeability. Cations held by clays can be replaced by other cations.

This means they are exchangeable. Hanshaw (1963) showed that the order of cation

exchange selectively is dependent upon whether clay is dispersed or compacted. In

fact, the negative charge on particles is compensated by attraction of cations on the

surface. In the case of bentonite high concentration of cations would also occur inside

the particles as the spacing between platelets of each particle could vary due to this

presence of weak van der Waals forces among the adjacent platelets in each particle. A

small fraction of cations on the surface of particles develop the inner compact layer

commonly referred to as the immobile Stern layer (Mitchell and Soga 2005; Deriszadeh,

2012). In fact, some researchers concentrated on the fluid and ionic flows through micro

pores and the interparticle space of the clays (Mitchell and Greenberg, 1973; Moyne

and Murad, 2002; Smith, 2005; Sherwood, 1994).

19

Figure 2.6: Schematic representation of the fixed and diffused double layer near a clay

surface (Dampier, 2004).

The drilling fluid is a colloidal clay system though its consistency is critical. It must be

fluid enough to be pumped and thick enough to keep the cuttings suspended. In order to

prevent the loss of fluid, it is necessary to minimize the amount of fluid entering the

porous formations. The colloidal clay in drilling mud contributes to the filter cake building

up against the wellbore. When a suspension of a finely divided precipitate is filtered, the

filtration is slow because the particles pack tightly on the filter. The filter cake deposited

by a coarser precipitate of larger particles is less dense and more porous. The

characteristics of the filter cake formed depend on the degree of peptization or

flocculation of the suspension (Zakaria et al.,2012). Stable (peptized) suspensions form

dense, compact sediments while flocculated suspensions form more voluminous

sediments. The filter cake formed from a stable suspension will be dense and relatively

impenetrable in comparison to that formed from a flocculated suspension. Thus a stable

suspension has more effective plastering characteristics (Baker, 2006; Schmidt et al.,

1987). Chemical changes in clay minerals use certain additives in order to modify the

properties of drilling fluid. These additives could result in a safe and speedy drilling with

a maximum productive capacity after completion of a well.

20

2.6 Nanoparticles

Nanoparticles are defined as particulate dispersions or solid particles with a size in the

range of 1-100 nm (Nabhani and Emami,2012; Zakaria et al.,2011). Using nanoparticles

in drilling fluid provide a new era in drilling industry. Amanullah and Al-Tahini (2010)

defined nano fluids as any fluids (drilling fluids, drill-in-fluids, etc.) used in the

exploitation of oil and gas that contain at least one additive with particle size in the

range of 1-100 nm. They also classified nano fluids as simple nano fluids and advanced

nano fluids. Simple nano fluids contain nano particles of only one dimension, whereas

advanced nano fluids are ones with multiple nanosize additives. Commonly used drilling

fluid additives such as bentonite and barite in the conventional drilling fluids have much

larger particle diameters, ranging between 100 nm to more than 100 microns (Srivatsa,

2010; Abrams, 1977; Cai et al., 2012). Figure 2.7 shows a scale of typical particle size

ranges. There have been several methods for the selection of the lost circulation

materials, which are based on the size for the purpose of keeping mud loss at minimum

and given in Appendix B.

Figure 2.7: Particle size scale (adapted from Abrams (1977) and Sensoy et al., (2009)).

In general pore-throat sizes (diameters) are greater than 2 μm in conventional reservoir

rocks, range from about 0.03 to 2 µm in tight-gas sandstones, and range from 0.005 to

0.1 μm in shales (Nelson, 2009). Figure 2.7 shows the pore size connection for

sandstone, tight sand and shales according to Rezaee et al.(2012). Al-Bazali et al.

(2005) also reported average pore throat sizes of variety of shales in the range from 10

to 30 nm. Even though, loss in shale formations is not a big problem, nanoparticles can

21

be ideal additives to minimize fluid seeping into such a sensitive formation (Chenevert

and Sharma, 2009).

Figure 2.8: Pore throat sizes in rocks (adapted from Rezaee et al., (2012)).

According to Smalley and Yakobson (1998), the laws that govern nanoscale material

behave)ior are completely different than the laws governing the macro and micro-scale

behavior. Nanoparticles possess very large surface area per volume as shown in Figure

2.9.

Figure 2.9: Surface area to volume ratio of same volume of materials (Amanullah and

Al-Tahini, 2009).

22

These particles are smaller than micro particles requiring a very low additive

concentration and hence provide superior fluid properties at low concentrations of the

additives (Amanullah and Al-abdullatif, 2010).This causes a very large potential for

interaction with other matter as a function of volume. These enormous surface areas to

volume dramatically increase the interaction of the nanoparticles with the matrix or

surrounding fluid (Monteiro and Quintero,2012). This property of nanoparticles‎ provide

them increased interaction with reactive shale to eliminate shale-drilling mud

interactions and the associated bore hole problems (Amanullah et al., 2011). In addition,

due to large surface area per volume, it is expected that less proportion of nanoparticles‎

need to be employed relative to micron-sized additives conventionally used to achieve a

similar effect. As nano based fluids require small volumes, it significantly decreases

drilling time and increases the productivity index of the drilling activity by increasing the

rate of penetration (ROP). The main application of nanoparticles would be to control the

spurt and fluid loss into the formation and hence control formation damage (Husein et

al.,2012a; Amanullah et al., 2011). The nanoparticles can form a thin and impermeable

mud-cake. Due to its high surface to volume ratio the particles in the mud cake matrix

can easily be removed by traditional cleaning systems during completion stages. Thus,

the nanoparticles can be used as rheology modifiers, fluid loss additives and shale

inhibitors at very small concentrations (Zakaria et al., 2012; Amanullah et al., 2011;

Amanullah and Al-abdullatif, 2010). Research showed that the thermal conductivity and

the convection heat transfer coefficient of the fluid can be largely enhanced by the

suspended nanoparticles (Xuan et al., 2003; Choi et al., 2001). These features make

the nanofluid very attractive in cooling or lubricating application in many industries

including manufacturing, transportation, energy and electronics, etc. Hence, the

enhanced thermal conductivity of drilling fluid will provide efficient cooling of drill bit

leading to an increase in operating life cycle of a drill bit. Micro and macro sized

particles used in drilling fluid accelerate the wear and tear of the surface and subsurface

equipment. Conversely nanoparticles due to its extreme tiny size, the wear and tear of

down hole equipment due to abrasive action is negligible because less kinetic energy

23

impacts by nanoparticles. It is explained by the role of kinetic energy or dynamic action

(sedimentation speed) of nanoparticles on the bit. Particles in suspension in liquid

medium are subjected to three kinds of forces: a) gravitational forces the particles to fall

down, b) viscosity of the liquid decreases the speed of their displacement and c)

Archimedes force is opposed to gravitation forces in this case. By applying fundamental

relation of dynamics, the expression of the steady sedimentation speed of a particle is

can be approximated by Stokes law:

Vt =

( ) (E 2.1)

From the equation E 2.1, it is noticed that the speed varies in proportion to the square of

the radius of the particle. Particles with larger radius will sediment much faster than

smaller ones. Hence, comparing with the larger sized particles (micro or macro

additives) nanosized particles, the kinetic energy will be much less due to its low

sedimentation speed. Therefore Amanullah et al.(2011) reported that nanoparticles

could not harm the downhole tools during the dynamic operation.

Nanoparticles could improve the electrical conductivity of drilling fluids by forming

electrically conductive filter cake that highly improves real time high resolution logs

(Monteiro and Quintero,2012). Due to low requirement of nano additive in mud

formulation, nanobased fluid could be the fluid of choice in conducting the drilling

operation in sensitive environments. The wettability of a formation can be changed by

nanoparticles. The use of nanoparticles to change rock wettability and its subsequent

effect on oil recovery has been reported by several authors (Qinfeng et al., 2010; Ju et

al., 2006). From experimental results, it is expected that some nanoparticles application

in EOR will maximize recovery and boost hydrocarbon production. In a parallel research

to this one, Nwaoji et al.(2013) and Nwaoji (2012) found nanoparticles with LCM blend

bridge the formation, act as an excellent propping and sealing properties of the

fracturing fluid and increased the core fracture breakdown pressure (fbp) resulting in

strengthening wellbores in shale formations.

24

2.6.1 Nanoparticle synthesis

Selection of nanoparticles is dependent on its properties and particle size. There are

different methods for nanoparticles synthesis which are categorized as dry and wet

methods. Dry methods consist of jet and ball milling, micronizer whereas wet synthesis

consist of solvent evaporation, chemical precipitation, spray drying and emulsion

method (Midoux et al., 1999). Husein and Nassar (2008) reported five main techniques

for preparation of nanoparticles; namely: (i) chemical co-precipitation, (ii)

electrochemical, (iii) sonochemical, (iv) sol-gel processing and (v) microemulsions.

Engineered nanoparticles are designed and manufactured with specific properties or

compositions (e.g., shape, size, surface properties, and chemistry). All these techniques

require the presence of a stabilizing agent to prevent aggregation of the resultant

nanoparticles. Among them, water-in-oil, (w/o), microemulsions serve as an excellent

media for the preparation of wide variety of colloidal nanoparticles. (w/o)

Microemulsions typically provide easy control over nanoparticles size and shape and

produce highly homogeneous nanoparticles due to their ability to mix reactants

efficiently at the molecular level (Husein and Nassar, 2008). Nanoparticles get stabilized

in (w/o) microemulsions by means of steric stabilization which is provided by the

adsorbed surfactant molecules on the surface of the nanoparticles (Nassar and Husein,

2007a; 2007b). The surrounding surfactant layer limits their growth and protects them

from aggregation, and hence, maintains their colloidal stability. Stability of colloidal

particles is dictated by the net between the repulsive and the attractive forces which

emerge as the particles approach one another due to Brownian motion and/or other

external forces. When repulsive forces dominate, stable colloidal suspension is

maintained, while net attractive forces lead to particle aggregation and precipitation.

Van der Waals force is an attractive type interaction and is inversely proportional to the

sixth power of the distance between the surfaces of the particles (Husein and Nassar,

2008; Nassar and Husein 2007a,b; Kostansek, 2003). (w/o) Microemulsions are

thermodynamically stable systems and are different in nature than the kinetically stable

invert emulsions typically used in drilling operations. Entropy of dispersion is very

important parameter for the formation of microemulsion systems. Entropy of dispersion

25

contributes to very effective mixing of water pools and, hence very high rate of

intermicellar exchange dynamics compared to invert emulsion systems. This high rate is

indispensable for formation of nanoparticles in (w/o) microemulsions (Husein and

Nassar, 2008). In-situ formation of nanoparticles in invert emulsions, on the other hand,

relies heavily on effective mixing and shearing. This research attempted to adopt (w/o)

microemulsion technique to prepare nanoparticles in invert emulsion drilling fluids with

this fact in mind. Figure 2.10 shows the NPs preparation in an invert emulsion drilling

fluid following chemical co-precipitation method. By adding small quantities of

nanoparticles, or preparing them in-situ, in drilling fluid ensuring mixing at the molecular

level, wrapping and interpenetrating network achieve this new class fluid that could be

used in down hole drilling. The nanoparticles will be tightly held in the water pools,

surrounded by surfactant layers that limit their growth and protect them from

aggregation. A number of investigations were performed using nanoparticles in drilling

fluid to ‎improve the functional characteristics described earlier (Cai et al.,2012; Monteiro

and Quintero,2012; Tour et ‎al.,2011; Manea, 2011; Srivatsa, 2010; Abdo and Haneef,

2010; Chenevert and Sharma, 2009; Sensoy, 2009; Agarwal et al., 2009; Roddy et

al.,2009; Paiaman and Al-Anazi,2008; Sayyadnezad et al.,2008; Jimenez et al.,2003),

but none, had in fact adopted in-situ preparation technique, and most used commercial

nanoparticles.‎

Figure 2.10: Schematic representation of NPs in invert emulsion fluid.

26

2.7 Nanoparticle-based drilling fluids

The spurt loss is considered one of the sources of solid particles and particulate

invasion into the formation. Beeson and Wright (1952) observed that spurt losses

ranging from 2.3 to 7 mL may take place in the formation having permeability in the

range of 7 to 469 md. Muds producing soft and thick cakes increase the potential of

differential sticking and formation damage. This highlights the importance of mud design

to produce clear filtrate with virtually no spurt, low filtrate volume and well-dispersed and

tightly packed thin mud cake. It is often impossible to fulfill certain functional tasks using

conventional macro and micro type mud additives. According to Amanullah and Al-

Tahini (2009), due to the scope of manufacturing of tailored made nanoparticles with

custom made functional behavior, ionic nature, physical shape and sizes and charge

density/volume opened the door to the development of a new generation fluid for drilling

which is expected to play a leading role in overcoming technical challenges associated

with the conventional macro and micro particles based drilling fluid. In order to increase

the penetration rate in deep drilling systems and prevent fluid loss researchers are

working on developing a new nano-particle-based drilling fluid and additives that can

improve the efficiency, extend the life of drilling fluids, control fluid loss and less

susceptible to degradation under high temperature and pressure (HTHP) operations.

Recent experiments have demonstrated that nano fluids have attractive properties for

applications in heat transfer, drag reduction, binding ability for sand consolidation, gel

formation, wettability alteration, and corrosive control (Mokhatab, 2006; Krishnamoorti,

2006).

Amanullah et al. (2011) disclosed a WBM with less than 1 wt% NPs, resulting in no mud

spurt loss. High potential for reducing differential pressure sticking problems while

drilling, reduce torque and drag problems in deviated, horizontal extended reach and

multi-lateral drilling operations. Tiny concentration of less than 1% w/w of nanomaterial

plays an important role in increasing rate of penetration. But more interestingly,

Amanullah et al. (2011) formulated their “nano-based drilling fluids” by mixing

nanoparticles with the base fluid, i.e. water. They did not use a real drilling fluid

27

formulated by industry. Therefore, they were in need for very active stabilizers to

maintain the nanoparticles dispersed. The polymeric viscosifier and surfactant additives

used are costly, i.e. not practical. Moreover, industrial drilling fluids contain many

additives that may compromise the stability of their drilling fluids and render them

ineffective. For all practical purposes their “drilling fluids” are model drilling fluids which

have no industrial applications. Also, looking at the fluid loss experimental results, it is

clear that fluid loss reduction for a period of 30 min was not improved at all or became

negative compared to the bentonite based mud (without nanoparticles as a control

sample). Calcium‎carbonate is a well-known weighting material in drilling fluid. Reducing

solid, i.e. clay, content by introducing CaCO3 is not a new idea. Their claim regarding

increasing ROP, decreasing formation damage and decreasing the coefficient of friction

are indirectly related to NPs addition. They only improved the rheological behavior of

fluid in terms of stability and gel strength. Srivatsa and Ziaja (2012) disclosed a WBM

with viscoelastic surfactant and 10 wt % NPs. It also tests higher amounts of 20 wt%

and 30 wt% NPs. 10 wt% is considered as the minimum concentration needed for fluid

loss and address differential sticking problems. The authors used model “drilling fluids”

formulated by adding different proportions of surfactants, polymers and/or

nanoparticles. No actual nano-based drilling mud was used. A large amount of

surfactant and polymers were used in fluid formulation. With the addition of NPs at

different concentration resulted at 20-40 % fluid loss reduction. Due to the large amount

of NPs with surfactant-polymer blend could make the drilling fluid practically undesirable

in terms of drilling cost and other functional activities. Water based mud in real

application do not mix with large amount of surfactant and polymers. Polymer-surfactant

blend suppress the NPs fluid mud cake with a desirable thickness so that differential

sticking problem can be eliminated. No experimental results and mechanism proved the

lubricity nature of their commercial silica NPs used in their works. Aston et al. (2002)

disclosed NPs at a concentration of 0.7 to 1.4 wt% and discussed preventing differential

sticking and formation damage avoidance as well as fluid loss reduction. Three

components were required for fluid loss control – emulsified brine, fine solids, and fluid

loss control chemical such as Gilsonite, asphalt or synthetic polymer. High

28

concentrations of fluid loss additives were not required since they did not improve fluid

loss control. The study focused on the effect of every component of an OBM on fluid

loss, and in fact formulated the fluids using the base oil and added one component at a

time. Once the solid particles were tested, they were not in suspension, Illite provided

loss prevention but it was in the micro-domain.

The use of engineered nanoparticles increases the intra-granular strength and reduces

permeability and porosity of formation. Oil-based muds offer a good solution to shale

instability problems. Nevertheless, development of water-based mud is also needed for

environmentally sensitive areas where nanoparticles might be effective in plugging pore

throat openings and stabilizing the wellbore. Sensoy (2009) disclosed a WBM with

nanoparticles. It states that 5 wt% NPs is not as effective as 10 wt% which is

considered the minimum needed for fluid loss. Reduced fluid penetration into Atoka

shale up to 98% compared to sea water. NPs were between 5 to 20 nm in size. No

actual nano-based drilling fluids were used. The author used dispersion of nanoparticles

in water. Permeability reduction was taken longer time with higher amount of silica NPs.

Extra additives were required to disperse the silica NPs in drilling fluid. Tests were not

covered for invert emulsion mud and HTHP conditions. The application of this fluid

pertains to nanopore throat reduction rather than considering the overall fluid loss. We

anticipate that this approach could damage the formation by forming internal plugging

into the formation and in this manner could significantly increase wellbore instability

problems. As it took longer time to reduce permeability means the penetration of drilling

fluid particles or NPs passes from the hole into the formation. Therefore we can

conclude that spurt loss is higher in this case before pore plugging occurred. Similarly

Chenevert and Sharma (2009) investigated permeability reduction of shale formations

using specific nanoparticles in the water based drilling fluids. By identifying the pore

throat radii of shale samples, the investigators were able to select fine particles that

would fit into the pore throats during the drilling process and create a non-permeable

shale surface. They formulated their water based mud with silica, iron, aluminum,

titanium or other metal oxides and hydroxides nanoparticles having size range of 1-500

29

nm and also composed of a surface active agent (alkyl amines, alkyl sulphates, alkyl

sulphates containing aromatic rings, Polyethylene glycol (PEG), Polypropylene glycol

(PPG) etc) . The minimum concentration required to reduce the fluid penetration was 10

wt% NPs. It was observed that addition of 10 wt% of silica NPs reduced fluid

penetration by 72% in 36 h for Atoka Shale and 50% in 23 h for GOM Shale. It was also

observed that high concentration of nanoparticles (41 wt%) completely plugged the pore

throats by 2 h for Atoka Shale. Thus, these fluids based on nano fluids can be used

effectively in horizontal and directional shale drilling as nanoparticles can easily

penetrate into the shale and hence drastically reduce the shale-drilling mud interactions

and stabilize the wellbore. It was also found by Huang and Crews (2008) that

nanocrystals with hydraulic fracture proppant reduced fines migration without disturbing

productivity. The NPs asscociate with VES (viscoelastic surfactant) micelles through

surface adsorption and surface-charge attraction to stabilize fluid viscosity at high

temperatures and produced a pseudofilter cake of viscous VES fluid on porous media

that reduced the rate of fluid loss significantly and improved fluid efficiency for hydraulic

fracturing.

Ying,(2012) disclosed the use of precipitated sub-micron barite as a weighting agent

in ‎drilling fluid. The precipitated barite showed less sag than conventional weighting

agents which led to a ‎decrease in pipe sticking. The precipitated barite was used in

amounts from 20-99.9% by volume and ‎did not lead to an unwanted increase in

viscosity. They also reported the use of polymers, which might bridge the particles and

produce thin mud cake as per literature. Thin mud cake leads to a ‎decrease in pipe

sticking. So, no clear distinction between the roles of the submicron particles or the

polymer in reducing pipe sticking problems was made in Ying,(2012) work.‎ Ballard and

Massam (2012) investigated precipitated submicron barite as a weighting agent in a

drilling fluid. The ‎precipitated weighting agents showed less sag than conventional

weighting agents. Preferred ‎precipitated agents are calcium carbonate, barium sulfate,

iron oxide, magnesium carbonate and a wide ‎variety of others. The precipitated

weighting agents have a particles sizes varying between 20 and 90 nm. They also

30

disclosed that precipitated barium sulfate can be made by mixing barium chloride

and ‎sodium sulfate. Their works mainly focused on optimizing the viscosity of a drilling

fluid and avoiding high viscosities while achieving high densities, an attribute especially

needed for high pressure drilling. Further, unlike the methods developed in this study,

the precipitated weighting agent is prepared ex-situ, where a coating material

(dispersant) ‎surrounds it, and is then added as a solid powder to the drilling fluid.‎

Paiaman and Al-Anazi (2008) suggested to reduce the thickness of mud cake utilizing

carbon black nanoparticles in drilling fluid. Carbon black NPs having specific Gravity=

1.9-2.1, initial diameter about 30 nm which after aggregation increased to 150-500 nm

was used in the works. The presence of carbon black particles reduced the thickness of

the mud cakes and increased the thermal stability up to 3000 °F (1649 °C). Results

have shown that adding 2 % by volume of carbon black to water based drilling mud

decreased mud cake thickness, mud viscosity and yield point which led to less

permeability and stuck pipe problems. Results also showed that thickness reduction

was found better at high temperature and pressure.

Griffo and Keshavan (2007) disclosed a drilling bit grease that comprises from 0.1-10

wt% at least one nanomaterial. The ‎grease is comprised of a common base stock such

as synthetic oil, petroleum oil, and mineral oil or ‎a combination thereof. Soaps, urea,

fine silica, fine clays and silica gel may be used as thickeners. ‎Preferred lubricating

nanoparticles include molybdenum disulfide, graphite, carbon black, lead oxide, ‎zinc

nanoparticles ranges from 0.5-50 nm.‎ In their works it was found that addition of

thickeners were necessary along with NPs. Although the invention did not claim that

lubricity was based on NPs only; most of the metals considered in their works were

heavy metals, which have a big environmental impact (e.g. lead). Uses of nanoparticles

in drilling fluid will also expand its area of application in fracturing fluid additive, cement

slurry additives, completion fluid additive. Nanoparticles increase the mechanical

strength of the fluid. Roddy et al. (2009) observed that addition of nano-silica having a

particle size in the range of about 1 nm to about 100 nm and present in an amount in

31

the range of about 1% to about 25% by weight to the cement slurry (comprised of

cement, water) reduced the cement setting time and increased the mechanical strength

of the resulting cement explains the fact that light nanomaterials can take overall load.

Some nanoparticles may carry magnetic property which can change the density of fluid

when necessary. Jimenez et al. (2003) made an effort to prepare nanoparticles treated

drilling fluid which was responsive to the change density state required to control

subsurface pressures, preserve and protect the drilled hole. The nanoparticles were

sized between 0.5 and 200 nm and formed into clusters having average size of between

0.1 µm and 500 µm. The clusters were formed by incorporating the nanoparticles into a

matrix of glass or ceramic. Group VIII metals Cd, Au and their alloys were found to

provide an excellent result in adjusting fluid density in a reversible manner. They have

shown that 90% of the supermagnetic nanoparticles from the treated drilling fluid from

the downhole location again recovered to a magnetic field at the surface resulting in the

adjustment of drilling fluid density within a short period of time and circulated the

magnetic nanoparticles to the surface level for reusing them in the drilling fluid. From

the study it is shown that viscosity could also change with the addition of nanoparticles.

Javora and Qu (2009) used an aqueous based well treatment fluid containing an

additive having a median particle size of the calcium carbonate nanoparticles less than

or equal to 1 µm as viscosifying additive. The amount of calcium carbonate

nanoparticles used in drilling fluid was approximately 20 wt%. The nanoparticles used in

well treatment fluid were capable of being suspended in the fluid without the aid of a

polymeric viscosifying agent. Nanoparticles suspended in a well treatment fluid even at

high temperature, e.g. 350 °F, typically exhibit sag no greater than about 8%. It was

observed from the study that addition of nanoparticles altered the viscosity of the fluid.

Using nanoparticles, Huang and Crews (2008) reduced the leak off viscoelastic

surfactant simulation fluids at high temperature for completion applications. They also

discovered that, micellar fluids such as surfactants can have wall building

characteristics when small concentrations of nanoparticles are added. The

nanoparticles pseudo-crosslink the elongated micelles in a manner similar to cross-

linking Polymers. They further investigated the pseudo-crosslink characteristics of the

32

worm like micelles in surfactant systems and concluded that nanoparticles first

associate with end caps which are energetically unfavorable and then this becomes

junctions for worm like micelles, which enhance the wall building characteristics of the

fluid system, improve thermal stability and also improve the viscosity of the fluids. Berret

(2004) investigated the interaction of nanoparticles with co-polymers and observed the

formation of “super-micellar” aggregates. Nanoparticles which have a hydroxyl group

(-OH) on the surface and causes nanoparticles to be agglomerated. This agglomeration

causes poor dispersions and addition of surfactants reduces this problem. Asphaltic

materials are also used as additive to bind the metal oxides at high temperature tend to

decrease the fluid loss. McGlothlin and Baggett (1972) invented an invert emulsion

drilling fluid employing manganese oxide with asphalt constituents. Addition of MnO2 the

fluid loss reduction was approximately 66 % than the control sample at 300°F with

substantially no breakdown of the emulsion. The amount of metal oxide or oxides

employed were from about 1 to about 10 wt%. Miller (1971) improved plastering

properties and reduced fluid loss properties at extreme conditions of borehole

temperature and pressure using asphalt material as a filler or plaster at high

temperature. He formulated the oil based drilling fluid containing a small amount of a

secondary weighting material inert to the fluid and having particle size of no more than 3

µm. Suitable inert materials for the secondary weight phase were the iron oxides and

titanium oxides. Each sample was tested for fluid loss by maintaining the fluid in a high-

temperature, high pressure filter press at 300°F and 500 psi for 30 min. The

investigations showed that iron oxides having size 3 μm had 12 % less fluid loss, TiO2

particles having 0.18 μm had 36 % and 0.19 μm showed 12 % less fluid loss than the

control samples. The investigation again showed that addition of TiO2 with fine barium

sulphate lower the fluid loss 38% than the control samples. Ravi et al. (2011) made an

effort to introduce a lost circulation composition into a lost circulation zone to reduce the

loss of fluid into the formation. The lost circulation composition comprised of portland

cement in an amount of about 10% to about 20 % by weight of the lost circulation

composition, nano-silica in an amount of about 0.5 % to about 4 % by weight of the lost

circulation composition, the nano-silica having a particle size of about 1 nm to about 100

33

nm, amorphous silica in an amount of about 5 % to about 10 % by weight of the lost

circulation composition, synthetic clay in an amount of about 0.5 % to about 2 % by

weight of the lost circulation composition, sub-micron sized calcium carbonate in an

amount of about 15 % to about 50 % by weight of the hydraulic cement and water in an

amount of about 60 % to about 75 % by weight of the lost circulation composition. It was

investigated that lost circulation compositions rapidly developed static gel strength and

remained pumpable for at least about 1 day. The sample was observed to gel while

static but returned to liquid upon application of shear. Thus mixed nanocomponents

with cement could reduce the setting time for mud cake formation and development of

gel strength. The retention of nanoparticles from concentrated dispersions in

sedimentary rocks has recently been investigated by Rodriguez et al. (2009). They

mentioned that NPs retention by the porous medium had consistent competition

between adsorption (vander Waals attraction between nanoparticles and solid surface)

and desorption (Brownian motion and hydrodynamic drag). Tour et al. (2011) used

drilling fluid including chemically converted nanoplatelet graphenes with functional

groups. The graphene comprised about 0.001% to about 10 vol.% . The functionalized

chemically-converted graphene sheets were about 1.8 nm to about 2.2 nm in thickness.

Whatman 50 allowed some graphene oxide to pass through the filter. Filtration rates

varied from 0.10 mL/min to 0.28 mL/min for graphene oxide and chemically converted

graphene solution on whatman 50 filter paper. Surface charge plays an important role

on the transport of nanoparticles and trapped in porous media. Poulton and Raisweel

(2005) reported that the natural spherical iron oxides nanoparticles (10-20 nm) in

sediments tend to aggregate at the edges of clay grains due to their surface charge

characteristics. Kosynkin et al. (2012) showed that using graphene oxide (GO) with a

concentration of 0.2 % (w/w) by carbon content exhibited 15.27 % fluid loss reduction in

water based drilling fluid compare to the water based control drilling fluid sample. GO

preparation technique was not user friendly. During spurt loss, NPs pass through the

filter cake which could damage the formation due to internal pore blockage by their NPs.

Saboori et al. (2012) also investigated mud cake thickness and water loss using CMC

nanoparticles having particle size distribution 27 nm to 930 nm with average size of 47

34

nm in water based mud. NPs were prepared by ball milling method. It was found that

existance of naoparticles caused the amounts of water loss and mud cake thickness to

decrease. Compare with the micron-sized CMC (10 gm) used in control samples, nano

CMC (10 gm) can decrease only 7 % fluid loss. Manea (2011) investigated nanoscale

polymer additives in water based mud to measure the fluid loss. 2.5 wt% nano Xanthan

gum and 3 wt% nanopolymer together in mud reduced 29 % water loss whereas mud

treated with NaOH and addition of 2.5 wt% nano Xanthan gum and 3 wt% nanopolymer

together reduced fluid loss by 48% led to the conclusion that in an alkali medium, the

efficiency of Xanthan gum was enhanced. On the other hand, increasing the sodium

hydoxide concentration in the mud also increase the pH of the mud which could lead to

the undesirable mud viscosity reported by Irawan et al. (2010). Deep water drilling has

been emerging as an important drilling activity to all oil and gas companies in order to

increase the daily production of crude oil. Drilling in deep water wells normally

associated with high temperature high pressure (HTHP) condition. HTHP wells are

generally considered to be those which encounter bottom hole temperatures in excess

of 350 °F (177 °C) and more than 500 Psi pressure. A research program initiated by

Tran et al. (2007) to develop nano particles based drilling fluids to perform in high

temperature drilling. Under their study, they assumed the benefits of using drag-

reducing polymer additives with nanoparticles could improve the drilling penetration

rate, lubrication, and cooling the drill bit.

Hydrogen sulphide which is corrosive, toxic and dangerous gas largely produced in gas

and petroleum industries. It can diffuse into the drilling fluid from formations during oil

and gas wells drilling. Investigations are being carried out to remove hydrogen sulphide

from this drilling fluid to reduce the environmental pollution, protect the health of drilling

personnel and prevent corrosion of pipelines and equipments. Sayyadnejad et al.

(2008) showed that utilizing zinc oxide (ZnO) nanoparticles synthesized by spray

pyrolysis method in water based drilling mud removed hydrogen sulphide completely

within 15 min where as bulk zinc oxide removed 2.5% of hydrogen sulphide in as long

as 90 min under the same operating conditions. The results obtained in this research

35

showed that ZnO nanoparticles in the range about 14-25 nm size, 44-56 m2/g specific

surface area used as an effective scavenger for removing hydrogen sulphide (H2S) from

drilling mud.

Since HTHP/environmentally sensitive/remote areas drilling are inherently

expensive, the choice of drilling fluids and its continuous phase require careful

evaluation to successfully handle the environmental challenges. Drilling fluids

formulated with low aromatic oil (contain less than 1% of aromatic fractions) are 5 to 14

times less toxic than diesel based fluid under the EPA criteria. LC50 currently has a limit

of 30,000 ppm for the suspended particulate phase of the whole mud (Sáchez et al.,

1999). Research concerning the toxicity of NPs is still in its infant stages. However a

study was investigated by García et al. (2011) showed that iron oxide NPs exhibit low or

no toxicity at low concentration (0.67 mg/mL) and reported LC50 = 2.3 x 10-4 mg/mL

which would be special interest to use it in drilling fluid.

The cost of nanobased drilling fluid may make them economically feasible. Srivatsa

(2010) reported the cost of a typical oil based drilling fluid containing silica NPs was

costly. A comparative cost between commercial silica NPs used by Srivatsa (2010) and

NPs prepared in ex-situ/in-situ method by Husein et al. (2012) shown in the following

table:

Table 2.1: Comparative cost analysis study of NP-based drilling mud (Srivatsa, 2010).

Component Volume Cost/unit

($) Cost/component

($)

Diesel Oil 0.8 bbl 42 33.6

Emulsifier 6 lbs 1.5 9.0

Water 0.14 bbl NA NA

Gel 5 lbs 1.20 6.0

Calcium Chloride 20 lbs 0.20 4.0

Lime 3 lbs 0.10 0.30

Total Cost (1bbl) 52.9

NP (silica) at 10 wt% conc. Srivatsa (2010) 35 lbs 2.5 87.5

In-situ/ex-situ NPs 1 wt% Fe(OH)3 conc. prepared by Husein et al. (2012a)

4.8 lbs NA 12.78

36

From the above table, total cost of the 1bbl diesel based mud is $52.90. Addition of

commercial NPs (10 wt%) contributed to $140.4/bbl, whereas in-situ/ex-situ NPs based

mud overall cost would be $65.71/bbl. So based on the literature in-situ/ex-situ NPs

costs are less expensive and more cost effective in terms of drilling cost per well. In

respect to filtration control of drilling fluid suggests that there should be a proper

criterion of particle size distribution. Experimental works presented by different

researchers in literature showed the single size of NPs used in mud formulation and

reached hardly near to the objectives. Whereas we suggest the cost effective in house

ex-situ and in-situ prepared NPs having wide range of particle size distribution could be

successful to handle the different permeabilities and porosities of the formation.

Nonetheless we can conclude that the addition of NPs to drilling fluid has a positive

effect on fluid properties. In this work we evaluate fluid loss performance along with

other characteristics of drilling fluid after blend of custom prepared engineered NPs in

drilling fluid. From earlier literature review a wide range of NPs are preferably selected

from metal hydroxides, e.g. iron hydroxide, metal carbonates, e.g. calcium carbonate

and metal sulfate and sulfide e.g barium sulphate and ferrous sulfide respectively.

2.8 General characteristics of drilling fluid filtration

The invasion of filtrate occurs once filtration starts during drilling. It is well known that

there is a mud spurt at the start of a filtration process before filtration begins. In the

drilling well, mud spurt may be much larger when filtration takes place against the more

permeable rocks. In fact they can be infinite (i.e. circulation is lost) unless the mud

contains particles of the size required 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 size can bridge these pores. Particles lager than the pore openings cannot enter

the pore, and are swept away by the mud stream. Particles considerably smaller than

the opening, on the other hand, invade the formation unhindered. Intermediate particles

of a certain critical size stick at bottle-necks 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 filtrate invades the

37

formation. The mud spurt period is very brief, a matter of a second or two. Krueger

(1963) produced a typical curve for cumulative filtration volume versus time as shown in

Figure 2.11. Equilibrium conditions are reached after about 6 to 10 h and fluid loss rate

becomes constant as indicated by the straight line portion of the curve. The first portion

of the curve is the dynamic fluid loss rate through equilibrium dynamic mud cake laid

down on newly exposed formation. The second portion of the curve represents the

duration when circulation was stopped and the static mud cake allowed to build up on

top of the dynamic cake. The third portion of the curve represents the period when the

circulation is resumed and equilibrium dynamic fluid loss rate is established. During T0

and T1 time period, deposition of particles occurs and is characterized by filtration

through a cake of constant thickness. Outmans (1963) suggested that during this

period, the rates of particle deposition and cake compaction must be equal. During T2

and T3, filtration rate, cake thickness and permeability are constant. The next phase of

the curve corresponds to the period of static filtration (T3 to T4) which occurs when

circulation is stopped. During this time filter cake thickness will start to increase. This

static filtrate affects the subsequent drilling operation. Therefore it is necessary to

predict the static fluid loss in which the filter cake forms upon a previously deposited

dynamic filter cake. It is logically thought that the lower the mud cake permeability, the

thinner the mud cake and the lesser the volume of filtrate from muds. The primarily

concern is to control the static filtration rate/loss in order to overcome drilling and

completion difficulties associated with filter cake growth to great thickness during long

period of static filtration, which is often the case during swabbing, fishing and trips.

Thick filter cakes restrict the easy passage of downhole tools and allow excessive

amount of filtrate to move into the formation, creating a potential cause of caving and a

long term formation damage problem as a result of fluid invasion (Maduka, 2009). In the

third phase (T4 to T6) of Figure 2.11, a new dynamic equilibrium filtration rate is achieved

and total resistance to filtration increases due to cake depositing in the static phase.

This theory was proposed by Ferguson and Klotz (1954) and also confirmed by Peng

(1990).

38

Figure 2.11: A characteristic filtration plot of drilling fluid during drilling (Krueger, 1963).

Darley (1965) reported that initial filtration rates depended on the concentration of solids

and particle size distribution in the mud. It was suggested that there was a critical size

range for bridging in the surface pores. Once the pores are bridged by appropriate size

particles, successively smaller ones are trapped, and a filter cake is established.

Schremp and Johnson (1952) described the drilling mud filtration process into two

steps: (1) bridging of openings in the filter medium, and (2) filtration of fluid throughout

the filter cake that developed on the filter medium as the filtration continues. Gates and

Bowie (1942) discussed the relationship between particles size distribution and filtration

properties. They showed that the best filtration control properties of the muds were

composed of approximately 65 wt% colloids, 30 wt% silt and 5 wt% sand, whereas the

poorest filtration control muds were composed of 1 wt% colloids, 94 wt% silt and 5 wt%

sand. And increasing temperature increased the filtration rates of the fluids tested.

Barkman and Davidson (1972) showed three characteristic shapes of the filtration curve

as shown in Figure 2.12.

39

Figure 2.12: Three Types of Filtration Curves (Barkman and Davidson, 1972).

They concluded that the shape of the curves depended on the sizes of the suspended

solids and filter medium. Figure 2.12a results when the suspended particles are larger

than the pores of the filter medium and no invasion takes place. The intercept of the

straight line at time zero becomes negative. Figure 2.12b occurs when the suspended

solids are much smaller than the pores of the medium and invasion takes place at the

early part of the experiment. Therefore a positive intercepts is apparent. In Figure 2.12c

S-shaped curve, which is not common, occurs when several filtration mechanisms

(plugging mechanisms, pore throat blockage and pore filling) operate simultaneously. It

is evident that filtration curves mostly depend on the particle/pore-throat size ratio,

filtration velocity and mechanism of capture of particles (Pang and Sharma,1997).

Jones and Babson (1935) investigated the filtration properties in artificial formation

prepared from the unconsolidated sand at pressures upto 4000 psi and temperatures

upto 275°F under dynamic conditions. They found that filtration flow rates attained a

constant value at the end of about 2 h. Regardless of the nature of mud or temperature,

40

variation in pressures above 500 psi had very little effect on the filtrate flow rate or on

the mud cake thickness. It was found that the fluid loss increased rapidly with reduced

apparent viscosity for weak thixotropic mud but unaffected for strong or moderately

strong thixotropic muds. Byck (1940) investigated the effect of formation permeability on

the filtration characteristics of drilling fluids. It was shown that the filtration rate was

dependent on the permeability of the cake as long as it was several orders of magnitude

lower than the permeability of the formation. Willams and Cannon (1938) performed

static filtration tests at pressures ranging from 30-1500 psi for a wide range of drilling

fluid and concluded that cake resistance increased by the addition of bentonite and rate

of filtration could be varied by adding weighting materials. Larsen (1938) reported the

following relationships drilling fluid filtration experiments at pressures ranging from 25-

8000 psi and temperatures in the range of 60-250°F.

Fluid loss α t1/2 (E 2.2)

Fluid loss α

(E 2.3)

Fluid loss α Cake thickness (E 2.4)

It was also found that fluid loss increased by calcium ion flocculation of the mud.

Flocculation of muds causes the particles to associate in the form of a loose, open

network causing considerable increase in permeability. Conversely, deflocculation of a

mud by the addition of a thinning agent causes a decrease in cake permeability.

Moreover, most thinners are sodium salts, and at high concentration, the sodium ion

may displace the polyvalent cations in the base exchange positions on the clay, thereby

dispersing the clay aggregates, and further reducing cake permeability. Thus, the

electrochemical conditions prevailing in a mud are a major factor in determining the

permeability of its filter cake (Maduka, 2010).

Fordham et al. (1988) proposed two fundamental models for dynamic filtration of a

drilling fluid. One of them is convection diffusion balance model based on the fluid loss

control between the convective transport of mud particles towards the filtration surface

41

by the filtration flux and the diffusion of particles away from the surface. Another one is

particle adhesion model which suggests that particle permanently stick at the cake

surface depends on both the filtration rate and the colloidal interactions between

particles. Once stuck, particles are unable to migrate away from the surface. Milligan

and Weintritt (1961) reported the effects of elevated temperature on filtration of drilling

fluids. They concluded that at an elevated temperature, muds might undergo irreversible

reactions or changes in composition. The increase in temperature reduces the viscosity

of the filtrate, and therefore, cumulative filtrate volume increases proportionally. Many

organic filtration control agents start to degrade significantly at temperatures above

100°C. Chemical degradation of one or more components used in mud can affect filtrate

properties. Electrochemical equilibria which govern the degree of flocculation and

aggregation, altering the permeability of the filter cake due to an elevated temperature

(Maduka, 2010).

The static filtration equation for mud filtrate through a mud cake is described by Darcy’s

law. In 1856, Henry Darcy published the following equation to describe the flow of fluid

through porous medium (Chelton, 1967):

mc

f

h

PkA

dt

dQ

(E 2.5)

Larsen (1938) found that if a mud was filtered through paper at constant temperature

and pressure, fluid loss Qf was proportional to square root of time √t . Although this

finding is not strictly true for all muds, but close enough for practical purposes. Carman

(1938) extensively studied the cake filtration and concluded that Darcy’s law is

applicable for mud filtration. It is shown that drilling fluid loss behavior with time has

been estimated from the experimentally derived data by using Darcy law in the literature

(Maduka, 2010; Kumar, 2010; Hoff et al., 2005).

42

In current study, API static filtration equation is described at Chapter 5 as a part of

sequential filtration for realistic drilling condition for predicting the drilling fluid loss

control by nanoparticles.

2.9 Filtration mechanism

Particles in suspension in a liquid medium are subjected to three kinds of forces: a)

gravitation forces, b) drag forces and c) Archimedes force (Martin and Ohmae, 2008).

Filtration occurs when fluid-containing particles are captured flowing through a porous

medium. Particles are deposited due to the different mechanisms as described by

Zamani (2010); Farajzadeh (2004) and deZwart (2007).

Particles transport by interception. This particle transport mechanism becomes

important when particles are larger in size. Interception occurs when particles following

a streamline hits the surface of a grain and attach to it. Particles having equal density of

the fluid follow the streamline in porous media at low velocities. When the particle is

retained by a previously deposited particle, it is referred to as bridging. Figure 2.13

shows the bridging of the particles in a pore throat with varying particles diameter. It

shows the bridging effect that occurs when two or more particles arrive at the same

moment to pass through or when one particle is already attached to the grain and

another particle wants to pass through.

Figure 2.13: Bridging effects with varying particles diameter in pore throat (de Zwart (2007), Farajzadeh (2004)).

43

Happel (1958) used the following relationship to calculate the probability of collision.

=

(E2.6)

As is the porosity dependent parameter and express as

=

(E2.7)

where, z= 2-3p+3p5-2p6 and p = (1-φ) 1/3 ; where φ is the porosity of medium.

Particles transport by Impaction. When the density of particle is larger than that of the

fluid, inertia deviate the approaching particles from the stream line and attach them to

the surface of a grain. This mechanism is responsible for collecting larger particles. The

inertial effect is characterized by the dimensionless Stokes number (Tien,1989 ; Ives,

1970) as

Nst=

(E 2.8)

Particles transport by Sedimentation. When the density of particles is different than the

density of the fluid, the fluid velocity will be different than the particles velocity. The

expression of steady state sedimentation velocity of particles in dilute suspension, V t

are given in E 2.1. Particles with large radius will sediment much faster than the small

ones and less kinetic energy will be impacted by small particles.

Particles transport by Diffusion. Small particles experience to random Brownian motion

that increases the collision frequency between particles and grains. The collision

probability due to diffusion is equal to

= 0.9 (

)

(E2.9)

44

Diffusion is important for small particles (dp<1 μm) and is usually neglected for larger

particles. Ives (1970) found that the Brownian motion is dominant in transporting the

submicron size particles but for the particles with greater than 1 μm in diameter, the

viscous drag of the fluid restricts this movement and the mean free path of the particle is

at most one of two particle diameter and therefore the mechanism is neglected. At a

short distance, below 50 nm, two particles attract themselves due to van der Waals

forces and it is expressed by the following relation:

Evdw = -

(E2.10)

Van der Waals forces become more significant when h is below 10 nm. It is often used

to describe the intermolecular interactions. For a better stability of a suspension, the

Brownian motion is certainly most interesting. For mass transfer of nanoparticles in the

presence of surface interaction forces over the filter grain dg, the Brownian diffusion

coefficient can be defined by the Stoke-Einstein equation:

(E2.11)

Pe =

(E2.12)

where Pe is the Peclet number (ratio of the convective motion of fluid to the movement

due to Brownian diffusion).

Particles transport by straining. When fluid containing particles approaches a pore

throat, particles which are too small to pass through, get stuck there. This phenomenon

is called straining or size exclusion (Farajzadeh,2004). It is determined by the ratio of

porous media (pore throat) diameter to the particle diameter. When this ratio is less than

10 (too small) cake will build up on the surface of the media (Farajzadeh, 2004). If the

concentration of particles is too high, they can also make a surface cake. In such cases,

45

many particles may reach the pore opening at a time and cram in it by aching action

(Zamani, 2010).

Particles transport by Electrical forces. Oppositely charged particles are attracted to a

charged fiber as shown in Figure 2.14. This collection mechanism is not limited to a

certain particle size and electrostatic charges may influence particle deposition.

van der Waals attraction forces and double layer repulsion forces are also significant for

the capture and detachment of the particles. They determine whether a particle will stick

to the grain or not. In other words, if the sum of the hydrodynamic and electrostatic

forces is attractive, a particle will be retained and if the sum is repulsive particles will not

adhere to the grain (Farajzadeh, 2004). All of the above mechanisms are summarized in

Figure 2.14. As NPs (dp<< 1µm) are used in the current work, it is believed that diffusion

is the most dominant mechanism as per Figure 2.15.

Electrostatic attraction

Figure 2.14: Overview of Filtration mechanisms (adapted from Wilcox et al., 2010).

+ -

46

Figure 2.14: Effect of particle diameter on collision probability (Farajzadeh, 2004).

Generally, three possible mechanisms may contribute to dispersed NPs deposition from

invert emulsion drilling fluid. Brownian diffusion, steric stabilization, and van der Waals

attraction forces. Double layer repulsion forces enhance particle stability in water based

muds.

a) NPs from the drilling fluid may physically plug or bridge across the flow paths in

the porous formation upon the first filter cake is formed by clay particles. NPs

plugging probability during drilling is shown in Figure 2.16. The movement or

transport of NPs by diffusion mechanism is believed to occur during fluid flow

through porous medium.

b) Chemical interaction between the fluid containing NPs and the formation rock

and drilling fluid may precipitate NPs or other semisolids that plug the pore

spaces. Adin et al. (1979) concluded that chemical interactions greatly affect the

47

attachment of particles on the surface. Raveendran (1993) found that water

molecules strongly bind with certain clays and minerals by the strong hydrogen-

bonding surface groups such as hydrated ions or hydroxyl (-OH) groups. Van der

waals attraction forces and electrical double layer repulsions can present in the

NPs based system depending on whether the surfaces have respectively unlike

and like potentials.

c) NPs with surfactant act as pseudo-solid and due to hydrophilic and hydrophobic

surface of surfactant effectively suspend NPs in the fluid system and create

stable drilling fluid through steric stabilization which potentially improves the lost

circulation.

Figure 2.16: NPs plugging mechanism during drilling (courtesy of nFluids Inc, reprinted by permission).

48

Chapter Three: Experimental Methods

This chapter describes the experimental procedures involved in the synthesis of the

different nanoparticles (NPs), preparation of the NP-based drilling fluid, and the

methods and instruments used to characterize the resultant NPs and the performance

of the product fluid. The particles were characterized using TEM, SEM with EDX and

ICP. The drilling fluid, on the other hand, was characterized using Fann mud balance for

density and Fann viscometer for viscosity and the gel strength measurements. The NP-

based drilling fluid lubricity was measured by Ofite lubricity tester and its filtration

properties were evaluated using API low temperature low pressure (LTLP) and high

temperature high pressure (HTHP) tests. Finally, the resultant mud cake was

characterized using SEM and EDX images and its thickness was measured by digital

caliper.

3.1 Drilling Fluid Samples

The invert emulsion muds used in this study were supplied by several Calgary based

drilling fluid companies, while the water-based mud was prepared in-house. Primarily,

two mixes of the invert emulsion drilling fluids were tested; namely 90 oil:10 water (V/V)

and 80 oil:20 water (V/V). The compositions of the invert emulsion and water-based

muds are shown in Table 3.1.

Table 3.1: Compositions of the invert emulsion and water based muds employed in this

work.

Invert Emulsion Mud Water based Mud

Oil: water (V/V) =90:10 or 80:20 Water= 500 mL

Base Oil= Low-aromatic oil Bentonite Clays= 10 g

Brine = 30% Calcium Chloride Surfactant = 0.5 g

Organophillic Clays =15 kg/m3 Xanthan= 1.5 g

Primary Emulsifier= 10 L/m3

Secondary Emulsifier = 5 L/m3

49

The loss circulation material (LCM) content of the drilling fluid, mainly Gilsonite, was

fixed at 1.6 wt% in the invert emulsion mud as recommended by one company. In some

experiments, nonetheless, no LCM was added in order to provide a bench mark to

evaluate NPs as sole loss prevention agents. The experiments mostly cover invert

emulsion based mud characterization, although a number of filtration tests of water

based mud were also performed. The in-house technique for the synthesis of the NPs

developed in this work is a chemo-mechanical process. The unique process has

enabled finely dispersed NPs formation in water-in-oil based drilling fluids as well as

water-based drilling fluids. The severity of the drilling process, nevertheless, may induce

particle agglomeration. However, the surfactants existing in a typical drilling fluid act as

stabilizers and limit agglomeration through steric hindrance (Husein and Nassar, 2008).

The method developed in this work is versatile and different types of precipitates of NPs

were prepared; including Fe(OH)3, CaCO3, FeS and BaSO4. Complete investigation

was, nevertheless, focused on the performance of Fe(OH)3 and CaCO3 NPs. The

method above uses two different approaches to prepare the NPs; namely in-situ and ex-

situ. The two terms essentially refer to whether the “birth place” of the NPs is inside or

outside the drilling fluid. The NPs concentration was varied between 1 wt% and 5 wt%

for both in-situ and ex-situ prepared particles.

3.2 NPs and NP-based drilling fluid formation

The choice of Fe(OH)3 NPs was based on the fact that the precursors are inexpensive,

and the product NPs are good scavengers for H2S that may evolve during drilling

(Husein et al., 2012a; Zakaria et al., 2012;Husein et al., 2010; Nassar et al., 2010).

Hydrogen sulfide is a very toxic and corrosive gas and may easily diffuse into the drilling

fluid from formation during drilling oil and gas wells (Sayyadnejad, 2008). Moreover,

H2S may evolve during completion when acid wash is used to remove metal salts scale

(Nasr-EI-Din et al., 2000). Similar to previous studies (Abdo and Haneef, 2010; Cai et

al., 2012; Srivatsa, 2010; Sensoy, 2009; Manea, 2011; Agarwal et al., 2009; Jimenez et

al., 2003; Paiaman and Al-Anazi, 2008; Roddy et al., 2009; Sayyadnejad et al., 2008),

50

and in order to establish the advantage of the in-house preparation method developed

in this study, commercial NPs were employed at an early stage. Fe2O3 NPs (Nano

structured and amorphous materials Inc, Texas, USA) were used to provide bench

marking. Fe2O3 is a thermally degraded form of iron oxide hydroxide (Balek and Šubrt,

1995). On the other hand, CaCO3 (fine/coarse) is a conventional lost circulation

materials used as a bridging agent and/or weighting material in oil based as well as

water based drilling fluids. The choice of nano CaCO3 in drilling fluid was due to its wide

range of applications in drilling fluid property manipulation from mud weight to fluid loss

control. Apart from those nanomaterials, fluid loss studies were also carried using FeS

and BaSO4. There is a synergy between Fe(OH)3 NPs and FeS NPs, as the latter is the

final product of the reaction between Fe(OH)3 NPs with H2S during drilling (Husein et al.,

2010; Nassar et al., 2010). Moreover, both FeS and BaSO4 (barite) are widely used as

weighting materials in drilling fluids (Chilingarian and Vorabutr, 1983; Moore and

Cannon,1936). They provide the high density needed to balance drilling operations.

3.2.1 Ex-situ preparation of NPs

In the ex-situ method the NPs are formed literally ‘out of place’ meaning the NPs

formation reactions take place outside the drilling fluid. NPs are formed from their

precursors in reaction vials at a standard condition. A general ex-situ NP-based fluid

preparation sequence is shown in Figure 3.1.

Figure 3.1: Schematic representation of the ex-situ method for NP-based drilling fluid preparation.

Reaction & mixing

at room temp

(25 °C and 200 rpm)

NPs mixing with drilling

fluid followed by high

shear action

(2500 rpm @ 30 min)

Drilling fluid

with desired

NPs

Aqueous precursor-2 Aqueous

precursor-1

Drilling Fluid

NPs

Slurry

51

3.2.1.1 Fe(OH)3 NPs

Ferric hydroxide NPs were prepared by first solubilizing a specific amount of anhydrous

iron (III) chloride powder (laboratory grade, Fisher Scientific Company, Toronto, ON,

Canada) in 2 mL deionized water to give final concentration of 2.5 M followed by

addition of a stoichiometric amount of NaOH(s) pellets (Fisher Scientific Company,

Toronto, ON, Canada) under 200 rpm of mixing and 25oC. The color of the aqueous

solution turned reddish brown signaling the formation of precipitate of Fe(OH)3(s) as per

reaction (R1).

FeCl3(aq) + 3NaOH(aq) Fe(OH)3(s) + 3NaCl(aq) (R1)

Part of the particles was recovered for characterization and the rest was mixed with the

invert emulsion drilling fluid in a slurry form. The fluids were mixed/sheared at 2500 rpm

to unifromly disperse the slurry using Hamilton beach mixer.

Similar to invert emulsion drilling fluid, ex-situ Fe(OH)3 NPs were also prepared in

the water based drilling fluid and tested only for LTLP fluid loss performance.

3.2.1.2 CaCO3 NPs

Calcium carbonate NPs were prepared by first solubilizing a specific amount of

anhydrous sodium carbonate powder (99% ACS reagent, Sigma-Aldrich Fine Chemical,

Toronto, ON, Canada) in 5 mL deionized water to give a final concentration of 2.26 M

followed by addition of 1 mL of 7.6 M stoichiometric amount of aqueous calcium nitrate

(99.5%, VWR, USA) under 200 rpm of mixing at 25oC. The color of the aqueous

solution turned white signaling the formation of precipitate of CaCO3(s) as per reaction

(R2). Part of the particles was recovered for characterization and the rest was mixed as

a slurry with 500 mL invert emulsion drilling fluid. The fluids were mixed/sheared at

2500 rpm to unifromly disperse the slurry ‎using Hamilton beach mixer.

Ca(NO3)2(aq) + Na2CO3(aq) CaCO3 (s) + 2 NaNO3 (aq) (R2)

52

Similar to invert emulsion drilling fluid, ex-situ CaCO3 NPs were also prepared in

the water based drilling fluid and tested only for LTLP fluid loss performance.

3.2.1.3 FeS NPs

Iron (II) Sulfide NPs were prepared by solubilizing a specific amount of aqueous sodium

sulfide (laboratory grade, Fisher Scientific Company, Toronto, ON, Canada) with

aqueous iron (II) chloride (laboratory grade, Fisher Scientific Company, Toronto, ON,

Canada). First, a specific amount of aqueous iron (II) chloride was solubilized in 1 mL

deionized water to give a final concentration of 3.4 M followed by the addition of 4 mL of

0.9 M stoichiometric amount of aqueous sodium sulfide and mixing at 200 rpm and

25oC. The color of the aqueous solution turned black signaling the formation of

precipitate of FeS(s) as per reaction (R3). Finally, the product was mixed with the drilling

fluid as a slurry using Hamilton beach mixer and sheared at 2500 rpm.

Na2S(aq) + FeCl2(aq) FeS(s) + 2 NaCl (aq) (R3)

3.2.1.4 BaSO4 NPs

Ex-situ preparation of barium sulfate NPs followed the same procedure as above. A

specific amount of aqueous barium chloride (Sigma-Aldrich, Toronto, ON, Canada) was

reacted with aqueous sodium sulfate (VWR, Calgary, Canada) under 200 rpm of mixing

at 25oC. First, a specific amount of aqueous barium chloride was solubilized in 3 mL

deionized water to give a final concentration of 1.14 M followed by the addition of 3 mL

of 1.14 M stoichiometric amount of aqueous sodium sulfate. The color of the aqueous

solution turned white signaling the formation of BaSO4(s) as per reaction (R4). Finally,

the product slurry was mixed with the drilling fluid using the Hamilton beach mixer and

sheared at 2500 rpm to achieve a homogenous mixture.

Na2SO4(aq) + BaCl2(aq) BaSO4(s) + 2 NaCl (aq) (R4)

53

3.2.2 In-situ preparation

The term in-situ NPs is used to mean that NPs were created in place. NPs were formed

from their precursors by reactions in the drilling fluid. Formation of NPs in this manner

minimizes particles aggregation and allowing easier handling than ex-situ prepared

NPs. A general in-situ NP-based fluid preparation scheme is shown in Figure 3.2. The

in-situ NPs preparation was used to prepare NP-based drilling fluid in invert emulsion as

well as water based muds.

Figure 3.2: Schematic representation of in-situ NP-based drilling fluid.

3.2.2.1 Fe(OH)3 NPs

In invert emulsion drilling fluid, this scheme of NPs synthesis followed the two

microemulsion method as per Husein and Nassar (2008). A 1 mL of 5 M FeCl3(aq) was

added to 250 mL of the drilling fluid and mixed at 200 rpm for 24 h. In a separate vial, 1

mL of 16 M NaOH(aq) (stoichiometric amount) was added to 250 mL of the drilling fluid

and mixed at 200 rpm for 24 h. The two vials were mixed and left overnight at 200 rpm

and 25oC. Two control samples were prepared, one containing the FeCl3(aq) in the

drilling fluid and another containing the NaOH(aq) in the drilling fluid, and the samples

were left to mix overnight at 200 rpm and 25oC. Finally, and in order to achieve a

uniform mixture of the fluid was sheared using Hamilton beach mixer at 2500 rpm. It is

worth noting that no phase separation was observed in the experimental as well as the

control samples, even after a period of 4 weeks.

Similar to invert emulsion drilling fluid, in-situ Fe(OH)3 NPs were also prepared in

the water based drilling fluid and tested only for LTLP fluid loss performance.

Aqueous precursor -2 with

drilling fluid

Reaction, mixing and high shear action

with drilling fluid at room temperature

(25 °C, 200 rpm and 2500 rpm @30 min)

Drilling fluid with

desired NPs

Aqueous precursor -1

with drilling fluid

54

3.2.2.2 CaCO3 NPs

Calcium Carbonate NPs were prepared in-situ in invert emulsion drilling fluids using two

different methods as per the reaction R2 and R5. The first method followed exactly

Fe(OH)3 NPs preparation presented in Figure 3.2. A 5 mL of 2.2 M sodium carbonate

was added to 250 mL of the drilling fluid and in a separate vial 1 mL of 7.6 M aqueous

calcium nitrate was added to 250 mL of the drilling fluid. The samples were left to mix

overnight at 200 rpm and 25oC. Finally, fluids were sheared at 2500 rpm again by the

Hamilton beach mixer before testing.

Similar to invert emulsion drilling fluid, in-situ CaCO3 NPs were also prepared in

the water based drilling fluid using the first method and tested only for LTLP fluid loss

performance.

The second method of in-situ synthesis of calcium carbonate NPs in invert

emulsion drilling fluid employed reaction (R5). Carbon dioxide, CO2 (99.9% purity,

Praxair, Edmonton) and Ca(OH)2 (92 % Purity, Canamara United Supply, Calgary) were

used as the precursors. First, 3 g Ca(OH)2 was added to 6 mL of water and then mixed

overnight at 200 rpm and 25oC. The calcium hydroxide solution was added to 500 mL

invert emulsion drilling fluid and sheared at 2500 rpm using a Hamilton beach mixer for

30 min. Carbon dioxide, CO2, gas was allowed into the drilling fluid sample through a

sparger for 10-20 min until pH changed to neutral. The pH of the invert emulsion drilling

fluid was roughly estimated using a pH paper. The mechanism leading to CaCO3 NPs

formation included CO2 transport to the water pools through the organic phase, reaction

in the water pools, Brownian collisions leading to material exchange, nucleation of

CaCO3, particle growth and may be aggregation due to Brownian collisions

(Bandyopadhyaya et al., 2001). Finally, to ensure uniform dispersion of the NPs ‎ a

Hamilton beach mixer was used at 2500 rpm. Figure 3.3 is a schematic representation

of the experimental procedure used to form the NPs as per reaction (R5). This approach

has serves within borehole NPs preparation, which may reduce particle aggregation

upon shearing drilling fluid through the drill bit. It may also help controlling CO2

emissions and converting it into value added product.

55

Ca(OH)2 (aq) + CO2(g) CaCO3 (s) + H2O (R5)

Reaction pathways as per Soony et al. (2002) involves the following steps.

CO2(g) CO2(aq) (R6)

CO2(aq) + H2O H2CO3 ; pH= 4 (R7)

H2CO3 H+ + HCO3

- (R8)

HCO3- H

+ + CO3 2- (R9)

Ca2++CO3 2- CaCO3 ; pH= 8 (R10)

Figure 3.3: Schematic of In-situ prepared CaCO3 NPs-based drilling fluid using CO2.

3.2.2.3 FeS NPs

In-situ Iron (II) Sulfide NPs were prepared by the method presented in Figure 3.2, and

only invert emulsion was employed here. First, 1 mL of 3.4 M aqueous iron (II) chloride

56

was added to 250 mL of the drilling fluid and in a separate vial 4 mL of 0.9 M aqueous

sodium sulfide was added to 250 mL of the drilling fluid. The samples were left to mix

overnight at 200 rpm and 25oC. Finally, to ensure good dispersion product NPs, the

invert emulsion mud was sheared at 2500 rpm by a Hamilton beach mixer for 30 min.

3.2.2.4 BaSO4 NPs

In-situ preparation of barium sulfate NPs followed the same procedure as above, and

again, only invert emulsion muds were employed. First, 3 mL of 1.14 M aqueous barium

chloride was added to 250 mL of the drilling fluid and in a separate vial 3 mL of 1.14 M

aqueous sodium sulfate was added to 250 mL of the drilling fluid. The samples were left

to mix overnight at 200 rpm and 25oC. Finally, to ensure good dispersion ‎product NPs,

the invert emulsion mud was sheared at ‎‎2500 rpm by a Hamilton beach mixer for 30

min. ‎

3.3 Characterization methods and techniques

3.3.1 Particle characterization

The ex-situ prepared NPs were characterized using X-ray diffraction patterns and

transmission electron microscopy, whereas the in-situ prepared NPs were characterized

as part of the mud cake following deposition on the filter paper using energy dispersive

X-ray spectroscopy.

The ex-situ prepared NPs were collected by centrifuging the aqueous colloidal

suspension at 5000 rpm for 30 min to recover the NPs followed by washing several

times with deionized water. The particles were left to dry at room temperature for 24 h.

The dried particles were ground using a pestle and mortar before been introduced to

Ultima III and Ultima IV Multipurpose Diffraction system. Ultima III operating at 40 KV

and 44 mA and Ultima IV operating at 40 KV and 40 mA use Cu Kα and Co Kβ radiation

respectively (Rigaku Corporation, USA) with a θ–2θ goniometer. Each scan used a 2°

step size from 0 to 90° for Ultima III and 5 to 90° for Ultima IV with a counting time of

2 ‎s/step. The structure was identified by comparing the diffractograms with spectra in

the JADE ‎program, Materials Data XRD Pattern Processing Identification &

57

Quantification. Ultima III and Ultima IV were used for Fe(OH)3 and CaCO3 based NPs

structure identification respectively. The particle size distribution was determined using

transmission electron microscopy, TEM. A small amount of the powder used for XRD

analysis was dispersed in 5 mL of methanol using sonication and one drop of the

methanol dispersion was deposited on a copper grid covered with carbon film, and was

left to evaporate for 24 h. In order to avoid possible aggregation upon methanol

evaporation, only a thin layer was deposited on the copper grid. The grid was then

introduced to a Philips Tecni (FEI USA Inc., Hillsboro, OR) TEM equipped with 200 kV

Field Emission Gun and Gatan ‎Imaging Filter (GIF) with a slow scan CCD camera. ‎

The in-situ prepared NPs, on the other hand, were characterized following their

collection on the filter cake. Filter cake of drilling fluid was dried at room temperature

(~25o C) for 4 days. The samples were then mounted and gold vapor applied to the

surface of the dried filter cake to create an electrically conductive layer necessary for

SEM photographs. In addition, an electrically conductive carbon particle suspension

was used to glue the filter cake sample to pedestal. Oil based mud needed more time to

achieve good gold coat, whereas water based mud took less time. Scanning electron

microscopy (SEM) analysis was performed on a FEI ESME XL30 (Philips XL30 ESEM,

USA). The instrument uses 20 kV acceleration voltage, WD= 10 mm with secondary

and backscattered electron signal. The images were taken under high vacuum mode

and recorded with a slow scan camera. Elemental analysis of the mud cakes was

performed with the energy-dispersive X-ray spectrometry (EDX) attached to the SEM.

3.3.2 Toxicity evaluation

The assessment of the environmental effects requires an evaluation of the NPs

ecotoxicity. The Microtox bioassay was used to assess the toxicity of the Fe(OH)3 NPs

only. Samples were analyzed using Microtox method (EC50 % at 15 min) in an external

lab (Kaizen Lab, Calgary, Canada). The effective concentration, EC50, is defined as the

concentration that produces a 50% light reduction (García et al., 2011) and was

measured after 15 min contact time. The test system measured the light output of the

luminescent bacteria of the NPs sample and compared it to the light output of a control

58

sample containing no NPs. A difference in light output between the sample with NPs

and the control was attributed to the effect of the NPs on the organisms. Microtox is a

qualitative test of toxicity and the protocol of testing is shown as follows (Kaizen Lab,

Calgary, Canada).

Aqueous extraction of solid samples

Create dilution Determine EC50 (15°C,15 min)

3.3.4 Emulsified water droplet measurement

Water-in-oil invert emulsions with primary emulsifier were prepared using 10 v/v%

water, which is the same as the drilling fluid, except for the fact that solids were not

added. The water droplet diameters of the invert emulsion were measured using

Morphologi G3 microscope (Malvern Instruments Inc, USA).

3.3.5 Drilling fluid characterization

The filtration properties of the different drilling fluids involved in this study were

measured according to API 30-min test (API RP 13B-2,2012; API RP 13B-1,2003).

Data were collected using a standard FANN filter press (Fann Model 300 LTLP, Fann

Instrument Company, USA) and filter paper (pore size 2.7 µm, Fann Instrument

Company, USA). The low temperature low pressure (LTLP) test was conducted

according to the following procedure. A volume of 500 mL of the drilling fluid was

poured into the filter press cup and 100±5 psi of pressure was applied through CO2

supply cylinder at room temperature of 25oC. The cumulative volume of permeate was

Pass : EC50 ≥ 75%; Non-toxic

Fail : EC50< 75%; Toxic

59

reported after 7.5 min and 30 min from the graduated cylinder reading. Three replicates

were prepared for every sample and 95% confidence intervals are reported in the

results. The smoothness of the final filter cake was reported through visual observation,

while the thickness was measured using a digital caliper (0-6˝ TTC Electronic digital

calipers model # T3506, Canada). The concentration of the NPs in permeate was

correlated to iron and calcium concentration measured by an inductively coupled

plasma (ICP) (IRIS Intrepid IIXDL, Thermo Instruments Canada Inc., Canada). A

portable 175-mL Ofite high temperature high pressure (HTHP) filter press (OFI Testing

Equipment Inc, USA) and filter paper (pore size 2-5 µm, specially hardened for filter

presses, Fann Instrument Company, USA) was used to study the filtration

characteristics of the mud at differential pressure of 500 psi and temperatures of 177°C

(350oF). Fann LTLP and Ofite HTHP filter presses are shown in Figure 3.4.

Figure 3.4: Drilling fluid loss apparatus for a) LTLP and b) HTHP tests.

It should be noted that the area of the filter paper used in the HTHP filter press is one-

half the area of the standard filter press. Therefore, the volume of filtrate collected in 30-

min is typically reported as double. Commercial invert emulsion drilling fluid without NPs

and LCM (Gilsonite) and with 1.6 wt% LCM only were considered as baseline drilling

fluids for comparative evaluation of API and filtration experiments. Invert emulsion

a) b)

60

drilling fluids containing NPs and 1.6 wt% LCM or NPs only were considered as the

nano-based drilling fluids of study.

Fann Model 140 mud balance (Fann Instrument Company, USA) was used to

measure the mud density. During the measurement care was taken in order to eliminate

any errors due to air entrapment. The pH measurements were performed using pH

papers (0-14) (VWR international, Calgary, Canada). Several readings were collected in

order to ensure precision, despite the limitations of pH determination using pH paper,

since a pH meter could not be inserted in the mud. In addition, and in order to provide

more reliability for the method, hydrated lime, similar to the one used in drilling fluid,

was dissolved in water at the same concentration as the water in the drilling fluid, and

its pH was measured using pH meter (Model: AccumetAB15+,Fisher Scientific, Toronto,

Canada ) and pH papers.

A rotational Fann 35A viscometer (Fann Instrument Company, USA) was used to

measure the rheological properties of the drilling fluid at six different speeds as shown

in Figure 3.5.

Figure 3.5: Fann Model 35A viscometer for measuring viscosity.

A volume of approximately 350 mL of the fluid was poured into the viscometer cup, and

the mud was sheared at a constant rate in between an inner bob and outer rating

sleeve. The system was left to rotate at a certain rpm until reaching the steady state

reading for 5 min. The readings were collected at 600, 300, 200, 100, 6 and 3 rpm.

These experiments were conducted at room temperature. The dimensions of bob and

61

rotor were chosen such that the dial reading on the viscometer is equivalent to apparent

viscosity in centipoises at rotor speed of 300 rpm. The apparent viscosities for all rotor

speeds were calculated using equation (E3.1) below (Fann 35 viscometer manual,

2008):

Apparent/ Effective viscosity, μa = 300

N

(E3.1)

where N is the rotor speed (rpm) and is the viscometer dial reading (o). The shear rate

can be calculated as per equation (E3.2) (Fann 35 viscometer manual,2008).

Shear rate, sec-1 = 1.7023N (E3.2)

The plastic viscosity and yield point are found using the following equations (Fann 35

viscometer manual, 2008):

Plastic viscosity, μp = ϴ600 - ϴ300 (E3.3)

Yield point, Yp = ϴ300 - μp (E3.4)

where μp plastic viscosity (cP), Yp yield point (lbf/100ft2), ϴ600 and ϴ300 are the dial

readings at 600 and 300 rpm, respectively.

Gel strength of the drilling fluid was measured at lower shear rate after the drilling mud

is static for a certain period of time. The readings at 3 rpm were taken after 10 sec and

10 min following stirring the drilling fluid at 600 rpm for 5 min. The first reading noted

after the mud is in a static condition for 10 sec is called 10 sec gel strength. The second

gel strength noted after 10 min is called 10 min gel strength. Gel strength is usually

expressed in pressure unit lbf/100ft2. The difference between the initial gel strength and

those taken after a 10 min test period were used to define how thick the mud would be

during round trips.

Lubricity test is designed to simulate the speed of rotation of the drill pipe and the

pressure the pipe bears against the wall of the bore hole (OFITE lubricity test manual,

2011). It also predicts the wear rates of mechanical parts in known fluid systems.

Lubricity property of the NP-based drilling fluid was evaluated by OFITE Lubricity Tester

(Part no: 111-00, serial: 07-09, Houston, USA) at 150 inch-pounds of torque which are

62

applied to two hardened steel surfaces, a block and ring rotating at 60 rpm rotational

speed as shown in Figure 3.6.

Figure 3.6: OFITE drilling fluid lubricity tester.

The test sample is completely immersed between the ring and block. The apparatus

runs for 5 min in order to coat the metal test pieces with the sample fluid. The torque

adjustment handle is then turned until 150 inch-pounds of torque have been applied to

the test block. The machine again runs a 5 min stabilization period. A friction coefficient

reading is then taken. Additional readings are taken every 5 min until three consecutive

readings agree within ±2 units. The drilling fluid lubricity coefficient can be calculated

using the following equation as given in the Ofite manual (Ofite lubricity tester manual,

2011).

Coefficient of friction = applied load torque lb

ring the turn to force lb =

100

Reading Meter (E3.5)

Coefficient of Friction (CoF) is used to quantify how readily two surfaces slide in the

presence of a lubricant or oil. It is a key factor which directly affects the torque and drag.

The lower the value of the coefficient of friction, the higher the lubricity, or vice-versa.

The torque reduction at a given load can be calculated using the following equation.

Percent torque reduction at given load = 100xA

B(A

L

LL ) (E3.6)

where AL = Torque meter reading of untreated mud (inch-pounds)

BL = Torque meter reading of treated mud (inch-pounds)

63

Chapter Four: Results and Discussion

Initially the experimental analysis was performed on the base drilling fluid, mostly invert

emulsion containing all ingredients, e.g. organophilic clays, primary and secondary

emulsifiers, brine, etc. except for the lost circulation materials (LCMs), to understand the

nature of the fluid loss, and later with drilling fluid containing conventional LCMs, i.e.

Gilsonite. The next step was to test these fluid systems in the presence of in-house

prepared nanoparticles (NPs), and commercial Fe2O3 NPs. Concentrations between 1-4

wt% NPs were used depending on the stability of the NPs in the drilling fluid (Husein et

al., 2012 a&b; Zakaria et al., 2012). Needless to say that low concentrations were

targeted to study large-scale application. Results pertaining to the detailed preparation

and performance of Fe(OH)3 NPs are presented first. Then, the preparation and

performance of CaCO3 NPs were considered in details, due to the wide application of

CaCO3 particles in drilling fluids (Manea, 2012; Simon et al., 2010; Whitfill at al., 2003).

Finally the preparation and performance of BaSO4 and FeS NPs were considered in

order to prove the applicability of the in-house preparation method developed in this

work to other NPs. It should be noted also that BaSO4 and FeS are widely used as

weighting material in drilling fluids (Scott and Robinson, 2010; Moore and Cannon,

1936). In all cases the NP-based fluids were compared with the corresponding control

DF samples. The use of NPs to increase the mud density, while minimizing sagging was

demonstrated by other groups (Amanullah et al., 2011). Therefore, the characteristics of

NP-based DF were evaluated by measuring mud weight, pH, viscosity, gel strength, API

LTLP and HTHP filter tests and lubricity as described in Chapter 3.

4.1 Fe(OH)3 Nanoparticles (NPs) Characterization

The structure of the ex-situ prepared Fe(OH)3(s) NPs was determined using X-ray

diffraction (XRD) patterns, whereas its particle size distribution was evaluated using

TEM photographs. Detailed particle characterization is provided herein.

64

4.1.1 X-ray diffraction analysis

The X-ray diffraction pattern of the ex-situ prepared NPs shown in Figure 4.1 indicates

no evidence of strong distinct peaks which would be expected from a crystalline

material. This said, the most likely product as suggested by the figure is Fe(OH)3(s). The

peak maximum around 2θ= 35° can be attributed to the presence of aggregates

dispersed in an amorphous phase (Zakaria et al., 2012). Streat et al. (2008) has also

prepared ferric hydroxide using ferric chloride and stoichiometric quantity of sodium

hydroxide in deionized water and reported the same XRD pattern.

Figure 4.1: X-ray diffraction pattern for the ex-situ prepared iron-based NPs.

Reaction pH might affect the final nature of the iron oxide/hydroxide product. The

optimum pH for the precipitation of Fe(OH)3(s) is found 5 (Zakaria et al., 2012). Liu et al.

(2005) reported the same initial pH level of the precipitated amorphous Fe(OH)3

prepared from aqueous FeCl3 and NaOH precursors. Phase transformation of Fe(OH)3

gel to α-Fe2O3 particles was impacted by the pH range. In a different study, Cai et al.

(2001) found that at room temperature the reaction pH affected the crystallinity of iron

oxide material. They reported narrow and distinct peaks for 1.5≤ pH< 4. At pH= 4 there

were two broad and less intense peaks similar to the ones appearing in Figure 4.1

suggesting poor crystallinity. At pH≥ 6, crystallinity was re-gained. It is to be noted that

65

amorphous iron (III) hydroxide can transform into α-Fe2O3, β-FeOOH or α-FeOOH with

the change in reaction temperature (Nassar and Husein, 2007a). The X-ray diffraction

pattern of a filter cake collected following LTLP test of a drilling fluid containing ex-situ

prepared Fe(OH)3 NPs and organophilic clays is shown in Figure 4.2. The pattern in

Figure 4.2 suggests that Fe(OH)3 NPs acted as intercalating agents and had entered

into the crystallite layers of bentonite clay. It is believed that such a structure of

nanometal-clay composite could improve drilling fluid properties; including loss

prevention and wellbore strengthening. In addition, this structure suggests good

compatibility, dispersion and communication between the NPs and the rest of the drilling

fluid constituents, which ultimately could offer better functionality than regular bentonite

clays. NPs embedded randomly on the surface of clay particles promote gelation of the

bentonite particles (Baird and Walz, 2006; Lee et al., 2010). Similar observation was

also reported by Fernandez et al. (2010) using acrylamide polymer with bentonite clays.

An important outcome of good gelation is inhibition of clay swelling as reported by Mei

et al. (2011).

Figure 4.2: X-ray diffraction pattern of ex-situ prepared Fe(OH)3 NPs collected on the filter paper.

66

4.1.2 Water droplet size distribution

Emulsion samples were prepared by mixing water with the similar type base oil (low

aromatic oil) and the primary emulsifiers used in the drilling fluids formulation but no

solids, i.e. bentonite, were used. The water droplets were observed in the oil phase by a

microscope. It was clear that 10/90 (V water/V oil) contained droplets size from 1-30 µm

with mean droplet diameter at 20 µm as shown in Figure 4.3. Water-in-oil (w/o)

microemulsions have been demonstrated as a very versatile and reproducible method

that allows control over nanoparticle size and yields particles with a narrow size

distribution (Lopez-Quintela, 2003), by virtue of their nanometer scale water pools. In

principle, invert emulsion drilling fluid can be employed in a similar manner to prepare

NPs. However, knowing the fact that invert emulsions typically contains much larger

water pools, as shown in Figure 4.3, mixing becomes a very important parameter. In

studies by Anisa and Nour (2010) and Fjelde (2007), it was shown that stirring speed

largely affects the droplet size distribution in (w/o) emulsions. Higher shearing and

duration of stirring lead to a very tiny droplets, which can act as nanoreactors. In this

works, the NPs were stirred at 200 rpm during preparation and sheared in the drilling

fluid at 2500 rpm, which enabled them to accommodate into the water pools effectively.

Figure 4.3: Particle size distribution histogram of water droplet obtained from a water-in-oil emulsion by dispersing water into base-oil with the aid of a primary emulsifier.

67

As can be seen in Figure 4.3, most of the droplets were 11-20 µm in diameter. It should

be noted, nevertheless, that the detection limit of the instrument used was down to 0.5

μm in size. Therefore, Figure 4.3 should be considered with caution. Similar experiment

has been performed by Fjelde (2007) for 25/75 and 5/95 (V water/V oil) emulsions in the

presence of primary and secondary surfactants and water droplet sizes between 3-50

μm were reported for both mixes at different temperatures. Generally, the water droplets

in an emulsion may vary in size from less than 1 μm to more than 1000 μm (kokal,

2006). Typically in oil based drilling fluids, macroemulsion, which may have droplet

sizes in the range from 0.1-100 μm (Kokal, 2006; Bumajdad et al., 2011), are used.

Generally microemulsions consist of nano-sized water pools dispersed within the

bulk organic phase which act as nanoreactors for the chemical reduction of the metallic

precursors and metallic nanoparticle preparation (Kitchens,2004). Size of the particles

can be controlled by surfactant/co-surfactant type, concentration of the reagents and

water/surfactant molar ratio (Zielińska-Jurek et al.,2012). The total concentration of

metal ions under the present experiment is so small that the influence of water droplet

size has no great influence on the nanoparticles formation and growth. Moreover, the

primary and secondary surfactant rich ‎stabilized invert emulsion fluid limit particle

growth and agglomeration of metal ‎particles in water pools and renders particle sizes in

the nm scale (Husein and Nassar, 2008). ‎

4.1.3 Size distribution of ex-situ prepared Fe(OH)3 NPs

The TEM photographs and the corresponding particle size distribution histogram for the

ex-situ prepared Fe(OH)3 are shown in Figure 4.4. The histogram shows a spread in the

size distribution with most of the population falling in the range between 1-30 nm. The

photograph confirms that there is good degree of agglomeration, which must have

resulted from the high degree of collision between the precipitated particles while

shearing, especially since no surfactants were added to the aqueous phase.

68

Figure 4.4: TEM photographs and corresponding particle size distribution histograms of the ex-situ prepared Fe(OH)3 NPs in the range between a) 1-120 nm and b) 1-30 nm.

Dispersing the ex-situ prepared NPs by ultrasonication in methanol for 10 min before

deposition on the TEM grid did not seem to eliminate aggregation, despite the fact that

the NPs were not found to exhibit magnetic properties. Therefore, it is concluded that

this agglomeration at room temperature is not due to magnetic attraction, but rather due

to the high surface energy of the particles (Bumajdad et al., 2011). Once mixed with the

drilling fluid, the surfactant rich stabilized invert emulsion, or water-based muds, limits

the aggregation of the ex-situ prepared particles, especially since the concentration of

NPs in the drilling fluid is kept very low, < 5 wt%. On the other hand, in-situ prepared

NPs are expected to be very well dispersed by virtue of the surfactant component of the

drilling fluid mix.

The wide size distribution of particles has prompted an investigation on the

filtration characteristics of LCM-free NP-based drilling fluid. The results of this

investigation are detailed below.

4.1.4 Determination of particle size of in-situ prepared Fe(OH)3 NPs

When NPs are prepared in-situ within invert emulsion drilling fluids, it is not easy to

separate the particles for characterization. Alternatively, these particles could be

characterized following their collection on the mud cake. SEM images of the mud cake

a) TEM Photographs Particle Sizes from 1 to 120 nm b) Particle Sizes from 1 to 30 nm

100 nm

Particle size (nm)

69

without and with NPs are shown in Figure 4.5a,b, respectively. The observed

morphologies of the two samples have some distinct features. The mud cake with NPs

was fully intact and displayed a very smooth surface with no visible cracks at even 48

times magnification. It is worth noting that the cake surface was covered with Fe(OH)3

particles, as was visually confirmed from the reddish brown color of the surface. The

texture of the mud cake in the absence of NPs was rough and full of cracks. It is,

therefore, plausible to believe that the voids and gap of pores were effectively filled with

NPs, and NPs acted as an effective filling agent. This said, one should also keep in

mind the effective intercalation between Fe(OH)3 NPs and the organophilic clays

reported earlier. These observations are very important for the explanation of fluid loss

prevention and lubricity reported in this study as well as wellbore strengthening reported

in a recent study, which employed the same method and NPs developed in this study

(Nwaoji et al.,2013; Nwaoji,2012). Lastly, effective adsorption/deposition of Fe(OH)3

NPs on the organophillic clays forming the ‎cake may also contribute to a surface

chemical reactivity, which can provide further sealing. Lai et ‎al. (2000) reported that

Cu2+ ions were effectively adsorbed onto iron oxide-coated ‎sand. Finally, it is important

to note that mud cakes tested were LCM free.

Figure 4.5: SEM Images at 48x magnification of mud cakes collected following API LTLP filtration tests a) without NPs, b) with 1 wt% in-situ NPs (90/10(v/v) oil/water invert

emulsion mud, no LCMs).

a) b)

Crack

70

The elemental distribution mapping of EDX for the sample of mud cake without NPs and

mud cake with NPs are depicted in Figure 4.6. Through elemental analysis it was

determined that 0.7 wt% of iron ion was found on the mud cake. Results indicated that

iron ions could trap into the micropores and mesopores of the cake-containing clays. It

can be also attributed to diffusion of adsorbed metal species from the surface into the

nanopores, which are the least accessible sites of adsorption. This is believed to

contribute to effective sealing while filtering.

Figure 4.6: Elements contained in mud cake a) without NPs, b) with Fe(OH)3 NPs as

per EDX analysis.

4.2 Drilling fluid Characterization

4.2.1 Stability of NP-based Fluid

Visual observation was used to assess the stability of the NP-based fluids. Stability

against agglomeration and sagging relates here to the ‘shelf life’ of the NP-based fluid.

Figure 4.7 shows photos of samples representing the original drilling fluid (90 vol. oil/10

vol. water) invert emulsion samples without and with in-situ prepared Fe(OH)3 NPs. The

photos show no sign of sagging or aggregation even after 4 weeks of setting at room

temperature and confirm that, at the concentration of the added NPs, no agglomeration

or sagging takes place. Therefore, no extra additives were required to stabilize the NP-

base drilling fluid. The stability of an invert emulsion system containing dispersed

particles can be attributed to a steric effect conferred by adsorbed materials, mostly

71

surfactant molecules, onto the particles (Husein and Nassar, 2008; Nassar and Husein,

2007a). Careful evaluation of the system stability as a result of NPs addition showed

that no sagging was experienced for the different NPs considered in this work; including

Fe(OH)3, CaCO3, BaSO4 and FeS, up to 5 wt% and stable samples are obtained for

several weeks. This applies to both, in-situ prepared NPs and NPs prepared ex-situ and

then mixed with the drilling fluid.

Figure 4.7: Photos comparing NP-based and original invert emulsion drilling fluids

(Invert emulsion (90 vol. oil/10 vol. water); 1 wt% Fe(OH)3 in-situ prepared NPs).

NPs that grow or agglomerate to sizes beyond the stabilization capacity of invert

emulsion fluid might settle under gravity, which was not apparent in the above photos.

This suggests that the mixing provided during the preparation of in-situ and ex-situ NPs

as well as during mixing of ex-situ prepared NPs with the drilling mud was sufficient to

provide dispersion at a molecular level, which, in turn, leads to the formation of very

small particles that are well dispersed and stabilized in the rest of the fluid. As stated

earlier, adsorption of emulsifiers on the surface of these particles helps further

stabilizing them within the fluid. A qualitative assessment of the stability of the

nanobased fluid was done by checking its rheology behavior after 1 month which is

detailed in the rheology section.

4.2.2 LTLP Filtration

Filtration property is dependent upon the amount and physical state of colloidal

materials used in the mud. When mud containing sufficient colloidal material is used,

NP-based Fluid Original Drilling Fluid

a) b)

NP-based Fluid Original Drilling Fluid

72

fluid loss can be minimized as these materials will deposit and contribute to cake

formation, which increases resistance for fluid permeation. This resistance is highly

dependent on the structure and integrity of the filter cake. Details on the integrity and

structure of the cake were provided earlier. On the other hand, the spurt loss of the

drilling fluid is considered as one of the sources of solid particles and particulate

invasion to the formation, which can cause serious formation damage as a result of

internal mud cake formation in the vicinity of the wellbore (Amanullah et al., 2011; Al-

Hitti et al, 2005; Peng, 1990). Internal pore throat blockage may create a flow barrier

which reduces oil and gas flow. Moreover, higher particle flocculation in drilling fluid

leads to a thicker mud cake which increases the probability of differential sticking and

stuck pipe problems (Amanullah et al., 2011). This highlights the importance of using

low concentration of dispersed NPs in fluid design with virtually no spurt loss, low filtrate

volume and good quality filter cake.

4.2.2.1 Commercial NPs

At first, commercial iron oxide NPs were introduced into the commercial invert emulsion

drilling fluid as per literature procedure (Agarwal et al.,2009; Amanullah et al., 2011;

Srivatsa, 2010; Abdo and Haneef,2010), which involved mixing at 2500 rpm for 30 min.

This experiment served as bench marking. The performance towards fluid loss

prevention was very poor as can be seen in Table 4.1. It is to be noted that the original

drilling fluid (DF) was completely LCM free. A large amount of small ‘fish eyes’ (lumps of

agglomerated commercial NPs) on the NP-based mud cake was clearly seen, as shown

in Figure 4.8. It appears that, even under the high shear mixing used to prepare the in-

house NPs, commercial NPs did not seem to effectively disperse into the drilling fluid.

This, in a way, limited their interaction with the clays and resulted in a poorly structured

filter cake. The mud cake in the absence of NPs was provided for comparison. The

thickness of the mud cake developed upon filtering commercial NP-based drilling fluid

was 0.76 mm, whereas the one obtained from filtering the invert emulsion mud was 0.31

mm.

73

Table 4.1: API LTLP loss of drilling fluid in the presence and abscense of 1 wt% commercial Fe2O3 NPs. NPs were thoroughly mixed with the invert emulsion drilling fluid. No LCMs added to both samples.

Figure 4.8: Mud cake of drilling fluid with commercial NPs and without NPs.

4.2.2.2 In-house prepared Fe(OH)3 NPs

Following the hypothesis outlined earlier; in-house prepared NPs may better

interact ‎with the drilling fluid, especially the in-situ formed ones, in-house prepared

Fe(OH)3‎ NPs were formulated inside, or added to, the drilling fluid. In-house ‎Fe(OH)3‎

NPs at 1 wt% and size varying from 1-120 nm had better plugging performance

than ‎commercial Fe2O3 NPs and will be detailed in the fluid loss experiments. ‘Fish

eyes’, which ‎appeared in the mud cake containing the commercial NPs, were minimized

in the presence ‎of the in-house; both ex-situ and in-situ, formulated NPs as can be seen

Samples Types

Commercial NPs Used (20-40 nm)

Time (min)

LPLT Fluid Loss (mL) Fluid Loss Reduction

% DF DF with

1 wt% NPs

90:10 (v/v) Oil: Water

Fe2O3/FeOOH

7.5 1.7±0.6 1.7±0.6 0

30 4.5±0.6 4.2±0.6 6.67

Commercial NP-based mud cake

Mud cake without NPs

fish eyes

74

in Figure 4.9 and 4.10. Moreover, NaCl, which is a by-product of the Fe(OH)3 formation

reaction, is commonly used as ‎a bridging solid to prevent clay swelling and clay

dispersion, which, in turn, lead to the ‎minimiiinm clay related formation damage (Mohan

et al., 1993; Crowe ,1990). ‎

Generally, the characteristics of the resultant filter cake depended on the degree

of peptization or ‎flocculation of the suspension. Stable (peptized) suspensions form

dense and ‎compact sediments, while flocculated suspensions form more voluminous

sediments ‎and particles are associated in the form of a loose, open network (Smith and

Hartman, 1987). Filter cake ‎formed from stable dispersion of NPs is relatively

impenetrable, and hence, creates more resistance to flow in comparison to that formed

from flocculated commercial NPs. This might explain why in-house prepared NPs

showed better performance. ‎In-house prepared NPs are better dispersed in the drilling

mud. Therefore, they effectively adsorbed into the pore space of clay platelets and

formed well dispersed plastering effect on the filter paper. This implies lower penetration

of drilling fluid into the formation and, hence lesser damage to the formation. In-house

prepared NPs progressively built up on the surface of the filter cake and acted as a shut

off valve. Effective mud cake resulted in much lower fluid loss as can be clearly seen in

Table 4.2.

Table 4.2: Comparative study of API LTLP fluid loss of drilling fluids with 1.6 wt% conventional Gilsonite LCM, and 1 wt% in-situ and ex-situ prepared NPs.

*Fluid loss reduction, %.

Original drilling fluid (DF) without NPs and LCM and drilling fluid with 1.6 wt% LCM were

considered as a baseline for comparative evaluation of fluid loss property of the ex-situ

Samples

Types

Time

(min)

LTLP Fluid Loss (mL)

DF DF +LCM DF+LCM with 1

wt% ex-situ NPs

DF+LCM with 1

wt% in-situ NPs

90:10 (v/v) Oil: Water

7.5 2.0±0.2 1.4±0.2 0.2±0.2 0

30 3.96±0.2 3.6±0.1 (9%*) 1.1±0.1 (72%*) 0.5±0.2 (87%*)

Cake thickness, mm 0.31 0.76 0.52 0.44

75

and in-situ prepared nanobased fluid. Based on the original DF, fluid loss over a period

of 30 min decreased by 9% for the drilling fluid containing 1.6 wt% LCM only, while it

decreased by 70% for the drilling fluid containing 1 wt% ex-situ prepared Fe(OH)3 NPs

and by more than 80% for the drilling fluid containing the in-situ prepared Fe(OH)3 NPs.

Both ex-situ and in-situ drilling fluid samples contained 1.6 wt% Gilsonite LCM. In-situ

prepared NPs, which, as stated earlier, immediately adsorb onto ‎neighbouring clay

platelets when squeezed through the filter cake by virtue of their ‎high dispersion, fill the

pores and the gaps of clay network and, hence tremendously ‎lower the fluid loss

compare to the ex-situ prepared NPs. On the other hand, for the ‎typical LCM, particles

larger than pore opening cannot enter the pore at first and ‎might be swept away by the

mud stream, of course under dynamic drilling.

During spurt loss period (t< 7.5 min), mud particles attempt to flow with the

filtrate ‎through the filter paper. The emulsion droplets provide sufficient surface area for

the ‎water-containing NPs to spread on the mud cake. This may have resulted in NPs

bridging across pore ‎throats to form the external mud cake immediately, and thus

lowering the spurt loss. ‎Iron oxides/hydroxides have affinity for negative charges

(Follett, 1965), while the edges of the betonite clay are negatively charged (Xu et al.,

2005; Lai et al., 2000; Follett, 1965). This ‎may explain the high particle-clay interaction

during filtration (Xu et al., 2005; ‎Lai et al., 2000; Follett, 1965). Fluid loss control of

drilling muds using similar approach was not ‎reported in the literature. Most of the

literature on NP-based drilling muds considered ‎water based muds employing

commercial NPs, and loss reduction of 40% was reported for 1-30 wt% NPs (Amanullah

et al., 2011; Srivatsa, 2010; Cai et al., ‎‎2011). Using similar explanation to Aston et al.

(2002), NPs probably acted at the ‎interfacial region between the emulsion droplets and

the oil phase when pressure is ‎applied during filtration and made the region viscous.

This phenomenon could slow ‎down the flow of oil through the cake and thereby lower

the fluid loss. Moreover, there ‎could be an additional effect from NPs acting as bridging

agents between long chain hydrocarbons, including those of LCM molecules, in the

invert emulsion drilling fluids.‎

76

The above results are particularly important when drilling in shale formations.

Even though shales have macro to nano pores, shales are very sensitive ‎to water loss

since they tend to swell easily (Chenevert and Sharma,2009). Conventional LCMs will

not be able ‎to block the nanopores due to their micron sizes. Therefore, smaller

particles, i.e. NPs, ‎are needed to better fit the nanopores.‎

In order to prevent drilling and completion problems, mud cake quality and build

up characteristics are also very important. Figure 4.9 includes photographs of the mud

cake formed in the presence and absence of NPs. Compared with LCM based cake, the

NP-based drilling fluid produced thin mud cake less than 1 mm. The NP-based DF

deposited a fine thin layer of iron (III) hydroxide NPs on the cake surface. Addition of

NPs did not cause an increase in the thickness of the mud cake, especially since small

concentrations of NPs were used in fluid formulation and these NPs are believed to be

located on the top of the clays and eventually filled the gap or holes in the clay platelets.

The NPs are subsequently captured within the clay layers. This multiple layer structure

provides much better sealing, prevents further flow through the pores, and subsequently

lower clay deposit and thinner filter cake. During filtration clays provided disordered

stacking and displayed the highest permeability. NPs reduced this roughness of clay

surface by the thickness of the deposit. It could be associated with dispersion ability of

nanoparticles to be well-distributed more effectively on the surface of bentonite clays or

‘intercalation’ of NPs in clay layers provided lower permeability. This eventually

decreases the volume of the cake leads to a minimum amount of fluid in the pores.

Moreover small concentrations of NPs were used in fluid formulation. On the

other hand, large sized LCM could not lodge in the porous space of the cake and the

cake exhibited sufficient porosity to permit continued flow through it as filtration

proceeds. This, in turn, led to more clay depositing onto the cake and particles

accumulation. Moreover, Figure 4.9 c-d shows that a layer of NPs was the last to

deposit on the cake surface leading to crack-free and smooth surface. Thin filter cake

suggests a high potential for reducing the differential pressure sticking problem while

drilling.

77

Figure 4.9: Mud Cakes with thickness of a) DF only, b) DF+LCM, c) DF +LCM with 1 wt% ex-situ NPs, and d) DF+LCM with 1 wt% in-situ NPs.

Because the fluid loss performance is improved dramatically with the Fe(OH)3 ‎NPs

additives, it raises the question as to whether conventional LCMs are still needed

to ‎control the fluid loss, especially in light of the fact that NPs displayed a relatively wide

size distribution, at least the ‎reported ex-situ prepared ones. These measurements

suggest that a wide size range of NPs can be used as substitutes for ‎conventional

LCMs in the mud, e.g. Gilsonite. The hypothesis ‎was larger size NPs would contribute

to blocking large formation pores and help ‎bridging large voids, and once a primary

bridge is established, successively NPs, down ‎to few nm, are trapped and thereafter

stop the filtrate from invading the formation. ‎The filtration properties of a drilling fluid

with NPs only also consider the wall/cake building ability of the NPs with the solid

components of drilling fluid such as clays. The results of the API low temperature low

pressure LTLP experiments are shown in Table 4.3. An interesting observation was that

a wide range of NPs size distribution gave the lower filtrate volume than the Gilsonite

LCM. A reasonably low fluid loss value and thin mud cake with a thickness of less than

1 mm significantly improved the performance of the NP-based drilling fluid. These

results are summarized in Figure 4.10.

a) b) c) d)

Thickness= 0.31mm Thickness= 0.76 mm Thickness= 0.52 mm Thickness= 0.44mm

78

Table 4.3: API LTLP fluid loss comparing drilling fluid and drilling fluid with Gilsonite LCM as base cases with drilling fluid samples containing the in-house prepared Fe(OH)3 NPs only with no LCM.

*Fluid loss reduction, %.

Figure 4.10: Mud Cakes of a) DF only, b) DF+LCM, c) DF with 1 wt % ex-situ NPs and d) DF with 1 wt % in-situ NPs.

4.2.3 Filtrate Characterization

Loss of fluid from invert emulsion drilling muds usually allows oil and chemicals into the

formation. In order to provide a measure of how much NPs seeped through the filter

cake during API LTLP filtration, the concentrations of iron and calcium in the filtrate

were determined using inductively coupled plasma (ICP). In the total filtrate volume, the

Fe(OH)3 NP-based fluid reduced the calcium content 500 times relative to the drilling

mud alone. It should be noted that typically the aqueous phase of the invert emulsion

drilling fluids contain calcium hydroxide in order to control alkalinity (Chilingarian and

Vorabutr, 1983). On the other hand, no iron was found in the original drilling fluid or the

NP-based drilling fluid, as shown in Table 4.4. The results can be attributed to the fact

that clays are negatively charged and adsorbed species with high affinity to negative

charges such as iron oxide/hydroxide (Xu et al., 2005; Lai et al., 2000; Follett, 1965), as

discussed earlier. Therefore, NPs provided bridges between the clay particles reducing

Samples

Types

Time

(min)

LTLP Fluid Loss (mL)

DF DF +LCM DF with 1 wt%

ex-situ NPs

DF with 1 wt%

in-situ NPs

90:10 (v/v) Oil: Water

7.5 2.0±0.2 1.4±0.2 0.15±0.1 0

30 3.96±0.2 3.6±0.1 (9%*) 1.25±0.2 (68%*) 0.9±0.2 (77%*)

a) b) c) d)

79

the area available for the fluid seepage and hence forcing Ca2+ ions to adsorb onto the

negatively charged clays.

Table 4.4: ICP results of the filtrate collected following API LTLP to determine the Ca and Fe content.

Those NPs interact with the formation and eventually plug the pore either internally

or ‎externally, preferably. If, on the other hand, it blocks the pore channel,

formation ‎damage may occur and oil and gas production will be interrupted. ICP results,

together ‎with the fact that filter cakes were thin, suggest only external plugging took

place.

In clays, often Na+

or Ca2+

, are too large to be accommodated in the interior of the

lattice and therefore may be easily exchanged by other cations when available in

solution (Deriszadeh,2012). NPs either ex-situ or in-situ could exchange cations with

clays and locate on the exterior surface and near the pore openings are kinetically more

accessible than the interior pore wall. Therefore NPs bridge across pore throats to form

the external mud cake immediately.

4.2.4 HTHP Filtration

The HTHP API filtration test simulates drilling in deep formation, where both the

temperature and the pressure of formation and the drilling fluid may reach high values.

High temperatures may alter the size, identity and surface morphology of the Fe(OH)3

NPs (Agarwal et al., 2009; Balek and Šubrt, 1995). This may ultimately lead to reducing

the NPs effectiveness. Generally, nanoparticle aggregation and the formation of large

irregular particles can be captured by visual analysis of the filter cake. The results

Filtrate Samples of

Drilling fluid (DF)

mg (In total volumes)

Ca Content Fe Content

Without NPs

With 1 wt % in-situ NPs

478

0.87

Nil

Nil

80

shown in Table 4.5 and Figure 4.11 provide details on fluid loss reduction and sealing

potential of the filter cake for invert emulsion muds. Based on the original DF, fluid loss

over a period of 30 min decreased by 24% for the drilling fluid containing 1.6 wt% LCM

only, while it decreased by 53% for the drilling fluid containing 1 wt% ex-situ prepared

Fe(OH)3 NPs and 61% for the drilling fluid containing the in-situ prepared Fe(OH)3 NPs.

In this experiment, both ex-situ and in-situ NP-based drilling fluids contained 1.6 wt%

Gilsonite LCM. The better dispersed in-situ prepared NPs exhibited lower mud cake

thickness than ex-situ prepared NPs as shown in Figure 4.11.

Generally, the higher loss of the drilling mud with and without NPs or LCMs when

compared with low temperatures is attributed to the lower viscosity of the fluid at 177oC.

Cake thickness is proportional to filtration loss (ASME, 2005). As the mud is not being

circulated, the filter cake grows undisturbed with the filtrate rate. Table 4.6 shows fluid

loss reduction and mud cake thickness under API HTHP conditions. As temperature and

pressure go up, lower mud cake thickness in presence of Fe(OH)3 NPs is obtained.

Similar observations were reported by Javeri et al.(2011) and Paiaman and Al-

Anazi,(2008). It is true that in the absence of NPs and LCMs filter cake displayed low

thickness, but it should be noted that the filter cake was not effective towards filtrate

reduction. NPs increase the tortuous flow path and travel time of the fluid to pass

through the filter cake and lower the fluid loss.

In the presence of NPs filtration rate became slow probably due to the high level

of interaction between ‎the NPs, Gilsonite and clays, which led to effective bridging even

at high temperature. ‎In addition, and as noted by Aston et al. (2002), water droplets with

sizes≥ 5.5 μm tend ‎to bridge the 3 μm pores on the filter paper. At the high

temperatures encountered in this ‎experiment, water pools may coalesce to form larger

droplets. At the low concentration ‎of NPs it is more likely that the particles interacted

with the rest of the mud constituent ‎rather than merely aggregating. Moreover,

temperature affects clays by changing the orientation of the adsorbed water pools in the

clay matrix. The rigid bonding of water may decrease dispersion of the clay and form a

more porous filter cake which allows a greater filtrate flow at high temperature (Fisk and

Jamison, 1989).

81

Table 4.5: HTHP filtration property of different drilling fluid samples.

*Fluid loss reduction,%

Figure 4.11: Filter cakes obtained following API HTHP tests on invert emulsion drilling

fluids with and without Fe(OH)3 NPs and Gilsonite LCMs.

Samples

Types

Time

(min)

HTHP Fluid Loss (mL)

DF DF +LCM DF with 1 wt%

ex-situ NPs

DF with 1 wt%

in-situ NPs

90:10 (v/v) Oil: Water

7.5 9±0.1

6.2±0.2

2±0.2

0

30 19±0.1 14.4±0.1

(24%*)

9±0.1 (53%*) 7.5±0.2 (61%*)

Cake thickness, mm 1.7 7.3 2.7 1.3

LCM+NPs (In-situ)

Filter cake

LCM+NPs (Ex-situ)

Filter cake

LCM

Filter cake

No LCM or NPs

Filter cake

Thickness= 1.3 mm Thickness= 2.7 mm Thickness= 7.3 mm Thickness= 1.7 mm

82

Table 4.6: Effect of operating conditions of API filtration test on mud cake thickness with and without Fe(OH)3 NPs and Gilsonite LCMs.

*Thickness improvement, x times compared to DF

In a similar manner, one run with NPs in the absence of the Gilsonite LCM was

performed. The results are shown in Table 4.7. The data on fluid loss at 30 min show

that in the absence of LCMs, the NPs are performing better. Based on the original DF,

fluid loss over a period of 30 min decreased by 79% for the drilling fluid containing 1

wt% ex-situ prepared Fe(OH)3 NPs, while it decreased by 86% for the drilling fluid

containing 1 wt% in-situ prepared Fe(OH)3 NPs. This observation can be attributed to

the fact in the absence of LCMs there seems to be higher interaction between the NPs

and the clays, which resulted in better sealing and more effective filter cake.

Table 4.7: HTHP fluid loss of different drilling fluid samples in the presence and absence of Fe(OH)3NPs. No LCMs were added. Cake Thickness at 30 min.

*Fluid loss reduction, %.

Temperature and Pressure

Mud cake thickness (mm) DF DF +LCM DF with 1 wt%

ex-situ NPs+LCM

DF with 1 wt% in-

situ NPs+LCM

25 °C,100 Psi 0.31 0.76 (2.5*) 0.52 (1.7*) 0.44 (1.4*)

177 °C,500 Psi 1.7 7.3 (4.3*) 2.7(1.6*) 1.3 (0.8*)

Samples

Types

Time

(min)

HTHP Fluid Loss (mL)

DF DF with 1 wt%

ex-situ NPs

DF with 1 wt% in-

situ NPs

90:10 (v/v) Oil: Water

7.5 9±0.1

0 0

30 19±0.1 4±0.1 (79%*) 2.7±0.2 (86%*)

Cake thickness, mm 1.7 2.1 1.1

83

Figure 4.12: Mud cakes obtained following API HTHP tests on invert emulsion

drilling fluids with in-house prepared NPs only. No LCMs added.

LCMs seem to consume NPs, which would otherwise interact with the mud cake to a

better extent than the interaction of the NP-LCM combination. The better dispersed in-

situ prepared NPs exhibited lower mud cake thickness than ex-situ prepared NPs as

shown in Figure 4.12. When subjected to high temperatures, NPs are likely to maintain

stability and dispersion in the water-in-oil ‎emulsions. Agarwal et al. (2009) used nano

CuO with 23-37 nm diameters and nano ‎alumina with 40-50 nm diameters in invert

emulsion drilling fluids and showed that ‎drilling fluids maintain their stability even at

175oC. It appears that when NPs are mixed with drilling fluid, clay suspensions may

bind with NPs ‎resulting space-filled structure. A sol-gel formation may be induced,

which finally ‎blocks fluid flow through the filter media, upon filtration. Addition of in-

house prepared Fe(OH)3 NPs, increases the ionic ‎strength of the fluid, due to the

formation of NaCl by-product, which causes stronger interaction with the clays during

HTHP filtration (Agarwal et al., 2009). As discussed earlier, the elimination ‎of spurt loss

observed in these experiments may reduce formation damage, and thin mud cakes

could ‎possibly reduce stuck pipe problems (Chilingarian and Vorabutr, 1983).

4.2.5 Effect of high shear on fluid loss control

High degree of mixing and shearing of the drilling fluid is essential to form NP-based

drilling fluid using the in-house preparation technique, as described earlier. This step is

important whether the particles are prepared in-situ or ex-situ. Shearing device may

NP-free DF Mud cake

Ex-situ NPs Mud cake

In-situ NPs Mud cake

Thickness 1.1 mm

Thickness 1.7 mm

Thickness 2.1 mm

84

significantly increase the dispersed phase fraction and dampens coalescence by

breaking agglomerated particles (Amanullah, 2011). Hamilton beach three blade high

speed mixer was used in addition to vigorous agitation of fluid during preparation. This

inexpensive equipment is used mostly in food processing. High-shear mixers provide

rapid micro-mixing and emulsification. Providing no blending displayed higher fluid loss

when compared with blending at 2500 rpm using Hemilton beach blender for same DF

with and without 1 wt% in-situ Fe(OH)3 NPs, as shown in Table 4.8. Figure 4.13 shows

that the mud cake collected following the filtration of unblended drilling fluid is full of

precipitates, agglomerates and ‘fish eyes’ as highlighted by the circles, while the one

collected from a blended sample is much more smooth and does not show

agglomerate. Very high mixing rates result in smaller particles in the mud as it serves

formation of very small water pools, in the case of in-situ prepared NPs, and minimizing

particle aggregation during the formation. Same effect was also observed during the

addition of ex-situ prepared NPs. It was found by Altun and Serpen (2005) that

variations in the mixing speed have important effects on fluid loss property and higher

mixing speeds yielded lower filtration loss. In a similar study, Newman et al. (2010)

showed that properties of drilling fluid were significantly affected when mechanical

mixing is applied. It was also understood that to obtain smaller droplets of uniform size

in water-in-oil emulsion, energy must be applied in the form of shear.

Table 4.8: Effect of shearing effect on LTLP fluid loss control in the presence and absence of NPs.

Samples Types

LTLP Fluid Loss (mL/30 min)

Unblended DF

(No NPs)

2500 rpm Blending

DF (No NPs)

2500 rpm Blending 1 wt%

in-situ NPs +DF

90:10 (v/v) Oil: Water 8±0.1 3.96±0.2 1.25±0.2

85

Figure 4.13: Quality of unblended and blended mud cake.

4.2.6 Effect of presence of organophillic clays on fluid loss

Table 4.9 shows the effect of varying the composition of organophillic clays from 12 to

15 kg/m3 in the presence and absence of 1 wt% Fe(OH)3 NPs.

Table 4.9: Effect of organophillic clays on LTLP fluid loss control.

As evident from the table, increasing clays concentration improves loss prevention. It

should be noted that clay content cannot be indefinitely increased. Solids content of the

drilling fluid is one of factors that causes formation damage and decreases rate of

penetration ROP (Newman et al., 2009). Solids are added to fulfill the functional tasks

of the mud such as increasing mud density, viscosity and fluid loss control. The higher

the amount of total solid in the drilling fluid the lower the rate of penetration, which in

turn increases rig days and reduces productivity index. Unlike the Gilsonite LCM,

Samples

Types

Amount of organophillic

clays used in DF

LTLP Fluid Loss (mL/30 min)

DF without NPs DF+ with 1 wt%

in-situ Fe(OH)3

NPs 90:10 (v/v) Oil: Water

12 kg/m3 4.9±0.1 2.3±0.1

15 kg/m3 3.96±0.2 1.25±0.2

Blended NP-based Mud cake

86

increasing the content of clays in presence of NPs increased fluid loss prevention, since

more clays are available to form the mud cake. NPs would still be performing their role

as bridging particles and will have higher surface to communicate with in the presence

of more clays.

A major outcome of the current study is that low NPs concentration can

significantly reduce fluid loss. In events were high solid concentration is not desirable,

for example due to the need to keep fluid density to a minimum, NPs can replace clay

additive. Addition of low concentrations of NPs did not have any effect on the mud

density, as will be detailed later.

4.2.7 Effect of Oil: Water ratio on fluid loss

Filtration behavior of emulsified oil is strongly influenced by oil/water ratio, additive

chemistry and concentration (Aston et al., 2002). Two formulations; namely 90:10 (v/v)

and 80:20 (v/v) oil: water mixes, were tested in the presence and absence of in-house

prepared NPs. This experiment is particularly relevant to in-house prepared NPs, since

aqueous precursors are added. The results shown in Table 4.10 reflect a decrease in

filtrate volume in the presence of Gilsonite LCMs and Fe(OH)3 NPs. Table 4.11 shows

the same trend in presence of NPs and absence of LCMs.

Increasing the water content from 10 to 20 percent by volume caused the fluid

loss to decrease 26% and 25% for drilling fluid control samples and drilling fluid

containing Gilsonite LCM, respectively. Addition of NPs, again, decreases the fluid loss

to 44% and 10% for ex-situ and in-situ method, respectively, due to the changed water

content from 10 to 20 percent by volume. The reduction of fluid loss was dramatic in the

case of ex-situ prepared NPs. This may suggest that extra water pools were originally

needed to disperse better the particles. In-situ prepared NPs are more readily dispersed

in the 10 percent water content. Therefore, in 20 percent water content, the fluid loss

reduction was not varied too much. Higher water content may increase collision among

water pools, which, in turn, may lead to more particle agglomeration (Husein and

Nassar, 2010). This, in a way, decreases the effectiveness of the NPs. Nevertheless,

one should not ignore the lower interaction between the organophilic clays constituting

87

the filter cake and the drilling fluid as the water content increases. Filtration rates

through hydrophobic membranes results in much lower permeate flux (Deriszadeh et

al., 2010). Aston et al. (2002) found the ‎similar trends and proposed major savings can

be attained by decreasing the oil to water ratio, while attaining more loss prevention.

Table 4.10: Effect of Oil: Water ratio on Fluid loss control in presence and absence of

LCM and in-house prepared Fe(OH)3 NPs.

Table 4.11: Effect of Oil: Water ratio on fluid loss control in presence and absence of

in-house prepared NPs. No LCMs added.

4.2.8 Rheology behavior of NP-based fluid

Drilling fluid with good pumpability exhibit lower viscosity at high shear rate and higher

viscosity at lower shear rate. This property of drilling mud is used widely where high

viscosities are required during tripping operation and low viscosities during drilling

operation to clean the cuttings from the bottom of the hole (Chenevert and Sharma,

2009; Fraser et al., 2003). The plot of apparent viscosity and shear rate as shown in

Figure 4.14 resembles the non-linearity of the curves at low shear rates and approach

Samples Types

Time (min)

LTLP Fluid Loss (mL)

DF DF+ LCM DF+LCM+ 1 wt% ex-situ NPs

DF+LCM+ 1 wt% in-situ NPs

90:10 (v/v) Oil: Water

7.5 2.0±0.2 1.4±0.2 0.2±0.2 0

30 3.96±0.2 3.6±0.1 1.10±0.1 0.5±0.2

80:20 (v/v) Oil: Water

7.5 1.0±0.2 1.0±0.2 0 0

30 2.9±0.1 2.7±0.2 0.62±0.1 0.45±0.1

Samples Types

Time (min)

LTLP Fluid Loss (mL)

DF DF with 1 wt%

ex-situ NPs

DF with 1 wt% in-situ NPs

90:10 (v/v) Oil: Water

7.5 2.0±0.2 0.15±0.1 0

30 3.96±0.2 1.25±0.2 0.9±0.2

80:20 (v/v) Oil: Water

7.5 1.0±0.2 0 0

30 2.9±0.1 0.8±0.1 0.5±0.2

88

linearity at high shear rates. The fact that addition of NPs created a slight change in the

rheology supports the theory that NPs behavior is governed by NPs grain boundary and

surface area/unit mass (Amanullah et al., 2011; Srivatsa, 2010). Although the addition

of small concentration of NPs is not sufficient to cause a significant rheology changes in

the system compared to the drilling fluid and drilling fluid with LCM only, particle size,

nature of particle surface, surfactants, pH value and particle interaction forces may play

significant role in altering the viscosity (Agarwal et al.,2009). The minor effect of NPs on

viscosity is attributed to the low concentrations employed in this study. Abu Tarboush

and Husein (2012) noted that NPs may increase the viscosity of heavy oil by bridging

between asphaltene molecules and aggregates.

The results are also highly dependent on the hydroxyl group (OH-) on the surface

of the NPs may lead to NPs agglomeration in an organic solution leading to a higher

mass of selective physisorption of organic clay suspension on the NP-free surface,

which may reduce the fluid viscosity slightly (Srivatsa, 2010). A small amount of NPs

exhibit stable rheological properties. Fluid with high viscosity may cause excessive

pumping pressure and decrease rate of drilling. Therefore, it is an important issue to

design a suitable fluid rheology. Lee et al. (2009), who investigated the application of

NPs for maintaining viscosity of drilling fluids at high temperature and high pressure,

reported that the rheological behavior may depend on the particle type, size,

concentration and inter-particle distance of NPs within the fluid. It was also reported that

adding very small amount of mixed metal oxide did not change fluid rheological

properties. It was shown that with an increase in temperature, the viscosity of drilling

fluid containing 0.05 wt% cobalt NPs unchanged at 100 cP and remained stable.

Therefore, potential application of NPs is to use them to stabilize in water-in-oil

emulsion where NPs (solid/semi solid) dispersed in clays and electrolyte (NaCl salt)

produced during the NP-based fluid formulation also work as a bridging material

between the platelets of organophillic clays to form gel structure. The rheological

properties of the in-house NP-based drilling fluid thus could suitably fulfill the drilling

requirements. The comparison of the gel strength behavior of the drilling fluid, drilling

fluid with LCM, drilling fluid with LCM and NPs together and NPs only are shown in

89

Figure 4.15. The gel strength property of the NP-based drilling fluid compared to the

progressive type gel strength of DF and DF+LCM also demonstrates superior functional

behavior of NP-based drilling fluid. Similar observation was reported by Amanullah et al.

(2011). Very high gel strength values are practically undesirable because they retard

the separation of drilled cuttings at the surface and also raise the pressure required to

re-establish circulation after changing bits. Furthermore, when pulling pipe, high gel

strength may reduce the pressure of the mud column nearby the bit. If the reduction in

pressure exceeds the differential pressure between the mud and the formation fluids,

the fluid will enter the hole and cause a blow-out (ASTM, 2005; Amanullah et al., 2011;

Chilingarian and Vorabutr, 1983).

Figure 4.14: Rheological behavior of drilling fluid containing a) LCM together with in-house prepared 1 wt% Fe(OH)3 NPs, b) 1 wt% Fe(OH)3 NPs no LCMs.

From Figure 4.16 and Figure 4.17 we observe the time dependent rheological

and gel strength behavior of the drilling fluid. The measurement was done immediately

after the preparation and also after 1 month. After 4 weeks the fluid was found

compliant with all specification for re-use. Analyses of the rheological profiles of the

drilling fluids shown in Figure 4.16 indicate no significant changes of the viscous profile

b) a)

90

of the NP-based fluid. The NP-based fluid immediately after preparation and static aging

after 1 month demonstrate that the short as well as long term stability exist in the NP-

based fluid. The 10 seconds and 10 minutes gel strength shown in Figure 4.17 also

demonstrate the short and long term stability of the NP-based fluid to fulfill its functional

task during drilling operation.

91

Figure 4.15: Gel strength behavior of drilling fluid a) with LCM and NPs together ex-situ and in-situ method b) in the absence of LCM,with NPs only ex-situ and in-situ method.

b)

a)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

DF DF+In-situNPs

DF+Ex-situ NPs

Gel Strenth (10 sec)

Gel Strenth (10 min)

lb /

10

0 f

t²

92

Figure 4.16: Shelf life of drilling fluid samples in terms of rheology behavior.

Figure 4.17: Aging effect of drilling fluid samples in terms of gel strength

behavior.

93

4.2.9 Drilling fluid density and pH

Mud density is one of the important drilling fluid properties, because it balances and

controls formation pressure and wellbore stability (Chilingarian and Vorabutr, 1983). A

mud density of 0.93 g/cm3 of the 90:10 (V/V) oil/water invert emulsion was found to be

constant for all samples with and without NPs as shown in Table 4.12. The addition of

NPs did not increase the mud weight given the fact that their concentration was low and

also due to the electrochemical behavior of NPs with clays. As discussed above, this is

advantageous since it is one way of improving fluid filtration properties while maitaining

the same mud density. Similar advantage of NPs was exploited to increase mud density

while maintaining low mud visocistiy.

A pH level of 12.5 was also found in all samples as also shown in Table 4.12,

even with NPs addition. It should be noted with the fact that NaOH was added at the

stoichiometric amount and NaCl was the reaction by-product, no changes in the pH of

the aqueous pools is expected. Generally, changes in the pH of the water pools of invert

emulsions could lead to unstability of the colloidal system by neutralizing charged

surfaces at the water/oil interface or particles.

Table 4.12: Density and pH values of drilling fluid in the presence and absence of LCM

and in-house prepared Fe(OH)3 NPs.

4.2.10 Drilling fluid lubricity

Even if a drilling fluid successfully meets all of the requirements, there is no guarantee

that the rate of penetration will be acceptable, since poor lubricity and high friction and

drag increase pipe sticking and drilling cycle (Amanullah et al., 2011). It needs to

overcome frictional forces which is very much encountered during all stages of well

Samples

Types

Properties

Test samples

DF DF +LCM DF+LCM with 1

wt% ex-situ NPs

DF+LCM with 1

wt% in-situ NPs

90:10 (v/v) Oil: Water

Density (g/cm3)

0.93±0.02 0.93±0.02 0.93±0.02 0.93±0.02

pH 12.5 12.5 12.5 12.5

94

construction; including drilling, completion and maintenance. Friction originates from the

rotation and/or sliding of a pipe inside the well in contact with either the wellbore (metal-

to-rock) or the casing (metal-to-metal). These forces hinder directional and extended

reach drilling by creating excessive torque and drag (Amanullah et al., 2011; Hoskins,

2010). Excessive torque and drag in highly directional and extended-reach wells can

exceed the mechanical limits of the drilling equipment, which may expedite wear and

tear of down hole tools and equipment and thereby limit production. These problems

can be minimized by using drilling fluid with high capabilities of lubricating the different

components. In fact, the switch from water-based to oil-based or invert emulsion muds,

despite the increase in cost, was originally proposed to help improving lubricity

(Kercheville et al.,1986).Friction dissipates energy and causes wear resulting in damage

to the equipment. The way to ensure that frictional effects are minimized is through

proper lubrication. In carrying out this function, lubricants create a lubricant film on

surfaces of moving parts.

The effect of the in-house prepared NPs on the lubricity of the invert emulsion

drilling fluid considered in this study was measured by evaluating the coefficient of

friction, as detailed in the experimental work. The hypothesis was that under the

conditions of load and temperature resulting from the contacting surfaces, these NPs

may furnish a thin film of lubricant layer on the contacting surfaces leading to reduced

friction between the surfaces. These NPs may act as nano-bearings and contribute

increasing the lubricity. Table 4.13 displays values for the coefficient of friction (CoF)

and the accompanying reduction in torque and drag in the presence of 1 wt% in-house

prepared Fe(OH)3 NPs.

Table 4.13: Co-efficient of friction (CoF) of drilling mud samples.

NPs and

conc. used

Coefficient of friction % torque reduction

DF without NPs

(control)

DF with ex-

situ NPs

DF with in-

situ NPs

DF with ex-

situ NPs

DF with in-

situ NPs

Fe(OH)3

(1 wt%)

0.095±0.002

0.081±0.004

0.039±0.002

14.73%

58.94%

95

It appears that in-situ prepared NPs disperse better and communicate better with the

mother drilling fluid as opposed to the ex-situ prepared ones. Therefore, in-situ prepared

NPs may carry a proportion of the load benefiting the improvement of antiwear property

more than NPs prepared ex-situ. Thus using tailormade NPs in drilling fluid can reduce

coefficient of friction and substantially increase lubricity. Improvement in lubricity

reduces energy consumption, which, in turn, increases profitability.

Oil-based drilling fluids have the inherent advantage of significantly lower

coefficients of friction (CoF). The typical CoF for an oil-based drilling fluid is 0.10 or less

(metal to metal) (Chang et al., 2011). In comparison, water has a CoF of 0.34 and the

CoF of water-base drilling fluids typically ranges between 0.2 and 0.5 (Chang et al.,

2011). Comparing between the typical oil based mud and NP-containing mud the friction

mechanism is most likely a transfer of NPs to the counterface. This suggests that NPs

in the contact zone act like ball bearings in the interface between the two surfaces. The

small size allows the particles to penetrate into the surface and van der Waals forces

ensure that the particles adhere to the surfaces. Regular lubricants, or oil as continuous

phase, in drilling fluid can only form a single oil film (Kostic, 2010; Mosleh et al., 2009;

Malshe et al., 2008), whereas NPs in drilling fluid can create an additional ball bearings

action leading to better lubrication effect. Nonetheless, iron oxide/hydroxide

nanoparticles act as a lubricious material (Reed, 2008). Sodium salts, e.g. NaCl salt

formed as a by-product during the Fe(OH)3 NP-based fluid formulation, may act as a

lubricant as per some literature (Scoggins and Ke, 2011; Ke and Foxenberg, 2010).

Table 4.14 shows that these side products, in fact, slightly increase the coefficient of

friction. Therefore, the increase in lubricity observed when iron-based NP-drilling fluids

are used can entirely be attributed to the nanoparticles only.

96

Table 4.14: Coefficient of friction (CoF) and % torque reduction in the presence and absence of 1 wt% NaCl salt in the invert emulsion drilling fluid.

Nanosized particles are much more readily dispersible than micron-sized ones (Canter,

2009). When dispersed in a drilling fluid, minimum agglomeration and settling occur and

a stable suspension form. The stable dispersion is also supported by the presence of

surfactant molecules. Both in-situ and ex-situ prepared NPs are so small in size that a

stable colloidal dispersion in drilling fluids can be achieved, which probably avoid the

undesired precipitation caused by gravitation. With the formation of a stable well-

proportioned dispersion through proper method, NPs are more prone to be trapped in

the rubbing surfaces due to its excessive surface energy. Besides, dispersed NPs are

deposited on the friction surface, trapped NPs at the interface and finally roughness of

the surface is reduced by its polishing effect (Wu et al., 2007; Mosleh et al., 2009).

Moreover, as the NPs tend to disperse uniformly, a more uniform contact stress

between the contacting surfaces may result (Chang and Friedrich, 2010). Moshkovith et

al. (2007) studied the lubricity properties of IF-WS2 and reported that dispersion impacts

the lubricity performance as the dispersed NPs possess solid lubrication properties due

to its stability. It was also found that the aggregate size of the NPs depend on the

mixing time. The in-house prepared NPs in this work can be engineered to have specific

size ranges so that they can find their way into intricate spaces and maintain lamellar

structure. It is, therefore, speculated that the coefficient of friction reduction is due to the

surface boundary films provided by NPs that slide easily over one another like ball

bearings. Similar findings have been reported in the literature on the effect of dispersing

carbon and metallic-based NPs on tribological performance of lubricating oils (Zhang et

al., 2009; Abdullah, 2008; Malshe et al., 2008; Verma et al., 2008). Specifically, a

reduction in the coefficient of friction by over 25 percent was observed when adding

nickel-based NPs to lubricants (Kostic, 2010).

Salt and conc.

Coefficient of friction

% torque reduction DF without salt

(control sample)

DF with salt

NaCl (1 wt%) 0.0980±0.002 0.100±0.004

-2%

97

In addition to reducing torque, higher lubricity also lowers the incidence of stuck

pipe, which can significantly lower drilling efficiency. Estimates by exploration

companies showed that stuck pipe while drilling costs more than $250 million each year

(Q’Max Technical Bulletin #7). Minimizing friction and the ability to transfer the weight to

the bit are very important factors in drilling highly deviated extended reach and

horizontal wells. Moreover, reduction in torque in the presence of NPs imply higher

extended reach wells at a given torque and load on bit.

From the aforementioned discussion it can be concluded that the ability of NPs to

increase lubricity depends on the following features:

1. NPs can adsorb physically on any metal surface due to van der Waals forces.

2. The size of the NPs is so small that they can easily enter a macroscopic sliding

contact.

3. The lubrication effect can be generated by the chemical nature of surfactant as

described by Yang et al. (2012) and NPs altogether or NPs alone. Dispersed

nanoparticles can help in reducing the agglomeration at the interface and

improving the co-efficient of friction. The role of surfactant molecules is to

improve the dispersion quality and stability of the NPs, since the level of

improvement is measured relative to a control sample that contains the same

amount of surfactant.

4. Coefficient of friction significantly reduced by NPs alone and the by-product salt

did not have a significant impact on lubricity.

4.2.11 Preparation and performance evaluation of Fe(OH)3 NPs in invert emulsion

drilling fluids provided by different suppliers

Three invert emulsion muds were obtained from three different suppliers having same

oil:water ratio (90:10 v/v). These drilling fluids mainly differ in terms of the amount of

organophillic clays and nature of emulsifiers. It is important to note that the nature of

drilling fluids emulsifiers was not studied separately in this current study. The in-house

in-situ and ex-situ NPs preparation techniques were employed to form Fe(OH)3 NPs and

the performance of the NP-based drilling fluids was evaluated. For a drilling fluid;

98

filtration, rheology, density, pH need to be suitable to fulfill the drilling requirements. The

results in Table 4.15 compare the density, pH and the API LTLP fluid loss at 30 min for

samples with and without Fe(OH)3 NPs. The density and the pH of the samples

remained constant, while fluid loss was decreased significantly. Moreover, the in-situ

prepared NPs displayed better performance and reduced fluid loss to a higher extent.

This fact can be attributed to better dispersion and communication with the rest of the

drilling fluid constituents, as discussed earlier.

Table 4.15: Effect of ex situ and in situ prepared Fe(OH)3 NPs on the performance of

three different invert emulsion samples of drilling fluids provided by three different suppliers. Concentration of NPs 1 wt%, composition of invert emulsion: (90:10) oil:water (v/v).

NPs were found to be suitable for use in drilling fluids due to its functional

characteristic of maintaining low viscosity and expected to minimize the drilling

problems. In-situ control of viscosity of drilling fluids in deep well bores is currently

Samples Density pH API LPLT fluid loss

(g/mL)

(mL/30 min)

Supplier A

DF 0.93±0.02 12.5 3.9±0.2

DF+LCM 0.93±0.02 12.5 3.6±0.1

DF+LCM+Ex situ NPs 0.93±0.02 12.5 1.1±0.1

DF+LCM+In situ NPs 0.93±0.02 12.5 0.5±0.2

Supplier B DF 0.93±0.02 12.5 16.5±0.3

DF+LCM 0.93±0.01 12.5 12.7±0.4

DF+LCM+Ex situ NPs 0.93±0.02 12.5 7.5±0.2

DF+LCM+In situ NPs 0.93±0.02 12.5 6.5±0.1

Supplier C DF 0.90±0.02 12.5 1.2±0.2

DF+LCM 0.90±0.02 12.5 1.0±0.1

DF+LCM+Ex situ NPs 0.90±0.01 12.5 0.5±0.1

DF+LCM+In situ NPs 0.90±0.01 12.5 0.1±0.1

99

limited (Lee et al., 2009). During operation, and as drill cuts suspend into the drilling

fluid, viscosity increases and alters the rheology in the subterranean wells (Adekomaya

and Olafuyi 2011; Herzhaft et al., 2006). Therefore, a gradual tuning of the rhelogical

properties of the drilling fluids is required to maintain good performance. As noticed

from a previous experiment, NPs, especially in-situ prepared ones, generally reduce the

viscosity of drilling fluids due to their electrochemical behavior with clays. The

decreased in viscosity with the addition of NPs may provide the in-situ

tunability/controllability of the fluid viscosity during drilling in subterranean wells. The

rheological properties of the drilling fluids obtained from different suppliers were

evaluated in the presence and absence of the in-house prepared NPs. Apparent

viscosities at 600 rpm were plotted for comparison in Figure 4.18. The results show

consistent decrease in the apparent viscosity in the presence of in-house prepared NPs.

This also confirms the fact that in-situ prepared NPs better interact with the drilling fluid

displaying more reduction in the apparent viscosity. Similar observation was reported by

Abdo and Haneef (2010) when using Montmorillonite NPs compare with regular

commercial bentonite particles.

Figure 4.18: Apparent viscosity at 600 rpm of 3 invert emulsion drilling fluids provided

by different supplies in the presence and absence of 1 wt% Fe(OH)3 NPs. Composition of invert emulsion: (90:10) oil:water (v/v).‎

Appare

nt vis

cosity (

cP

) at

600 r

pm

100

The data in Figure 4.18 suggest that in the presence of NPs lower the viscosity can

reduce the pumping power requirement without compromising the carrying capacity of

the drilling fluid to transport and drop off cuttings efficiently.

4.2.12 Performance of Fe(OH)3 NPs in water based mud (WBM)

Fluid invasion into porous formations can damage reservoirs and reduce productivity by

blocking hydrocarbon exit flow paths or causing formation damage. Fluid penetration

while drilling using WBM can lead to shale formation swelling and, subsequently, well

bore instability (Sensoy et al, 2009). Currently most fluid loss additives in WBMs have

formulations based on bentonite clays, lignite, asphaltite and organic polymers

(Kosynkin et al., 2011; Moore et al., 1974). In terms of environmental and economical

considerations, WBM would be preferred if the interaction between the fluid and the

shale could be controlled.

Water based muds are suitable only for relatively low temperature and pressure

drilling operation (Agarwal et al., 2009). Xanthan gum is a rheological modifier that

typically used in WBM to increase viscosity and improve dispersion stability. In the

current experiment, few mg of surfactants (Dioctyl sodium sulfosuccinate) were used to

provide adequate stability of NPs. Table 4.16 displays fluid loss results after 30 minutes

for WBM containing 0.60 wt% Fe(OH)3 NPs formed ex-situ and in-situ methods.

Table 4.16: API LTLP WBM fluid loss at 30 min in the presence and absence of 0.60

wt% in-house prepared Fe(OH)3 NPs.

*Fluid loss reduction, %. Comparison of the fluid loss behavior of invert emulsion and water based fluids ‎shows

that emulsion fluids exhibited a better fluid loss control. The types of clays used in both

muds were different. Organophillic clays (surface modified bentonite clays) which are

normally treated with amines in order to make them oil dispersible and largely used in

invert emulsion mud formulation, whereas regular unmodified bentonite clays are used

Water based DF Water based DF+ ex-

situ NPs Water based DF+ in-

situ NPs

8.8±0.6 8.0±0.2 (9%*) 6.3±0.2 (28.4%*)

101

in water based mud. Due to surface modification, organophillic clays extended the clay

properties in terms of rheology, carrying capacity, fluid loss control, etc. (Paiva et

al.,2008; Juppe et al.,2003).

Comparing the results for WBM alone to the samples containing the NPs it can

be concluded that the NPs probably contributed to the mud cake and resulted in an

overall reduction in fluid loss. It can also be inferred that the in-situ prepared NPs

interacted better with the mud cake, again, probably as a result of better dispersion in

the original mud. Cei et al. (2011) used commercial and non-modified silica NPs with

sizes ranging from 5 nm to 22 nm in a WBM and showed that concentrations ≥10 wt%

NPs reduced fluid loss by 25-30%. Similarly Kosynkin et al. (2011) showed that using

graphene oxide (GO), a combination of large flake GO and powdered GO in a 3:1 ratio

to perform, in WBM with a concentration of 0.2% (w/w) based on carbon content

resulted in 15.3% fluid loss reduction. Amanullah et al. (2011) used 0.14% (w/w) silica

NPs in WBM formulation along with different additives including polymeric additives.

Their study showed that the API fluid loss behavior of nano-based fluid showed similar

fluid loss as the original water based mud and did not improve the fluid loss reduction

performance.

It was observed by Flask and Jamison et al. (1989) that fluid penetration into the

pore space was controlled by the size of the aggregates in the drilling fluid relative to

the pore size of the formation. The size of aggregates in the drilling fluid controlled initial

bridging and formation of the filter cake. There are few guidelines used in oil and gas

industry to choose particle size in order to optimize their role as bridging materials.

Among the guidelines, Suri and Sharma (2004) showed that in order to form bridges,

particle sizes should not be larger than one third of the pore throat size. This implies

that wide range of NPs used in the current work makes it more effective in plugging

nano to micro pore throat. Of course, NPs plugged primarily the ones that fit that size

and then in some cases aggregate together to plug the big pore size. Similar

observation was reported by Sensoy et al. (2009).

Overall, WBM loss reduction contributes to reducing wellbore instability

problems.

102

4.2.13 Toxicity evaluation Fe(OH)3 samples

In recent years, the usage of inorganic NPs has increased exponentially for a variety of

applications (Buzea et al., 2007). In this context, it is necessary to assess the

environmental and biological risks of NPs used in this work. It should be noted,

nevertheless, one of the advantages of the in-house methods developed in this work is

the fact that the resultant NPs is contained within a liquid mixture rather than being

airborne. This minimizes inhalation and reduces health risk significantly. In Alberta, on-

site disposal of the drilling waste is allowed, provided that criteria imposed by the

Alberta Energy and Utilities Board (AEUB) are met (AEUB, G50, 1996). To be rated

essentially non-toxic, an EC50 of at least 75% of the original waste concentration must

be obtained. This threshold limit is compatible with pass/fail results in bioassay. It is

therefore useful to have an estimate of the ferric hydroxide NPs concentration that could

cause bioassay failure, i.e EC50 (15 min)< 75% (Ashworh and Walker,2006). Iron

hydroxide NPs were broadly used and selected for their low toxicity. Commercial Fe3O4

NPs exhibited toxicity at 45% of initial concentration (García et al., 2010). Iron hydroxide

NPs prepared in this work only showed toxicity when NPs concentration increased to

concentrations > 15 % by volume, as per Table 4.17. Therefore, it is safe to say that at

the concentrations employed in this work, 1 wt%, Fe(OH)3 NPs is nontoxic.

Table 4.17: Microtox bioassay of Fe(OH)3 NPs.

Test Fe(OH)3

Concentration

(V/V)

Results Interpretation (AEUB,G-50)

Pass/Fail; (Pass if EC50>75%)

EC50 50% 18.7 Fail

EC50 35% 22.5 Fail

EC50 35% 37.5 Fail

EC50 20% 64.8 Fail

EC50 15% 100 Pass

103

4.3 CaCO3 Nanoparticles (NPs) Characterization

As described in Section 3.2 of the experimental work, two methods were adopted to

form CaCO3 NPs; namely ex-situ and in-situ. Moreover, in-situ CaCO3 was prepared

from salt precursors as well as CO2 injection.

4.3.1 X-ray diffraction analysis

The structure prepared by precipitating CaCO3 NPs ex-situ starting from aqueous

Ca(NO3)2 and Na2CO3 precursors is shown in the X-ray diffraction (XRD) patterns of

Figure 4.19. The X-ray diffraction pattern shows that there is evidence of distinct peaks

which would be expected from a crystalline material.

Calcium carbonate has three crystal structures; calcite, aragonite and vaterite

(Kabalah-Amitai et al., 2013). Typically, calcium carbonate exists as calcite, which is the

most thermodynamically stable structure at ambient temperatures and pressures (Lee

et al., 2001), whereas vaterite is most unstable. Aragonite is less stable than calcite

and commonly found in marine organisms (Tai and Chen, 2008). The data obtained

from the X-ray diffraction patterns in Figure 4.19 demonstrates the crystalline nature of

the sample under analysis. The X-ray diffraction pattern of the synthesized calcium

carbonate exhibits characteristics peaks at 2θ values of 34.28°. These are the strongest

peaks observed in the X-ray diffraction patterns of the analysed samples which

represent calcite (CaCO3). Knowles and Freeman (2004) and later Jamaluddin and

Ahmad,(2010) stated that CaCO3 crystals overlapping each other induced the fibrous

morphology of the crystal, thus creating rough glaze surface. Due to its fiborous nature,

calcite can cement the existing rock grains or fill the fracture and act as a bridging

agent. All the precursors were assumed to react completely to form calcium carbonate.

This observation is agreeable with previous studies (Yao et al., 2010; Rahman and

Oomeri, 2009; Tong et al., 2004).

104

Figure 4.19: X-ray diffraction pattern of ex-situ prepared CaCO3 NPs starting from the aqueous precursor salts.

4.3.2 Size distribution of ex-situ prepared CaCO3

The TEM photographs and the corresponding particle size distribution histogram for the

ex-situ prepared CaCO3 are shown in Figures 4.20 and 4.21. The histogram shows a

wide size distribution with most of the population falling in the range between 31-60 nm

and a mean particle size of 60 nm. Figure 4.20b demonstrates a high degree of NPs

crystalline nature of the sample taken from an illuminated area of crystals are known to

form easily under laboratory conditions, especially when calcium carbonate precipitates

quickly (Lee et al. 2001). It can be assumed that Ca2+ ion concentration on the surfaces

of nanoparticles resulting in the growth of CaCO3 crystals as shown in Figure 4.20a.

The TEM image shows that the ex-situ prepared NPs have different shapes, from

subrounded to subangular. The picture clearly shows that there is crystalline particles

overlapping and affinity towards formation of polycrystalline. The images were

consistent with the D spectra which showed the crystal structure of CaCO3 NPs.

During NPs preparation, higher collisions rate of NPs might increase the probability of

105

the formation of larger particles and aggregation of fine particles (Nassar and Husein,

2007a), especially since no capping agents were used to prevent aggregation and

shearing was the only way to control particle size.

Figure 4.20 : TEM photographs of ex-situ CaCO3 NPs at two different

magnifications.

Figure 4.21 : Particle size distributions of ex-situ prepared CaCO3 NPs.

N

um

ber

of

Pa

rtic

les

, %

a) b)

106

4.3.3 Determination of particle size of in-situ prepared CaCO3

Calcium carbonate NPs are considered ideal bridging materials of the pore throats of

the mud cake. The particle sizes in nano domain might generate slurries and

suspensions in drilling fluid that will show a reduced tendency to sediment or sag and

minimize the differential sticking problems (Ballard and Massam,2009). TEM results

show that the ex-situ prepared CaCO3 NPs are bigger than their Fe(OH)3 counterparts.

SEM images of the mud cake without and with in-situ prepared CaCO3 NPs are shown

in Figure 4.22. It should be noted that WBM was used this time due to limitations

associated with EM imaging of oil-based mud cake, especially gold coating, which

took much longer time . As discussed earlier, SEM imaging of the filter cake was the

only way of evaluating in-situ prepared NPs. One thing to note at this stage of

discussion, there is no smooth surface like the one obtained with in-situ Fe(OH)3 NPs

formulated in invert emulsion mud. This fact is reflective of the overall poor fluid loss

prevention when the two cases are compared. The mud cakes surface roughness was

varied from cake to cake and particle sizes varied as well. Texture changes could also

happen during evaporation of water and gold coating of the surface of water based mud

cakes (Chenevert, 1991). Mud cake without NPs was rough and the surface was

composed of chunks of large mud particles, which explains the poor loss prevention

capability of the mud cake. In the presence of in-situ NPs prepared from aqueous

precursors as per reaction (R2), the surface became smoother, although roughness is

still apparent. Nevertheless, particles were much smaller in size and could tighten up

fluid intrusion through the surface. Similarly mud cake with in-situ NPs prepared by

bubbling CO2 as per reaction (R5) surfaces was even smoother than the in-situ NPs

prepared per reaction (R2). Comparing the results of with and without NPs clearly

demonstrates that CaCO3 NPs are effective in forming bridges. Clay surfaces were

covered with CaCO3 particles in both NP-based mud cake, as was evident from the

white color of the cake surface. The observed morphologies of the samples showed

distinct features in terms of particle sizes and as well as surface composition evaluated

by EDX of Figure 4.23.

107

Figure 4.22 : SEM images of mud cake a&b) without NPs ;c&d) in-situ CaCO3 NPs (R2); and e&f) in-situ CaCO3 NPs (R5).

100 nm

100 nm

100 nm

a) b)

c)

f) e)

d)

108

From the SEM images it is evident that the pore openings in the mud cake without NPs

were filled with NPs leading to reduced fluid loss, as will be discussed in the following

section. EDX spectra of Figure 4.23 provide elemental composition of the surface of the

mud cakes as shown in Figure 4.22. Magnification of mud cakes showing the grain

sizes and particles size distributions were estimated by using ImageJ software as

shown in Figures 4.24 and 4.25. The SEM images of mud cake without NPs revealed

that the grains mostly irregular from subrounded to subangular (Figure 4.22a). It is also

noted that the pore opening sizes range from 4 nm to 180 nm with average pore

openings were located in 31-60 nm range. These pore openings are resembled as a

pore throat in shale formation. The EDX analysis revealed that Ca content in mud cake

without NPs was below the detection limit or trace amount of Ca content was present

(Figure 4.23a). The in-situ NPs formation was confirmed by the presence of Ca content

and increased amount of Ca element, as per EDX images of Figure 4.23b-c.

Figure 4.25 shows the particle size distribution of in-situ prepared CaCO3 NPs

used as fluid loss additive (bridging agent) in our experiment. The NPs particle size

distribution (1-200 nm range) were confirmed by the SEM images and reported only

those were on the surface of the cake. More than 50% of NPs were 1-30 nm range as

shown in Figure 4.25 and potential to plug the nanometer sized pore openings of the

mud cake as shown in Figure 4.24. A wide particle size distribution was available that

covered some large particles available to bridge across large openings or fracture.

However, it is also to be noted that some NPs might diffuse into the cavity of the pores

permanently and were not counted in the measurements. A magnified photograph of

NPs indicated that particles entangled together to form aggregates and deposited on

the clay pores to form a low permeability mud cake as shown in Figure 4.22c-f.

109

Figure 4.23 : Elements containing mud cake a) without NPs,b) with insitu NPs (R2) and c) with insitu NPs (R5) from EDX data.

a)

b)

c)

110

Figure 4.24 : Available pore openings (nm) in mud cake of DF without NPs.

Figure 4.25: Particle size distribution of in-situ CaCO3 NPs, prepared by reactions (R2) and (R5), in the mud cake.

0

10

20

30

40

50

60

70

1-30 31-60 61-90 91-120 121-180

Pore openinigs in mud cake (nm)

Fre

qu

en

cy,%

111

The range of NPs size is narrow and it can be seen that the method followed is an

appropriate method for NPs production. The selection of CaCO3 NPs as bridging

material with a specific particle size distribution was in accordance with the physical-

chemical characteristics of formations to be drilled. It is important to have a substantial

colloidal fraction of particles in the mud with a broad PSD (particle size distribution). The

criterion of selection of particle size of bridging agent is particles about one-seventh to

one-third the size of the maximum “pore throat” and the fluid must maintain a significant

concentration of those particles throughout the interval (Cargnel and Luzardo,1999).

Thus, if NPs, for example, is larger than the diameter of the pores, it will simply sit on

top of that pore. There are other various guidelines used in industry to choose the

particle size of bridging materials that can form an efficient external filter cake. A median

particle size of the bridging agent equal or slightly greater than one third of the median

pore size of formation and concentration of bridging agents must be at least 5% by

volume in final mud mix (Abrams,1977), 90% of the particles are smaller than or equal

to the pore size of the rock ( Hands et al.1998). In the mud cake pore throat size

diameter falls below 10% at 61-180 nm range and 90% of nanoparticles lies between 1-

60 nm range. According to the relationship stated above, the NPs size is more suitable

for bridging materials at the cake surface and ensure that tailor made NPs are potential

to reduce the permeability of mud cake. As the dimension of NPs lies in the contiguous

area between the clusters and the macroscopic materials, they will not directly dictate

macroscopic properties, but bring their own unique effects such as surface effect, size

effect etc (Nabhani et al.2011).

Mud cake contains interconnecting pore spaces more like those of permeable

rock considered as a theoretical pore throat diameter of shale and is just an

approximation. A mud having wide range of particle size distribution adsorbed by clay

might slow down the filtrate. It presumably would seal the surface pores, stuck on the

surface of the clay and filter paper and establish the formation of low permeability filter

cake, whereas without NPs mud containing only clay start to form highly permeability

filter cake. The correct particle size distribution provides better compaction medium with

constrained flow of liquid from the drilling fluid. Therefore, drilling fluid containing CaCO3

112

NPs of sizes ranging up to 200 nm, the requisite maximum were able to effectively

bridge the formation and formed filter cake.

The relative pore size openings of the mud cake of drilling fluid without NPs

explained that the SEM result was found to be in good agreement with NPs considered

as a bridging or plugging agent to reduce the fluid loss.

4.3.4 LTLP Filtration of in-house prepared CaCO3 NPs

Drilling muds can cause large irreversible damage to fractures and dramatically reduce

the productivity of wells. Leoppke et al. (1990) found that if the particle size is not

compatible with the fracture width, a stable bridge cannot be formed and therefore

tailored particle size distribution provides the best plugging capabilities. It is essential to

drill the wells with minimum cost in loss of fluids. Al-Riyamy and Sharma (2004) used 5

wt% CaCO3 of narrow size distribution and found that volume of the filtrate decreased

when CaCO3 was used and reduced the invasion of emulsion droplets into the

formation, although granular CaCO3 LCM were found much less effective (Jiao and

Sharma, 1996). Currently all types of CaCO3 are used largely as fluid loss control

additives in drilling fluid. But the current size range of CaCO3 used does not serve the

purpose of the complete fluid loss control. More interestingly, the nanometer CaCO3

could result in much thinner filter cakes than those obtained using large sized CaCO3.

Isambourg and Matri (1999) showed how much force required to free a stuck pipe with a

change in mud cake thickness. This highlights the importance of nano drilling fluid with

thin mud cake development. The first step when choosing the particle size distribution of

bridging agent specifically CaCO3 in drilling fluid is the petrophysical characterization

and pore geometry determination of the rock. In consolidated sands, the criterion of

selecting particle size as bridging agent is (Cargnel and Luzardo,1999) :

1/7 DPore throat < Dparticle < 1/3 DPore throat

Bentonite clay particles have sizes of ~1-2 μm (at dispersed phase) according to

the supplier. During the migration of particles through the paper filter (2.7 µm pore

diameter) in our case resembling pore throats of the rock, they began to accumulate in

113

the filter surface. To avoid internal blocking, it is necessary to create a mud cake at the

surface of the pore throat in wellbore.

Static filter loss tests are relative, i.e, they can compare on a qualitative level

which mud systems are preferable and widely used by drilling crews for routine field

tests (Nyland et al., 1988). Total fluid loss is the indicator of the filtration controllability of

mud and its additives. It is apparent from the Table 4.18 that more than 60% of the API

fluid loss reduction occurred using ex-situ NPs and 55% when used in-situ CaCO3 NPs

prepared by reaction (R2). NPs primarily form a impermeable filter cake surface on the

filter paper. Permeability decreases with increasing fraction of colloids and is affected

strongly by particle size and shape of NPs. Flocculation causes particles to form a loose

and open network leading to higher filtration rate as indicated by the drilling fluid without

NPs where clays are dominant. After adding CaCO3 NPs in the drilling fluid probably

acted as a cementing and bridging agent that stabilized the bentonite clay aggregates

and could decrease the clay swelling and prevented their disintegration. Due to CaCO3

concentration in drilling fluid, which probably below the flocculation value caused the

migration of dispersed NPs into the pores of mud cake during filtration.

Comparison between Tables 4.18 and 4.19 shows that total filtrate is the highest

for the water based mud than the invert emulsion based mud. In both cases, the invert

emulsion NP-based mud had a very low filtration rate during the first 7.5 min and the

rate still lower during the 30 min. The water based NPs mud had a much higher initial

filtrate rate but after 30 min it was still closely two times greater than the invert emulsion

NP-based mud. CaCO3 in water based mud might have a flocculating effect seen by the

relative fluid loss performance with respect to invert emulsion mud and also affected by

the difference in mud constituents in the two fluid systems. Using 3 wt% nano CMC and

nanopolymer as a fluid loss additives in water based mud, 14% and 19% respective

fluid loss reduction was noticed in literatures (Saboori et al,2012; Manea et al.2012),

whereas 3 wt% CaCO3 NPs addition in the current experiments yielded 30% fluid loss

reduction.

114

Table 4.18: API LTLP fluid loss comparing invert emulsion drilling fluid as base case with invert emulsion drilling fluid samples containing 4 wt% in-house prepared CaCO3 NPs using reaction (R2). No LCMs added

*Fluid loss reduction,%

Table 4.19: API LTLP fluid loss comparing water based drilling fluid as base case with water based drilling fluid samples containing 3 wt% in-house prepared CaCO3 NPs by reaction (R2).

*Fluid loss reduction,% When Ca(NO3)2 and Na2CO3 precursors are added in the drilling fluid in order to form

CaCO3 NPs, a bi-product of NaNO3 salt was also produced. This formation increased

the ionic atmospheric charge on clay sheets. The attractive force between the ionic

atmosphere might force the individual clay sheets to regrouping, decrease the pore

openings and interlock at random angles, thereby fluid loss reduction would happen.

Since, such regrouping is a matter of statistical probability, some clay sheets may still

have openings difficult to move and, therefore, complete fluid loss reduction was not

possible. On the other hand, sodium nitrate itself acts as nitrogen based fertilizer.

Adding nitrates encourages the proliferation of nitrate-reducing bacteria in the oil-

seawater mixture. When present in the appropriate numbers these bacteria help loosen

oil from the rocks containing the reservoir. For this reason, since long Statoil Norway is

injecting sodium nitrate along with seawater to pump oil from the underground reservoir

(RSC,2003).

Samples

Types

Time

(min)

LTLP Fluid Loss (mL)

DF DF + ex-situ NPs DF +in-situ NPs

90:10 (v/v) Oil: Water

7.5 3.9±0.2

0.7±0.6

1.1±0.6

30 8.7±0.2 2.8±0.6 (68%*) 3.9±0.3 (55%*)

Samples Types LTLP Fluid Loss (mL/30 min)

DF DF + ex-situ NPs DF +in-situ NPs

Water based DF 9.5±0.2 6.5±0.2 (32%*) 6.8±0.2 (28%*)

115

A 4 wt% CaCO3 NPs concentration represents the optimum concentration in

which the volume of filtrate reached minimum values and better arrangement of

particles occurred in the filter cake surface turning into an impermeable cake. Besides,

the spurt losses are found lowest. Comparing between Tables 4.18 and 4.20, it can be

easily seen that in-situ CaCO3 NPs reaction (R5) yields lower spurt loss, filtration rates

and total filtration volume. The cake formation was instantaneous and effective. During

the initial stage of filter cake forming, NPs plug or bridge the near surface pores and

reduces formation permeability. Bailey et al. (1999) showed that the particle bridging

reduced the spurt loss. Because of this bridging tendency, quick external filter cake

formation is obvious in case of the in-house prepared CaCO3 NPs using reaction (R5).

Table 4.20: API LTLP fluid loss comparing invert emulsion drilling fluid as base cases with invert emulsion drilling fluid samples containing 4 wt% in-house prepared CaCO3 NPs using reaction (R5).

*Fluid loss reduction,%

NPs of CaCO3 modify the structure of clay due to Ca2+ cation exchange leading to

agglomeration of clay particles, which also increases internal friction among the

agglomerates and thereby reduced permeability of mud cake. As far as these NPs are

considered as fluid loss reducing agent, it gains better properties of keeping the

cumulative volume of filtrate at low values. It is due to the fact that when these NPs are

brought into the clay particles, the interaction area was considerably increased due to

higher surface area of NPs. Increased area to volume ratio in NPs cause the increase of

ionic group molecular weights for adsorption on the clay particle surface and attached

them to each other leading to form more colloidal particles. In addition, the presence of

CaCO3 NPs may induce a Brownian diffusion with clay particles. Spurt losses observed

Samples

Types

Time

(min)

LTLP Fluid Loss (mL)

DF DF +4 wt% in-situ NPs

90:10 (v/v) Oil: Water

7.5 2.6±0.2

0.5±0.3

30 6.4±0.3

2.2±0.4 (66%*)

116

with CaCO3 are acceptable from a drilling point of view. Looking at the above tables, it

is obvious that the effect of NPs presence intensified the flow resistance of the system.

4.3.5 HTHP Filtration of in-house prepared CaCO3 NPs

High temperature filtration rates could not be predicted from low temperature filtration.

Filtration rates increased at high temperature that could be attributed to the reduced

viscosity of oil alone. The samples exhibited a very low fluid loss at low temperatures

and same relative performance at high temperatures. Nanos having an excellent

thermal conductivity are expected to be the materials of choice in HTHP wells (Agarwal

et al., 2009). When the pressure is applied (500 psi differential in the HPHT tests) the

filtration became slow due to the bridging/agglomeration tendency of NPs at high

temperature. The HTHP fluid loss property of CaCO3 NP-based drilling fluid prepared

through reaction (R2) is shown in Table 4.21. It showed that NPs concentration equal to

4 wt% reduced more than 70% of the fluid loss. Similar trends were observed at

previous Fe(OH)3 NPs. The differences are thought to be a result of the surface

morphologies of the two different NPs. In order to determine the HTHP fluid loss with

CaCO3 NPs prepared in reaction (R5), results are reflected in Table 4.20. It is shown

that NP-based mud provided the lowest spurt loss and total filtrate loss. Addition of

optimum concentration of NPs improved filtration properties, however the extent of the

improvement depend on the mud type, nature of NPs material, NPs synthesis method,

size distribution of NPs and concentration of surfactants. As it can be seen, the total

filtrate passes through is minimum in the case of in-situ CaCO3 NPs prepared through

reaction (R2). But more interestingly, fluid loss towards zero is apparent in reaction

(R5).

117

Table 4.21: HTHP fluid loss comparing invert emulsion drilling fluid as base cases with invert emulsion drilling fluid samples containing 4 wt% in-house prepared CaCO3 NPs using reaction (R2).

*Fluid loss reduction,% Table 4.22: HTHP fluid loss comparing invert emulsion drilling fluid as base cases with

invert emulsion drilling fluid samples containing 4 wt% in-house prepared CaCO3 NPs using reaction (R5).

*Fluid loss reduction,%

High pressure and high temperature filtration tests clearly demonstrated that

CaCO3 NPs have the ability to reduce filtrate loss to the formation. No emulsion droplets

were observed in the filtrate means emulsion droplets containing NPs were needed to

form a stable external filter cake. The acid soluble CaCO3 NPs could be concentrated at

oil/water interface as like Fe(OH)3 NPs and restricted the flow of fluid to the porous

media.

4.3.6 Drilling fluid density and pH

As different formations are encountered with depth increases, the densities of drilling

fluids shall be proportionately adjusted to balance the drilling system while drilling. So

weighting materials should be continuously added to drilling fluids. CaCO3 NPs used as

weighting materials in adjusting drilling fluid density could reduce the extra additive cost.

Samples

Types

Time

(min)

HTHP Fluid Loss (mL)

DF DF+ ex-situ NPs DF+in-situ NPs

90:10 (v/v) Oil: Water

7.5 10±0.2

2.4±0.1

1.8±0.2

30 19.2±0.2 5.6±0.1 (71%*) 5.4±0.2 (72%*)

Samples Types

Time

(min)

HTHP Fluid Loss (mL)

DF DF +4 wt% in-situ NPs

90:10 (v/v)

Oil: Water

7.5 10±0.2

0

30 19.2±0.2 0 (100%*)

118

As shown in Table 4.23, a significant change in the drilling fluid density was seen. Due

to its distinctive nature as weighting material, addition of CaCO3 NPs can increase the

fluid density. But no change in pH was observed. It can be observed that the pH is 12.5

remains constant even after addition of CaCO3 NPs. This effect is probably explained

by the absence of chemical interactions between CaCO3 NPs with other materials used

in drilling fluids. Apart from that the NPs concentration was too low to increase and/or

decrease of density and pH of CaCO3 NP-based invert emulsion fluid. NPs may be

embedded in the clay matrix.

Table 4.23 : Density and pH measurements of drilling fluid samples in the presence

and absence of 4 wt% in-house prepared CaCO3 NPs.

4.3.7 Rheology behavior of NP-based fluid

When rheological properties are considered, CaCO3 NPs provide satisfactory mud

system. Mud rheological properties (viscosity vs. shear rate) were measured at room

temperature and pressure. Analyses of the rheological properties of the drilling fluids

shown in Figure 4.27 indicated a viscous profile of the nano-based fluid at low shear

rate, but at high shear rate there was no significant changes. Using CaCO3

microparticles, the viscosity of drilling fluids were increased substantially as shown by

Manea et al. (2012). Properties measured at 10-sec and 10-min gel strength were also

shown in Figure 4.28.

Samples

Types

Properties

Test samples

DF DF+ 4 wt% ex-

situ NPs

DF+ 4 wt% in-

situ NPs

90:10 (v/v) Oil: Water

Density (g/cm3)

0.89±0.06 0.93±0.02 0.93±0.02

pH 12.5 12.5 12.5

119

Figure 4.27: Rheological behavior of invert emulsion drilling in presence and absence of 4 wt% in-house prepared CaCO3 NPs. No LCMs added.

Figure 4.28: Gel strength behavior of invert emulsion drilling fluid in presence and

absence of 4 wt% in-house prepared CaCO3 NPs. No LCMs added.

1

10

100

1000

1 10 100 1000 10000

DF DF+4wt% Ex-situ NPs(R2)

DF+4wt% In-situ NPs (R2) DF+4wt% In-situ NPs(R5)

Shear Rate (Sec-1)

Ap

pa

ren

t V

isc

osit

y

(cP

)

120

All the muds had almost similar viscosities in the high shear range although NP-

based muds were slightly thicker in the low shear range than conventional invert

emulsion based muds (DF). The shear thinning properties of drilling fluid would be

advantageous in providing better hole cleaning. Similar trends were observed by

Simpson, (1979) and Sensory, (2010). With the addition of CaCO3 NPs (both ex-situ

and in-situ), the gel strength was also increased. Similar observations were also

reported by Manea et al.(2012). CaCO3 NPs as a bridging agent might have proper

mechanical and chemical consistency to be used in NP-based drilling fluid design.

Chemically, it is acid soluble so that CaCO3 laden mud cake can be removed easily

from the porous matrix to recover the permeability of the rock.

On the basis of the results, we believe NPs size and ability to keep particles

dispersed throughout the mud system allowing fluid loss reduction. CaCO3 NPs based

mud system proved versatile enough to provide better fluid loss control while retaining a

consistent viscous profile.

4.3.8 Drilling fluid lubricity

Lubricity is a very important parameter that was considered due to long extended reach

characteristics of wells. Coefficient of friction less than 0.1 or less is generally

advantageous as it helps the cuttings to travel as discrete particles over shaker screens

(ASME, 2005). The choice of CaCO3 NPs as a lubricity additive is due to its availability.

With the help of surface activity of CaCO3, NPs can effectively lower the friction

between drilling tools and borehole walls and reduce the difficulty in drilling of highly

deviated wells and horizontal wells. 4 wt% of CaCO3 NPs induced appreciable reduction

in the coefficient of friction. A comparative performance with regular base drilling fluid is

shown in Table 4.24. The friction of coefficient was reduced to 2.1% and 38% by using

ex-situ and in-situ prepared CaCO3 NPs, respectively. Using commercial silica as NPs

in drilling fluids improved lubricity by 3-5% as reported by Riley et al. (2012). High

drilling cost can be caused by slow drilling rates due to improper lubricity quality.

Therefore the results indicated the feasibility of CaCO3 NPs that could permit the

improved drilling rate.

121

Table 4.24 : Coefficient of friction of invert emulsion drilling fluid in presence and absence of 4 wt% in-house prepared CaCO3 NPs. No LCMs added.

4.4 Invert emulsion drilling fluid API fluid loss characterization using other NPs

From a practical point of view, the presence of NPs in drilling fluid is worth to be studied,

evaluated and therefore found promising. The advantages of using these NPs are that,

there is formation of more continuous and integrated mud cake. Having low permeability

and low porosity mud cake, there is less volume of filtrate entering into the formation

and mud cake thickness is less compared to DF and DF with LCM cases. Regular mud

systems contain large quantities of fine solids that penetrate the productive formation

causing irreversible plugging and influence negatively in the productivity of the wells.

Barite (BaSO4) of drilling fluids is such a component and when it invades the productive

area, it creates internal block within the formation which is difficult to remove (Cargnel

and Luzardo,1999). In normal, barite in drilling fluid is used as weighting material.

Conventional powdered barite exhibit an average particle diameter in the range of 10-30

microns. To adequately suspend these materials requires the addition of gellant such as

bentonite. However more gellants addition could increase fluid viscosity and undesirable

fluid properties (Ballard and Massam, 2009). The nano barite could alleviate those

problems in drilling fluid and may control colloidal interaction of particles.

Different NPs are prepared in the novel method described in the experimental

section to validate the prepared method. The techniques provide novel insight in fluid

loss reduction problems while drilling wells and help in counting the problems in a more

efficient and environment friendly manner. BaSO4 and FeS NPs were tested to check

their specific performance in fluid loss control as shown in Table 4.25. Addition of 3 wt%

NPs in drilling fluid reduced fluid loss from 68 to 85 % for BaSO4 and 90% for FeS NPs.

The adsorption process in clay involves a negative global charge in its surface, is

Co-efficient of friction % torque reduction

DF without NPs (control)

DF+ex-situ NPs

DF+in-situ NPs

DF+ex-situ NPs

DF+in-situ NPs

0.095 0.093 0.059 2.1% 37.89%

122

generally balanced by inorganic cations (Ba2+, Fe2+) in the internal and external

surfaces of clay material. Some nanoparticles may also aggregate and deposit into the

cake, therefore the Brownian motion of the particles and the van der Waals and

repulsive forces for the nanoparticles are equally important.

Table 4.25 : API LTLP fluid loss of invert emulsion DF in presence and absence of 3

wt% in-house BaSO4 and FeS NPs. No LCMs added.

*Fluid loss reduction,%

4.5 Summary of the API fluid loss study of different NPs in Invert emulsion

The most frequently encountered problems while drilling oil and gas wells are lost

circulation causing substantial financial loss. This problem occurs during the

uncontrolled flow of drilling fluids into vicinity of porous and permeable zones. The

proposed method incorporates using different nanoparticles in drilling fluid to decrease

fluid penetration and mud cake thickness. In the current study, the multifunction of NPs

in drilling fluid system is considered as the new hotspot in the field.

Results show that existence of nanoparticles caused the amount of fluid loss

reduction and mud cake thickness to decrease. By forming a thin, low permeability filter

cake NP-based fluid simulates sealing pores and other openings in the formations

penetrated by the drill bit. Different NPs were used as bridging agent to control the fluid

loss. Figure 4.29 and Figure 4.30 showed the effectiveness of different ex-situ and in-

situ NPs respectively in terms of fluid loss over a period of 30-min.

mL/ 30 min

DF DF+ ex‐situ BaSO4 DF+ in-situ BaSO4

10.95±0.3

3.5±0.3 (68%*) 1.6±0.3 (85.3%*)

DF+ ex-situ FeS DF+ in-situ FeS

1.15±0.3 (89.5%*) 0.93±0.1 (91.5%*)

123

Figure 4.29: LTLP fluid loss behavior of different ex-situ NPs in invert emulsion.

Figure 4.30: LTLP fluid loss behavior of different in-situ NPs in invert emulsion.

The nanoparticle concentration in drilling fluid was varied from less than 1 wt% to 4 wt%

(1 wt% Fe(OH)3 NPs, 4 wt% CaCO3 NPs, 3 wt% BaSO4 NPs and 3 wt% FeS NPs).

From the experimental trials, it can be concluded that zero spurt loss was achieved with

thin mud cake in all NP-based fluid. However, in-situ NPs responded with lower fluid

0

20

40

60

80

100

Fe(OH)3 CaCO3 (R2) CaCO3 (R5) FeS BaSO4

Fe(OH)3 CaCO3 (R2) CaCO3 (R5) FeS BaSO4

Flu

id lo

ss r

ed

uctio

n, %

124

loss than ex-situ NPs. Due to different physio-chemical nature, sizes and interaction

potential with clays caused the NPs to behave differently. Therefore, optimum

concentrations of NPs were required to meet the desired fluid loss control property.

If the NPs formed were not well dispersed, stabilization of fluid systems were

hindered and then particles tempted to agglomerate and precipitate. Figure 4.31 shows

the different ex-situ prepared NPs and their stability performance. Adequate usage of

emulsifiers prevented the phase separation through steric or electrostatic means of

adsorbed dispersed NPs on clay matrix and/or emulsifiers surface. Increased viscosity

of the continuous phase system can also prevent separation of dispersed phase (Riley

et al., 2012). As nanoparticles addition in drilling fluid did not increase the fluid viscosity

significantly, it is therefore, best prefer in our current works to support steric or

electrostatic way of NPs stabilization.

Mud having low filtration characteristics deposited as thick filter cakes (Nyland et

al.,1988). Conversely, good filtration characterized by all the NP-based fluids yielded

thin mud cakes as shown in Figure 4.32. The thickness of the mud cake and its integrity

can reveal that NPs deposited on the cake with optimum concentration establish an

effective seal. For combating formation damage, NP-based invert emulsion as well as

water based mud would offer better protection. The filtrate invasion from the invert

emulsion would be less at the downhole pressure and offer better protecting against

differential pressure sticking.

125

Figure 4.31: Different NPs and NPs-containing drilling fluid stability evaluation.

Figure 4.32: NPs-containing drilling fluid filter cakes (thickness <1 mm).

FeS NPs BaSO4 NPs CaCO3 NPs Fe(OH)3 NPs

CaCO3NPs filter cake Base filter cake filter

cake

FeS NPs filter cake

Fe(OH)3 NPs filter cake

BaSO4 NPs filter cake

126

Chapter Five: Modelling

In this chapter, Darcy filtration equation was used to model drilling fluid loss for LTLP

filtration experiments only. Based on the results, permeability of the mud cake at

different times was estimated. In this section, time-dependent behavior of filter cake

build up model was also evaluated. An attempt to address how the nanoparticles

transport during filtration was made. Bingham plastic rheology was used to predict the

relationship between the shear rate and shear stress for the drilling fluids under study

and allowed to calculate several other important attributes of the fluid.

5.1 LTLP API filtration model using Darcy’s law

Drilling fluid filtration rate and its behavior with respect to time were estimated from the

experimental results using Darcy’s law (Maduka, 2010; Kumar, 2010; Hoff et al., 2005;

Donaldson and Chernoglazov, 1987; Ferguson and Klotz, 1954; Williams and Cannon,

1938). The LTLP API filtration test was static, dead end filtration, as the mud was not

circulated during filtration and the filter cake was allowed to grow without disruption by

shear forces. Under this condition, certain volume of a stable suspension is filtered out

against a permeable substrate, e.g. filter paper, with time. At time t, certain volume of

filtrate is removed by filtration at constant temperature and pressure (25°C, 100 Psi).

During the filtration process, filter cake accumulates and the volume of the mud sample

in the filter press is decreased. Assuming constant density (temperature must be

constant in order for Darcy’s law to be valid. If temperature changes with time, density is

also a function of time), the material balance equation expressed as a volume balance

for mud filtration process can be written as follows.

Drilling fluid volume in the filter press = Wet Filter cake volume + Filtrate Volume

127

The filter cake is assumed uniform throughout and the rate of growth of the filter cake is

proportional to the rate of filtrate. According to the volume balance, if a unit volume of a

stable suspension of solids is filtered against a permeable substrate and x volume of

filtrate is collected, then 1-x volume of cake (solids plus liquid) will be deposited on the

substrate. The following equation can be written (Maduka, 2010; Hoff et al., 2005).

rx-1

x

Q

Q

f

C (E 5.1)

where Qc and Qf are the volumes of the filter cake and filtrate at a given time,

respectively, r is the ratio between the volume of the filter cake at a given time to the

volume of the fluid filtrated in the filter press. From our experimental works and

filteration curves provided by Barkman and Davidson(1972), we can assume the

following criterion for the ratio of f

C

Q

Q.

if f

C

Q

Q <1 ; initial fluid pass through the filter cake and happens only at the

early stage of filtration

if f

C

Q

Q =1 ; fluid loss during filtration approximated by the linear relationship

between volume of filtrate vs. time. It can be termed as equilibrium filtration

if f

C

Q

Q >1 ; saturation occurs, curve departs from linear relationship

between volume of filtrate vs. time and slows fluid loss due to the compact cake

layer formation. Filtrate decrease with increase of cake volume

128

The cross sectional area of the filter cake A is constant under static filtration. The

volume of the filter cake, Qc, is given by the product of cross sectional area of the filter

press, A, and the thickness of the mud cake at a given time, hmc.

mcc A.hQ (E 5.2)

Therefore,

A

.Q

A

Qh fc

mc

r (E 5.3)

From Darcy’s law, the flow rate of filtrate through the mud cake (an unconsolidated

porous medium) is given by,

mc

f

h

PkA

dt

dQ

(E 5.4)

Substituting (E5.3) into (E5.4) gives,

f

2f

Q

PkA

dt

dQ

r

(E 5.5)

Integrating (E5.5) assuming constant permeability, viscosity and pressure difference

gives,

r

t P2kAQ

2

f

2 (E 5.6)

It should be noted that the ∆P was maintained constant throughout the filtration process.

Darcy’s law is obtained empirically and defines the permeability k as a proportionality

coefficient in the relationship between flow rate and pressure gradient (Costa, 2005).

Cake permeability is much lower than the permeability of filter medium.Finally, Darcy’s

129

law under the above assumptions leads to the following expression of filtrate volume

versus time.

At P2k

Qf .r

(E 5.7)

The rates of mud filtration and mud cake formation are both function of time and

proportional to each other during filtration. Therefore, under equilibrium filtration

assumption, r can be taken as 1 (Hoff et al., 2005) and consequently,

A P2k

kwheretkQ //f .;

(E5.8)

It is well known that when an external mud cake begins to form and grow, the filtrate

volume is proportional to (Kumar, 2010; Hoff et al., 2005). From (E5.8) and (E5.3),

the thickness of the mud cake, hmc, at any time, t, during the filtration process, can be

simplified to,

Pt k 2hmc

(E5.9)

At the initial exposure of a permeable formation to a drilling fluid, three stages of

mud cake build up evolve: 1) spurt loss, which corresponds to the initial loss of fluid to

the formation, 2) buildup of filter cake, during which fluid filtration is proportional to the

square root of time as reported by many researchers (Kumar, 2010; Hoff et al., 2005;

ASME,2005) and 3) filter cake growth, which might be limited by the erosive action of

mud stream within the dynamic context of real time drilling (Outmans, 1963). It should

be noted that the last stage does not exist under static filtration. The surface of the

“dynamic” filter cakes erode to an extent that depends on the shear stress exerted by

the hydrodynamic force of the mud stream relative to the shear strength of cake’s upper

layers (Caenn et al., 2011). Spurt loss can be obtained by extrapolating filtrate volume

versus t to zero time and is given approximately by the y-axis intercept of the plot as

shown in Figures 5.1 and 5.2. In all cases the NP-based fluids were compared with the

corresponding control DF samples. Although DF samples were composed using the

130

same constituents and obtained from the same suppliers, nevertheless, this does not

guarantee no variation from one batch to another. Therefore, NP-based fluid

performance was compared with respect to its own DF control samples. From the

current experiments, addition of NPs significantly reduced the bridging time and,

therefore, the spurt loss. In Figure 5.2, symbols (R2) and (R5) stand for the reactions

used in the experimental methods described in Chapter 3. The plot of filtrate volume, Q f,

versus ‎ t suggests a different filtration mechanism in the presence of NPs. The very

low Y-intercept for the case were NPs are used suggest that time needed to completely

bridge the porous mud cake to reduce the fluid loss is much faster in case of NP-

mediated DF, and more specifically in the case of Fe(OH)3 NPs relative to CaCO3 NPs.

This suggests that Fe(OH)3 NPs are more effective than CaCO3 NPs at bridging across

the face of fracture of porous formation. A coefficient of determination, R2, approaching

1 for a straight line fit, not going through the origin, between Qf and t with positive

intercept in the case of no NPs, i.e. control samples and drilling fluid with LCM, indicates

that spurt loss is important, whereas for the NP-based fluids it was negligible. Spurt loss

is largely caused by the tendency of the particles to pass through the filter paper until its

pores become partially plugged, which eventually leads to linear relation between filtrate

volume and square root of time. Typically, a linear relationship w.r.t. square-root of time

represents wall building fluids (Clark,1990; Chin, 1995). Conversely, a region of non-

linear relationship appeared in the case of NP-based fluid at the beginning of filtration

due to the absence of spurt loss. An extrapolation of the linear portion of these curves

can lead to a negative spurt loss value. So it is evident from the trend that early portion

of curve did not follow the Darcy’s law. Therefore, (E5.9) does not provide a good

estimate of cake growth in the case of NP-based mud. The role of Brownian diffusion

during the initiation of filter cake will be explained in the next section.

131

Figure 5.1: Filtrate volume variation with square root of time in the presence and absence of in-situ and ex-situ prepared Fe(OH)3 NPs.

Figure 5.2: Filtrate volume variation with square root of time in the presence and

absence of in-situ and ex-situ prepared CaCO3 NPs. (R2) and (R5) refer to the reaction used to prepare the CaCO3 NPs per Section 3.2.

The permeability of filter cake is the fundamental parameter that controls static filtration

(Caenn et al., 2011; Byck, 1940). During filtration of mud with and without conventional

LCMs, the trend seem to follow Darcy’s equation throughout, and mud cake

permeability (fitted parameter) was reduced exponentially with time as shown in Figures

R² = 0.99

R² = 0.98

R² = 0.99

R² = 0.97

-0.5

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

0 200 400 600 800 1000

DF

DF+LCM

DF+Ex-situ NPs

DF+In-situ NPs

V

olu

me

of filtra

te , 𝑄

𝑓(c

m³)

No spurt loss

R² = 0.98

R² = 0.99

R² = 0.98

R² = 1.00

0

1

2

3

4

5

6

7

8

9

10

0 200 400 600 800 1000

DF

DF+Ex-situ NPs (R2)

DF+In-situ NPs (R2)

V

olu

me

of filtra

te , 𝑄

𝑓(c

m³)

Time ( ), sec½

Time ( ), sec½

132

5.3 and 5.4. The figures show a different behaviour for the mud cake permeability with

time for the NP-based fluid. The term /k were determined during each filtration times

from the data points of the Figure 5.1 and 5.2. Then using (E5.8), mud cakes

permeabilities were estimated at each time interval. The results were presented for the

sake of simple comparison between the drilling fluid with and without NPs using Darcy

equation. Spurt loss was nil during the initiation of the filter cake of NP-based mud, and

when filtration was stopped at a certain time, the mud cake growth reached a constant

value. It appears that the highly compacted filter cake contributed to significantly less

permeability and static filtration rate as evident from the smaller thickness of the filter

cake.

Figure 5.3: Mud cake permeability variation with time in the presence and absence of

in-situ and ex-situ prepared Fe(OH)3 NPs. Permeability obtained from fitting E5.8.

0

0.002

0.004

0.006

0.008

0.01

0.012

0 500 1000 1500 2000

DF

DF+LCM

DF+Ex-situ NPs

DF+In-situ NPs

P

erm

eability o

f m

ud c

ake

(n

D)

Time (sec)

133

Figure 5.4: Mud cake permeability variation with time in the presence and absence of

in-situ and ex-situ prepared CaCO3 NPs. Permeability obtained from fitting E5.8. (R2) and (R5) refer to the raction used to prepare the CaCO3 NPs per Section 3.2.

The permeability was calculated from the equation (E5.8) and compared with the base

drilling fluid after 30 min filtration as shown in Table 5.1. It is shown that more than 90%

mud cake premeability reduction was achieved in the presence of NPs, whereas

conventional LCM reduced the permeability by 77%. The extent of permeability

reduction varied with NP type and method of preparation. Data in Table 5.1 suggest

potential use of NPs to plugg shale formation. As indicated in the results section, NPs

are capable of providing effective sealing since they displayed wide range of size

distribution, which could effectively bridge between clay particles initially forming the

cake.

Table 5.1: Permeability reduction of mud cake in the presence and absence of NPs. (R2) and (R5) refer to the reaction used to prepare the CaCO3 NPs per Section 3.2.

NPs Test Fluid Cake Permeability after 30

min filtration (nD) % Reduction in

Permeability

DF(Control samples) 2.59X10-3 0 DF+LCM 5.84 X10-4 77.6

Fe(OH)3 DF+Ex-situ NPs 2.59 X10-4 90 DF+In-situ NPs 1.62 X10-5 99.4

DF( Control samples) 1.04 X10-2 0 DF+Ex-situ NPs (R2) 5.84 X10-4 94.4

CaCO3 DF+In-situ NPs (R2) 2.59 X10-4 97.5 DF+In-situ NPs (R5) 2.59 X10-4 97.5

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0 500 1000 1500 2000

DF

DF+Ex-situ NPs (R2)

DF+In-situ NPs (R2)

DF+In-situ NPs(R5)

Pe

rme

ab

ility o

f m

ud

ca

ke

(n

D)

Time (sec)

134

On the other hand, inappropriate size of plugging materials results in the formation of a

thick and permeable filter cake leading to continuous filtration, which seems to be the

case for drilling fluids alone and DF with conventional LCMs. Therefore, pore throat

diameter of formation must be known to help ensuring effective bridging. A general rule

of thumb for estimating the unknown pore throat diameter (microns) is to take the

square root of the permeability in milli Darcies as per (E5.10). For effective bridging 20-

30% by weight of the bridging material should be one-third of the pore size in microns

as per (E5.11) (kumar, 2010).

Pore throat diameter (microns) = ty Permeabili (mD) (E5.10)

D(50) = 3

diameter Pore (E 5.11)

Following Darcy equation, (E5.4), calculation gave permeability available of the

filter cake of drilling fluid control sample after 30 min filtration was approximately from

0.0001 to 0.000051 mD= 102 to 51 nD. This permeability is low and actually is of the

same order magnitude of as in shale (Chenevert and Sharma, 1991). According to the

above equations, estimated pore throat size of the control sample mud cake after

filtration is 17-34 nm. It does suggest that pore throat sizes are on the nm scale within

the cake. NP-based drilling fluids contain 70% of NPs with sizes ranging from 1 to 30

nm as shown in Figure 4.4, which could easily ensure effective bridging and plugging of

the pores. Therefore, when NPs were introduced, further reduction in cake permeability

occurred, which was not achieved by using conventional LCMs.

Materials accumulation, which is commonly referred to as fouling may arise from

particles deposits on the filter surface, adsorbed on the cake surface, or within the cake

pores. The concentrations of fouling materials at the cake surface typically increases

with time. Consequently, resistance to permeate flow increases with time. NPs

accumulate in the cake surface due to convective deposition and reach a threshold

where the formation of cake layer can be predicted. This cake provides significant

135

resistance to fluid loss. Cake growth performance can be evaluated in terms of

permeate flux as follows:

Permeate (filtrate) flux = S . t

V (E5.12)

where V is the volume of permeate in mL, t is the permeate collection time in sec and S

is the filter effective surface area = 62.06 cm2, from LTLP apparatus.

Figure 5.5 shows that permeate flux decreased with time probably due to cake layers

formation. The results show that the flux declined rapidly during the first 300 and 450

sec for DF and DF+LCM, respectively, followed by gradual decrease during the period

of 500 and 1500 sec. This type of decline is indicative of fouling, or resistance, gradually

building up on the surface of the filter paper. Although not on the same scale, permeate

flux for DF+NPs experienced sudden increase with time and approached a steady flow

as shown in Figures 5.5. The trend in permeate flow for NP-containing fluids can only

be explained by changes in cake structure. This influence of particle size on the

permeate flux was also described by Sethi (1997). Smaller particles (less than 0.01 µm)

formed thin filter cakes, which resulted in higher permeate flux, whereas monotonic

decline in flux occurred when larger particles (greater than 1 µm) were used.

136

Figure 5.5: Comparison of permeate (filtrate) flux with time in the presence and

absence of in-situ and ex-situ prepared Fe(OH)3 NPs.

Figure 5.6: Comparison of permeate (filtrate) flux with time in the presence and

absence of in-situ and ex-situ prepared CaCO3 NPs. (R2) and (R5) refer to the reaction used to prepare the CaCO3 NPs per Section 3.2.

NPs+DF

NPs+DF

137

It is believed that the cake build up was first initiated by clay particles, which were large

particles, even before the actual filtration started. These particles may deposit under

gravity effect by virtue of their large size. During the NP-based fluid filtration, NPs move

towards the large particles only by virtue of bulk flow. Similar observation was reported

by Chellam and Wisber (1997) for smaller particles transport in cross flow membrane

filters. Moreover, with large accessible surface area NPs can be adsorbed onto the cake

surface readily following travelling for short distances only. This leads to plugging

formation near the surface of the filter cake at early stages of its maturity and Brownian

motion becomes important in controlling particle deposition, since the pore openings are

generally not large.

Tran (2011) modeled the mud cake thickness for deadend filtration as an

exponential function of time until it attains its maximum thickness. Based on our

observation, the mud cake without NPs reached its maximum permeability reduction

after 30 min, whereas more reduction of the mud cake permeability was attained when

NPs were used. Therefore, a phenomenological model for cake growth of NP-based

fluid is proposed based on the experimental results obtained in this work. Similar

modelling approach was generally used for population and biomass growth (Edwards

and Edwards, 2011; Spier et al., 2009). The initial stage of growth is approximately

exponential, and then as saturation begins, the growth slows and at maturity, growth

stops. With this in mind, the following equation can be written to describe cake

thickness in the presence of NPs.

X

hgh

dt

dh mcmc

mc 1 (E5.13)

Equation (E5.13) can be integrated using separation of variables approach as follows.

gdt

X

hh

dh

mcmc

mc

1

(E 5.14)

138

Rearranging the LHS of (E5.14) gives,

mcmcmcmcmc

mc

hX

1

h

1

hXh

X

X

hh

1

1 (E5.15)

The steps below follow,

dt ghX

dh

h

dh

mc

mc

mc

mc (E 5.16)

C gth-X lnh ln mcmc (E5.17)

C gt

h

h-Xln

mc

mc (E5.18)

C gt

mc

mc eh

h-X (E5.19)

gt

mc

mc Aeh

h-X , where c e A (E5.20)

XAeh gtmc 1 (E5.21)

gtmcAe

X (t)h

1 (E5.22)

where, A relative cake growth rate=

0

0

h

h-X, h0= initial cake growth at time t=0

measured experimentally before the actual filtration starts, i.e. hmc(0) = h0 and relative

cake growth rate coefficient, g, is obtained from the exponential differential equation

dt

dhmc =ghmc, g is positive and X is the maximum cake thickness in mm after 30 min.

139

Finally, a model for cake thickness estimation in the presence of NPs can be written as

follows.

gt

0

0

mc

eh

h-X

X (t)h

1

(E5.23)

Assuming that all mud cakes attain their maximum thickness of X (mm) after 30 min of

mud filtration, the proposed cake thickness model (E5.23) can be used as the best

approximation to the mud cake deposition throughout the static filtration period for NP

based fluid. Bezemer and Havenaar (1966) proposed that for certain mud additives, the

mud cake permeabilities can be reduced significantly during the mud filtration process.

Equation (E5.10) shows that mud cake build up can be modeled as an exponential

growth. Mud cake porosity decreases as the NPs deposition increases and,

consequently, the filtrate flow through the cake decreases. With the above assumption,

the increase in cake layer thickness over time can be fitted from the initial cake

thickness until a steady state region is reached. From the API LTLP filter press

experiments after 30 min, the filter cake thickness was measured to be 0.35 mm for the

Fe(OH)3 NP-based fluid and 0.52 mm for the CaCO3 NP-based fluid as shown in

Figures 5.7. The model, (E5.23), was used to estimate the time for filter cake buildup

with the following parameters h0 =0.01 mm (initial cake growth at time t=0 measured

experimentally before the actual filtration starts which is natural cake build up) and X

=0.35 mm for Fe(OH)3 NP-based fluid and h0 =0.01 mm (initial cake growth at time t=0

measured experimentally before the actual filtration starts which is natural cake build

up) and X =0.52 mm for CaCO3 NP-based fluid. From the model curves mud cake

thickness attained its maximum thickness, hmc= 0.35 mm at 40 min for Fe(OH)3 and

0.52 mm 40 min for CaCO3 NP-based fluid, showing a fair agreement with the

experiments. The time-dependent behavior of the filter cake build up of NP-based fluids

to a maximum value was captured. Similar time dependent mud cake growth behavior

was also investigated in the literature (Tran, 2011; Wu et al., 2005 ; Mackley and

Sherman,1992).

140

Figure 5.7: Variation of Mud cake thickness with time for NP-based fluid.

5.2 NP based fluid transport using Stoke-Einstein equation

Nanoparticle transport during filtration is predominantly influenced by convection and

Brownian diffusion with negligible contributions from gravitational settlings and shear

induced diffusion (Hwang et al., 1998; Mcdonogh et al., 1984; Song and Elimelech,

1995). Previous studies found that between microparticle and nanoparticle transport

models, particulate deposition greatly depends on the particle size (Ding and Wen,

2005; Kleinstreuer et al., 2008). This current model suggested that the nanoparticles get

dispersed due to diffusion and convection, whereas microparticle transport is governed

by convection and sedimentation. For micron sized particles Brownian diffusion

mechanism is not important (Ives, 1970).

0.52 mm

0.35 mm

0.52 mm

0.35 mm

141

The general diffusion equation for nanoparticle transport can be written as

0 Jt

CNP

. (E5.24)

At a given location within the flow, the total flux, J of particle migration can also be

described at any time as (Zamani, 2009; Ding and Wen, 2005),

advectionDiffusion JJJ (E5.25)

The diffusion flux of NPs, DiffusionJ (moles/cm2-s), can be expressed as a function of the

concentration gradient using the first law of Fick, which in one dimension can be written

as

x

C-DJ NP

NPDiffusion

(E5.26)

For unsteady diffusion, Fick’s second law is,

xx

NPNP

NP CD

t

C (E5.27)

Equation (E5.26) can be simplified as:

NPNPDiffusion CDJ . (E5.28)

where, NPD is the diffusion co-efficient of nanoparticles due to Brownian effect. In the

second term of (E5.25), advectionJ , is the overall convection or flow and considered as an

associated flux called advection flux and can be expressed as:

advectionJ = U. NPC (E5.29)

where, U is the particle velocity induced by fluid flow (cm/s).

142

Then, the convection diffusion, equation (E 5.24), simplifies to

0 )C U.C.D(t

CNPNPNP

NP

. (E 5.30)

Equation E5.30 can be further simplified to (Ghesmat et al., 2011; Zamani, 2009; Bird et

al., 2002):

NP2

NPNPNP C.DC .(U.t

C

) (E 5.31)

Diffusion happens randomly due to molecular motion. In the absence of surface

interaction forces, nanoparticles diffusivity due to Brownian motion replaces ordinary

diffusion coefficient (Tien, 1989). Small particles can be diffused to the cake surface

where they can stick. Therefore, the diffusion coefficient of nanoparticles in liquid can be

calculated by using the Stokes-Einstein relationship:

NP

BNP

d3

TkD

(E 5.32)

where Bk = 1.38X10-23 J/K is the Boltzmann constant, T is the absolute temperature, μ is

the dynamic viscosity, and dNP is the nanoparticle diameter. If mass deposition rate

(micro or nanoparticles) from the drilling fluid over API filter paper having 2.7 μm grain

size is considered, Peclet number, Pe, for mass transfer is defined as

NP

g

D

UdPe

(E 5.33)

where, is the characteristic velocity determined from the filtrate flow rate till 30 min

and is the filter grain diameter. Peclet number reflects the ratio of particle migration

due to convection to that due to Brownian diffusion. If Pe >>1, transport by convection

and/or sedimentation, in the case of microparticles, is the main driver, whereas if Pe

<<1, diffusion dominates transport (Russel et al., 1999). Pe was calculated using

143

(E5.33) employing Brownian diffusion coefficient, (E5.32) and base oil viscosity of 3.44

cP at 25°C and different particle sizes at LTLP API filtration condition. The effect of

Peclet number is associated mainly with particle size as shown in Figures 5.8 and 5.9.

The calculated data tables were listed in Appendix C. From Figure 5.8 it can be shown

that Peclet number increases rapidly with increasing particle sizes. Both DF control

samples exhibited similar trend close to each other. Particles in the micro domain, e.g.

2-200 μm which is typical for conventional LCMs, display relatively high Peclet number,

Pe>>1. This, in turn, suggests that for LCM and clay particles the effect of Brownian

motion is not important. Conversely, many earlier studies have found that the migration

of NPs can be described by Brownian motion (Phillips et al., 1992; Lam et al., 2004;

Ding and Wen, 2005). In the present study, in the low-Pe regime, dispersion, molecular

diffusion, plays a very important role in NPs transport as shown in Figure 5.9. Although

both Fe(OH)3 and CaCO3 NPs show a similar trend, the slight differences are assumed

to be their size ranges. Migration of nanoparticles to the filters cake may also

encourage formation of clusters/aggregates due to diffusion on the surface.

Microscopically, it is the balance between different forces (hydrodynamic, van der

Waals, electrostatic, steric due to the presence of surfactant rich NPs) that affects the

force-distance dependency between the particles. Note that shear thinning behavior of

fluid was not considered in this study. This is because particle concentration of NPs

used in drilling fluid formulation, 1-4 wt%, was low and no significant effect on fluid

viscosity was reported as detailed in the results and discussion section.

144

Figure 5.8: Effect of particle sizes of DF (dp =2-200 μm in DF) on Peclet number.

Figure 5.9: Effect of Fe(OH)3 and CaCO3 NPs size in DF ranges from 0.001-0.3 μm (1-300 nm) on Peclet number.

dp, microns

P

ecle

t n

um

be

r,P

e

dp, microns

Pe

cle

t n

um

be

r,P

e

145

5.3 Rheology model of NP based fluid

A rheological model describes the relationship between the shear stress and shear rate.

Most drilling fluids are non-Newtonian. Analysis on correlation between shear stress

and shear rate for drilling fluids showed that they generally follow the Bingham plastic

model (Jihua and Sui, 2011; Lee et al., 2009; Agarwal et al., 2009; Peng, 1990).

(E5.34)

In order to better understand the effect of adding nanoparticles on the rheological

properties, data were best fitted to the model in (E5.34). In all cases, the coefficient of

determination, R2, was 0.99 suggesting an excellent fit.

All the muds had almost similar viscosities in the high shear range although

NP-based mud was slightly thinner than the base invert emulsion mud as shown in

Figures 5.10 and 5.11. Compared with the drilling fluid and the drilling fluid with LCMs,

increasing the shear rate did not increase the shear stress to the same extent in the

presence of ex-situ and in-situ prepared Fe(OH)3 NPs. This observation can be

attributed to the fact that smaller particles were dispersed more effectively than the

larger particles and, in the case of Fe(OH)3, NPs provided bridging between clay

particles due to the larger degree of interaction than the CaCO3 NPs. In the low shear

range, however, the CaCO3 NP-based mud had similar effective viscosities. The shear

thinning property of NP-based fluid would be advantageous in providing better hole

cleaning. Similar fluid flow behavior trends were reported by Simpson et al. (1979),

Baird and Walz (2006) and Srivatsa (2012).

Plastic viscosity (PV) measures the internal resistance to fluid flow attributable to

the amount, type and size of NPs present in the fluid system (Cai et al.,2012). If PV is

excessive, the equivalent circulating density will be excessive. This results in an

increased risk of lost returns (Shelton, 2005). PV is generally desired to be as low as

possible. Conversely the yield point is the resistance to initial flow and it represents the

stress required to start the fluid movement. This resistance is believed to be due to the

electrical charges located on or near the surface of the particles (Lyons, 1996). If this

146

property is too high, the consequence will be the same as for high plastic viscosity

(Shelton 2005). By fitting the experimental results to the model, the yield strength and

plastic viscosity were obtained for different drilling fluids as shown in Table 5.2. Both

plastic viscosity and yield strength decreased upon adding nanoparticles except for

CaCO3.

Figure 5.10: Bingham Plastic model in the presence and absence of Fe(OH)3 NPs

Figure 5.11: Bingham Plastic model in the presence and absence of CaCO3 NPs. (R5)

and (R2) refer to the reaction used to prepare the particles as detailed in section 3.2.

y = 0.0253x + 3.3138

y = 0.0263x + 2.9952 R² = 0.9932

R² = 0.9932

y = 0.0179x + 1.3536

y = 0.0179x + 1.3536 R² = 0.9985

R² = 0.9985

0

5

10

15

20

25

30

35

0 200 400 600 800 1000 1200

DF

DF+LCM

DF+Ex-situ NPs

DF+In-situ NPs

S

he

ar

Str

ess (

Pa

)

Shear rate (Sec-1)

y = 0.0143x + 1.6547 R² = 0.9905

y = 0.0154x + 1.682 R² = 0.9887

y = 0.0167x + 1.2007

y = 0.0167x + 1.2007

R² = 0.9945

R² = 0.9945

0

2

4

6

8

10

12

14

16

18

20

0 200 400 600 800 1000 1200

DF

DF+Ex-situ NPs (R2)

DF+In-situ NPs (R2)

DF+In-situ NPs (R5)

S

hear

Str

ess (

Pa)

Shear rate (Sec-1)

147

Due to the differences in the nature of NPs, their concentration, each sample behaves

differently. Moreover, surfactant act as a dispersion agent of NPs generally decreases

the viscosity. Model and experimental results were very close for Fe(OH)3 NPs and

CaCO3 NPs, nevertheless. Figures 5.10 and 5.11 show a perfect straight line. Drilling

fluid and drilling fluid with LCM displayed high plastic viscosity (PV) and yield point (YP),

which could result in increased equivalent circulating density (ECD) (Nicora, 2001) and

also could show detrimental effect on the rate of penetration (ROP). An increase in the

population of solid particles in fluid slows down the rate of penetration (ROP). A

desirable invert emulsion fluid would be the one which not only has a low PV but shows

good low shear viscosity and yield point (Okrajni and Azar,1986; Becker et al.1991;

Maghrabi et al.2011). NP-based fluid with low PV and low shear yield point serves

improved sag resistance (Bern, 1996) and cuttings carrying capacity. It was observed

that, in case of Fe(OH)3 NP-based fluid, PV and YP decreased by 26% and 60%,

respectively, in comparison with the base formulations. But in case of CaCO3 NPs,

these changes were insignificant. Javeri et al. (2011) and Paiaman and Al-Anazi (2008)

investigated the plastic viscosity and yield point of drilling fluid after addition of NPs.

With the addition of 2~3 vol% carbon black and silica NPs to the mud, these

researchers reported decreased plastic viscosity and yield point.

148

Table 5.2: Experimental and Bingham Plastic viscosity and Yield point.

In conclusion , the study shows that:

a) A plot of fluid loss versus square root of time gave a linear relationship for drilling

fluid without NPs arising from spurt loss. Conversely a non-linear relationship

was applicable to NP-based fluid, due to absence of spurt loss.

b) It was found that more than 90% of mud cake premeability reduction occurred

using NP-based fluid, whereas conventional LCM reduced the permeability by

77%.

c) It was also shown that nanoparticles transported in filtration was predominantly

influenced by the Brownian diffusion.

d) Bingham plastic model satisfied for both Fe(OH)3 and CaCO3 NP-based drilling

fluid.

NP Used Samples Types PV and YP values, Pa

Exp. PV Bingham,PV Exp.YP Bingham,YP

Fe(OH)3

DF

0.023 0.025 3.4 3.3

DF+LCM

0.023 0.026 5.3 3.0

DF+exsitu NPs

0.017 0.018 1.3 1.4

DF+insitu NPs 0.017 0.018 1.4 1.4

CaCO3

DF

0.015 0.017 2.4 1.2

DF+exsitu NPs (R2)

0.013 0.014 2.4 1.7

DF+insitu NPs (R2)

0.013 0.015 3.4 1.7

DF+insitu NPs (R5)

0.015 0.017 2.4 1.2

149

Chapter Six: Conclusions‎, Contributions to Knowledge‎ and Recommendations

In this work novel methods for in-house preparation of stable nanoparticles in drilling

fluids were introduced. The product nanoparticles were very stable in the invert

emulsion as well as the water-based muds. Different nanoparticles were prepared and

at the low concentrations involved, these particles did not alter the properties of the

drilling fluid; including pH, viscosity, density (except for weighting materials based NPs)

etc. Despite the low concentrations, the particles significantly reduced the fluid loss and

improved the drilling fluid lubricity property.

6.1 Conclusions

The specific conclusions of this research program are summarized as follows.

1. Various in-house NPs were prepared by ex-situ and in-situ method

2. In-house NPs addition to drilling fluids dramatically improved fluid performance

3. Low concentration of NPs were used ( 1-5 wt%) for fluid formulation

4. Addition of ex-situ NPs and in-situ NPs reduced the LTLP fluid loss by 70-80%,

exhibited thin mud cake and similar performance was obtained at HTHP filtration

5. Density and pH of the drilling fluids were unaltered after addition of in-house

NPs; the only exception to the weighting materials type of NP-based drilling fluid

6. Introducing NPs to drilling fluid did not change the rheology of the drilling fluid at

low as well as high shear range

7. Addition of ex-situ NPs and in-situ NPs enhanced the lubricity property of the

drilling fluid

Drilling fluids have a wide range of chemical and physical properties which are

specifically optimized for drilling conditions and the special problems that must be

handled while drilling a well. In this work, water and invert emulsion based muds have

been formulated to incorporate in-house prepared nanoparticles starting from aqueous

precursors at a very reasonable cost. Relative to commercial NPs, in-house prepared

ones produced much better fluid loss control, probably due to limited dispersion

150

associated with the commercial counterparts. The in-house techniques developed in

this study for the preparation of NPs and NP-based drilling fluids followed two different

routes; namely ex-situ and in-situ. The ex-situ technique for nanoparticle preparation

consisted of providing aqueous-based precursor solutions for forming the nanoparticles,

mixing the precursor solutions under 200 rpm at 25oC, and adding the mixed precursor

solution to the drilling fluid. To form the nanoparticle-containing fluid, the fluids were

mixed/sheared at 2500 rpm to achieve a uniform mixture using Hamilton beach mixer.

The purpose of stirring is to form a stable and homogenous emulsion by breaking large

liquid drops into smaller drops. In case of in-situ nanoparticles aqueous precursor salts

are added to two separate samples of the drilling fluid under 200 rpm of mixing at 25oC.

NPs were formed via precipitation reactions, which took place in the drilling fluid upon

mixing the two drilling fluid samples and shearing at 2500 rpm to achieve a uniform

mixture using Hamilton beach mixer. Formation of nanoparticles in this way minimizes

particles aggregation allowing easier handling than ex-situ prepared NPs. The method

developed in this work is versatile and different types of precipitates of NPs were

prepared using both techniques; including Fe(OH)3, CaCO3, FeS and BaSO4. Complete

investigation was, nevertheless, focused on the performance of Fe(OH)3 and CaCO3

NPs. Moreover, a method was developed for the in-situ synthesis of calcium carbonate

NPs using carbon dioxide, CO2, gas allowing inside the wellbore to generate this NPs.

Comparing the two different methods of in-situ CaCO3 NPs preparation, the second

method showed the lowest spurt loss and total filtrate volume. The mechanism leading

to CaCO3 NPs formation in this method involved the nucleation of CaCO3, particle

growth and may induce a Brownian diffusion with clay particles. Spurt losses observed

with all the in-house NPs are acceptable from a drilling point of view.

With all the different particles prepared in this study, the incorporation of the NPs

in invert emulsion fluid system reduced the fluid loss substantially at relatively low

concentration of the NPs. NPs as lost circulation materials are typically designed to

accomplish three goals: 1) to bridge across the face of fractures, 2) to prevent the

growth of any fractures that may be induced during drilling and 3) to change the mud

density while keeping the viscosity almost constant when NPs are used as weighting

151

materials. It was found that muds were quite stable and offered a wide range of

nanoparticle sizes that controlled the fluid loss efficiently and effectively, with and

without conventional loss circulation materials (LCMs), e.g. Gilsonite. Commercial invert

emulsion drilling fluid without NPs and LCM and with 1.6 wt% LCM only were

considered as baseline drilling fluids for comparative evaluation of API filtration

experiments. The NPs concentration was varied between 1 wt% and 5 wt% for both in-

situ and ex-situ prepared particles. At the low concentration ‎of NPs, it is most likely that

the particles interacted with the rest of the mud constituent ‎rather than merely

aggregating. ‎These muds also offer good lubrication and are, therefore, appropriate for

applications in drilling extended reach wells.

Test results showed that in-situ and ex-situ prepared NPs were not uniform in

size and shape. Testing also demonstrated that NPs covered a wide range of particle

sizes from nanometer to micrometer scale, but 70-80% of the prepared particles fell into

1-60 nm range. Properly prepared, these NPs have the potential to build structural

barriers in the fluid loss paths according to their size. Although different shapes of NPs

were visible in the TEM images but this property was not investigated in the current

study. In-house prepared NPs per current work behaved as a new generation fluid loss

additives. For example, it was found that Fe(OH)3 NPs at 1 wt% showed excellent

performance. They prevented fluid loss of invert emulsion mud up to 70-80%, and

contributed to a thin mud cake, < 1 mm, for LTLP and < 2 mm for HTHP conditions. At

this low concentration, the in-house prepared NPs out performed conventional LCMs,

e.g. Gilsonite to a great extent. 4 wt% CaCO3 NPs represent the optimum concentration

in which the volume of filtrate reached minimum values and better arrangement of

particles occurred in the filter cake surface turning into an impermeable cake. Besides,

the spurt losses are found lowest using NPs. Approximately, 60 % of the API fluid loss

reduction was happened using both ex-situ and in-situ prepared CaCO3 NPs. The

samples exhibited a very low fluid loss at low temperatures and same relative

performance at high temperatures. Similarly, addition of 3 wt% BaSO4 and FeS NPs in

drilling fluid reduced the API LTLP fluid loss more than 80%. The low concentration of

NPs addition to mud was mainly dependent on the physio-chemical nature of NPs and

152

stability of the mud system. Therefore, NP-based drilling fluids discussed in this study

are viable and robust systems and compatible to other systems that already exist in the

market. It could potentially have an impact on the total drilling cost, especially in

complex and troublesome drilling operations.

The experimental results also indicated an unchanged viscosity at the low

concentration of NPs, although a slight decrease in gel strength was experienced. It

was also found that Fe(OH)3 and CaCO3 nanoparticles exhibited good load-bearing

capacity, anti-wear and friction-reducing properties. The in-house NPs can be more

effectively dispersed in the drilling fluid if they are formed in-situ and more than 40%

friction of co-efficient is attainable.

Darcy’s law was used to interpret the LTLP fluid loss experimental only. Due to

the similar degree of fluid loss reduction close to LTLP one, HTHP results can also be

interpreted by Darcy’s law which is not included in the current study. A plot of fluid loss

versus square root of time gave a linear relationship for drilling fluid without NPs arising

from spurt loss. Conversely a non-linear relationship was applicable to NP-based fluid,

due to absence of spurt loss. It is also noted that NP-based drilling fluid did not follow

Darcy equation at the initiation of filtration and, therefore the initial region was found flat.

It was found that more than 90% of mud cake premeability reduction occurred using

NP-based fluid, whereas conventional LCM reduced the permeability by 77%. The

model results indicated that nanoparticles reduced the premeability instantly and fluid

invasion decreased siginificantly. It was also shown that nanoparticles transported in

filtration was predominantly influenced by the Brownian diffusion. Compare with the

drilling fluid alone and drilling fluid with LCM, increasing shear rate did not increase the

same extent of shear stress of NP-base fluid (both ex-situ and in-situ prepared), which

can be attributed to the fact that smaller particles were dispersed more effectively than

the larger bulk particles and provided bridging between clay particles due to their larger

surface area. In all cases, the coefficient of determination, R2, for Bingham rheological

model, was 0.99~0.98.

Applying the in-house techniques to prepare NPs in water based muds ‎resulted a

very progressive behavior. Both Fe(OH)3 and CaCO3 NPs reduced the fluid loss by

153

approximately 30%. NP-based water mud had a much higher initial filtrate rate but after

30 min it was closely two times greater than the invert emulsion NP-based mud.

From the aforementioned discussions, it can be concluded that the use of

nanoparticles in the drilling fluid at the right concentration and adoption of a specific

preparation method leads to a fluid with desirable properties in terms of mud density, pH

and rheological behavior. The addition of nanoparticles does not alter the optimum

values for these properties from the base fluid. Formation damage due to filtrate and

solids invasion is a major contributor to cost, lost time and lost production. One of the

critical factors in avoiding formation damage during drilling is obtaining surface bridging

on the formation face with minimum in-depth solids penetration. Nanoparticles work in

emulsion based fluids, even at high temperatures, providing a thin filter cake that gives

maximum formation protection at minimum concentration and cost. Thin filter cake is

important for reducing differential sticking problem and excessive drag in extremely

permeable formation. Tailor made nanoparticles with specific characteristics is expected

to play a promising role in solving circulation loss and other technical challenges faced

with commercial drilling fluid during oil and gas drilling operations.

Results obtained from this study, give rise to the following characteristics for the

nanobased drilling fluid.

Thin and firm filter cake.

Minimal fluid invasion, since sealing takes place at the surface, which leads to

minimal formation damage.

Extreme high temperature stability.

Only low NP concentration is required.

Time and cost savings.

These properties entail the following practical applications.

• Less Fluid loss = Money saving.

• Lower torque and drag = Increase extended reach well.

• Reducing differential pressure sticking problem = Less non-productive time.

154

• Less solid concentration in mud = Reduce formation damage and increase

productivity index.

6.2 Original contributions to knowledge

1. Development of new in-house methods for preparing wide variety of NPs in drilling

fluids. As per these methods, the birth place of NPs can be outside the drilling fluid, ex-

situ, or inside the drilling fluid, in-situ. In-situ prepared NPs communicated better with

the resultant NP-based drilling fluid, and generally led to high fluid loss prevention,

under LTLP and HTHP conditions.

2. The in-house methods developed in this study were applicable for both invert

emulsion and water based drilling fluids.

3. Use of nanoparticles as a fluid loss additives, lubricity additives and weighting

materials for the drilling fluid, wherein the drilling fluid comprises a base fluid and

nanoparticles present in an amount of about 5 wt % or less.

4. The nanoparticles have a particle size between about 1 and 120 nm, wherein a

majority of the nanoparticles have a particle size between 1 to 30 nm.

5. NPs addition reduced the total solid content usage in the drilling fluid.

6. NPs were found more effective in 80:20 (V/V) Oil/water invert emulsion drilling fluid

interms of fluid loss control.

7. Addition of NPs unchanged the density (except for weighting materials based NPs),

viscosity and pH of the final fluid.

8. The nanoparticles were selected from the group consisting of metal hydroxide (iron

hydroxide) metal oxide (iron oxide), metal carbonate (calcium carbonate), metal sulfide

(iron sulfide) and metal sulfate (barium sulfate). Nanoparticles were formed in situ in the

drilling fluid or formed ex situ and added to the fluid.

6.3 Recommendations for future research

The following recommendations are proposed for the future studies:

• Based on the static filtration tests, it is necessary to conduct the dynamic filtration

test to simulate the bottom hole conditions. Dynamic filtration test determines if

155

the fluid is properly conditioned to drill through highly permeable formations. For

this purpose, FANN 90 dynamic filtration apparatus can be used. It utilizes

ceramic cores available in a range of different permeabilities. A new filtration

model can be applied to measure the dynamic filtration rate. This model could

consider the effect of shear-induced migration where NPs move regions of higher

shear rate to regions of lower shear rate and deformation of NPs/LCM under

stress that exist in a filter cake. Cake deposition index (CDI) can also be

measured from the dynamic filtration data.

• One of the future filtration studies could be to investigate the formation damage

of known core samples by different nanoparticles. Most of the NPs used in

literature entered into pores and plugged them in the producing

direction, reducing the flow and permeability of the rock for production and finally

leading to formation damage. The NPs used in the current experiments provided

a better bridging effect on the filter cake and did not pass through the filter paper

into the filtrate. Therefore, it is believed that it would be more realistic to test the

filtercake core permeability to investigate the flow and production and return

permeability issues caused from plugged filter cakes. The Permeability Plugging

Apparatus (PPA) can be utilized to measure fluid loss using ceramic discs

available in a variety of permeabilities (5 micron to 190 micron) to simulate

reservoir pore throat diameters.

• Future efforts on field tests could enhance the potential development of low cost

drilling fluid with NPs.

• More studies should be performed on the usage of all existing drilling fluid

additives in nanoscale. A recommendation for future work is the incorporation of

pre-mix bentonite rich NPs, polymeric materials with NPs, nano emulsifiers and

cattle manure with NPs in the drilling fluid. XRD results in the current study

suggests that Fe(OH)3 NPs acted as intercalating agents and had entered into

the crystallite layers of bentonite clay. It is believed that such a structure of

nanometal-clay composite could improve drilling fluid properties and may offer

better functionality than regular bentonite without the requirement of other

156

expensive additives. Nanometer polymer could penetrate to pores, deposit on the

surface and form low permeable mud cake. The mud cake produced by the

insoluble deformable polymer with colloidal materials could reduce the pressure

differential between the hydrostatic pressure of the mud and the pore pressure

around the surface of wellbore and thus prevent differential sticking, promote

borehole stabilization and avoid formation damage (Benaissa, 2006). Similarly

Manea (2011) designed nanopolymer gel using Xanthan gum and fluid loss

additives, and Saboori et al. (2012) used nanoCMC polymer to decrease the

water loss and mud cake thickness in mud drilling. Using the aqueous polymer

base, crosslinked with NPs used in our study can also form nanocomposite

polymeric gel to cure the fluid loss. Lécolier et al. (2005) showed that

nanocomposite polymeric gel could be pumped into naturally fractured

formations or voids, and plug off a wide range of cracks. Nano-emulsion

prepared through a one-step method (Mei et al., 2011) showed a good lubrication

and long term stability. Similarly nano asphalt emulsion can be tested to increase

the lubrication and fluid loss control. Cow dung (manure) is a material highly

cherished by rural dwellers used as a binding agent for plastering houses

(Issaka, 2012). In many developing countries manure with clays are mixed and

used in boring pipe into an underground aquifer to lift water for irrigation. Manure

serves as a binding agent and gives plaster more body. It also contains small

natural fibers that provide additional tensile strength as well as reduce cracking

and water erosion (Guelberth and Chiras, 2003). Manure therefore, could prevent

fluid loss as well enhance lubrication.

157

References

AADE.,“Shale Shaker and Drilling Fluids Systems:Techniques and Technology for Improving Solids Control Management”, Gulf Professional Publishing, USA,(1999)

Abdo,J., and Haneef,D., “Nanoparticles: Promising solution to overcome stern drilling problems,”N TI-Nanotech 2010,www.nsti.org,ISBN 978-1-4398-3415-2,Vol.3, (2010) Abdullah, .F.,“Nanoparticle(Capped Wolfram (VI) oxide) as a additive in lubricant”,

ICCBT,pp 347-356,(2008) Abrams,A., “Mud design to minimize rock impairment due to particle invasion”,JPT,pp

586-592,(1977) Abu Tarboush,B., and Husein,M., “Adsorption of asphaltenes from heavy oil onto in situ

prepared NiO nanoparticles”, J. Colloid Interface ci.,378,pp 64– 69,(2012) Adekomaya, O. A., and Olafuyi, O., “An experimental study of the effect of contaminants

on the flow properties of oil based drilling mud”, Petroleum & Coal 53 (4) 315- 319,(2011)

Adin, A., Baumann, E.R., and Cleasby, J.L., “The application of filtration theory to pilot- plant design”, J. AWWA, 71, (1979).

AEUB (ALBE TA ENE GY AND UTILITIE BOA D), “Guide 50: Drilling Waste Management”,http://www.aer.ca/documents/directives/Directive050_1996.pdf, (1996)

Agarwal,S.,Walker,M.L.,Prieve,C.D., and oong,Y., “Using Nanoparticles and Nanofluids to tailor transport properties of drilling fluids for HPHT operations”,AADE NTCE-18-05,(2009)

AL-Bazali, M.T., Zhang,J., Chenevert,E.M., and Sharma,M.M., “Measurement of the Sealing Capacity of Shale Caprocks”, SPE Annual Technical Conference and Exhibition, 9-12 October 2005, Dallas, Texas,(2005)

Al-Hitti,H.A.,Al-Assaf,A.S.,and Abrahim,S.D., “Reduction of formation damage due to drilling muds”,Journal of Engineering,11(1),pp 21-32,(2005)

Al-Riyama,K.,and Sharma,M.M.,“Filtration properties of oil-in-water emulsion containing solids”, PE Drilling & Completion,pp. 164-172,(2004)

Altun,G., and Serpen,U., “Investigating Improved Rheological and Fluid Loss Performance of epiolite Muds under Elevated Temperatures”,Proceedings World Geothermal Congress 2005 Antalya, Turkey, 24-29 April (2005)

Amanullah,M., Al-Arfaj, K. M, and Al-Abdullatif.Z, “Preliminary Test esults of Nano- based Drilling Fluids for Oil and Gas Field Application”, PE/AIDC 139534, 1-9, (2011)

Amanullah,M., and Al-Abdullatif,Z., “Preliminary test results of a water based nanofluid”, The 8th International Conf.& Exhib. on chemistry in industry, Manama,Bahrain, 18-20 October (2010)

Amanullah,M., and Al-Tahini,M.A., “Nano-Technology – Its Significance in Smart Fluid Development for Oil and Gas Field Application,” SPE Saudi Arabia Section Technical Symposium, AlKhobar, Saudi Arabia, (2009)

158

Amoco., “Training to educe Unscheduled Events, 3rd Edition.Amoco EPTG Drilling Technology Teams”, Amoco Production Company,(1996)

Anisa,I.N.A., and Nour,H.A., “Affect of Viscosity and Droplet Diameter on water-in- oil (w/o) Emulsions: An Experimental Study”, World Academy of Science, Engineering and Technology, 38, (2010)

API RP 13B-2,“ ecommended Practice for Field Testing Oil-Based Drilling Fluids”,Fifth edition, (2012)

API RP 13B-1 “ ecommended Practice for Field Testing Water-Based Drilling Fluids”, Third Edition,(2003)

Argiller,F.J., Audibert,A., and Longeron.D., “Performance Evaluation and Formation Damage Potential of New water-based drilling Formulas”, SPE Drill & Completion, 14(4), pp 266-273,(1999)

Ashworh,J., and Walker,V., “Protocol for Non-toxic Concentrations of Drilling Fluid Additives”, Western Canada Microtox Users Committee (WCMUC) eport, http://www.wcmuc.com/links.htm,(2006)

A ME, “Drilling Fluids Processing Handbook”, Elsevier, U A,(2005) Aston,M.,Mihalik,P.,Tunbridge,J.,and Clarke,S., “Towards zero fluid loss oil based

muds”, SPE 77446, SPE Annual Technical Conference and Exhibition, San Antonion, Texas, USA,29 September-2 October, (2002)

Ballard, A.D., and Massam, J., “Precipitated weighting agents for use in wellbore Fluids”,WO2008033838A2, (2009)

Bandyopadhyaya,R.,Kumar,R., and Gandhi,K.S.,“Modelling of CaCO3 Nanoparticle Formation During Overbasing of Lubricating Oil Additives”, Langmuir,17, pp.1015-1029,(2001)

Balek,J., and Šubrt ,J.,Thermal behaviour of iron oxide hydroxides,Pure & App. Chem., 67(11), pp. 1839-1842, (1995)

Baker Hughes INTEQ, Drilling Engineering Workbook”, (1995 ) Bakerhughes, www.bakerhughes.com, “Drilling fluids reference manual”,(2006) Bailey, L., Bock, E., Jacques, S., Boassen, T., Selle, O., Argillier, J-F., and Longeron,D.,

“Particulate Invasion from Drilling Fluids, SPE 54762, SPE European Formation Damage Conference, The Hague, The Netherlands, 3 1 May-l June ,(1999)

Baird,J.C., and Walz,Y.J., “The effects of Added nanoparticles on aqueous kaolinite suspensions I. tructural effects”, Journal of Colloid and Interface Science 297, pp.161–169, (2006)

Barkman, J.H., and Davidson, D.H., “Measuring Water Quality and Predicting Well Impairment”, JPT, pp. 865-87,(1972)

Becker, T.E., Azar, J.J. and Okrajni, S.S., “Correlations of Mud Rheological Properties with Current Transport Performance in Directional Drilling.” PEDE, pp 16-24,(1991)

Beeson,C.M., and Wright,C.C., “Loss of mud solids to formation pores, Pet Eng.,pp 40- 52,August,(1952)

Benaissa,S.,“ ealant improved drilling in depleted sands”,Drilling Contractor,pp 53- 54,(2006)

159

Bennion, B.D.,Thomas,B.F.,Jamaluddin, AKM., Ma, T.,and Agnew,C., “Formation Damage and Reservoir Considerations for Overbalanced and Underbalanced CT Operations”,6th International Conference onCoiled Tubing Technologies held in Houston, Texas,Oct. 27-29, (1997)

Bern, P.A., “The Influence of Drilling Variables on Barite ag.” PE 36670, (1996). Berret,JF., “Interactions between Polymers and Nanoparticles: Formation of Hybrid

Aggregates”, Colloids and Surfaces, (2004) Bezeme,C., and Havenaar,I., “Fltration behavior of circulating drilling fluids,” PE

Journal,Trans,AIME,237, SPE-1263-PA,6(4):292-298,(1966) BHI., “Fluid Facts Engineering Hand Book”, Part # 008902097,Barker Hughes Inteq,

Rev B, (1998) Bird, R.B., Stewart, W.E.,and Lightfoot, E.N., “Transport Phenomena”, Wiley, New York,

(2002) Browning, W.C., Perricone, A.C., “Clay Chemistry and Drilling fluids”, Drilling and Rock

Mechanics Symposium, 23-24 January 1963, Austin, Texas,(1963) Bumajdad,A., Ali,S., and Mathew,A., “Characterization of iron hydroxide/oxide

nanoparticles prepared in microemulsions stabilized with cationic/non-ionic surfactant mixtures”,Journal of Colloid and Interface Science 355, pp.282–292,(2011)

Buzea,C., Pacheco Blandino,I.I., and obbiie,K., “Nanomaterials and nanoparticles: ources and toxicity”, Biointerphases,2(4), pp M 17 - MR172,(2007)

Byck,H.T., “The effect of formation permeability on the plastering behavior of mud fluids”,AP Drilling and Production Practice, 40-040,(1940)

Caenn,R.,Darley,HCH.,and Gray,R.G.,“Composition and properties of drilling and completion fluids”, sixth edition, Gulf professional publishing publications, UK,(2011)

Canter,N.,“Boron nanotechnology-based lubricant additive”, Tribology and lubrication technology,(2009)

Cai, J., Chenevert, M.E., Sharma, M.M., and Friedheim, J.,“Decreasing Water Invasion Into Atoka Shale Using Nonmodified ilica Nanoparticles”, SPE Drill & Compl 27

(1): 103-112. SPE-146979-PA, (2012) Cai,J.,Navrotsky,A., and Suib,S.L.," Synthesis and anion exchange of tunnel structure

akaganeite,Chem Matter,13,pp 4595-4602,(2001) Cargnel,D.R.,and Luzardo,P.L., “Particle size distribution of CaCO3 in Drill-in Fluids:

Theory and Applications”, PE 53937, PE Latin American and Caribbean Petroleum Engineering Conference held in Caracas,Venezuela,21-23 April,(1999)

Chang,Li., and Friedrich,K., “Enhancement effect of nanoparticles on the sliding wear of short fiber- reinforced polymercomposites: A critical discussion of wear mechanisms”, Tribology International,43 ,2355–2364,(2010)

Chang,Z-Y., Breeden,D.,McDonald,M., “The use of Zinc Dialkyl Dithiophosphate as a Lubricant Enhancer for Drilling Fluids particularly Silicate-based Drilling Fluids”, SPE International Symposium on Oilfield Chemistry, The Woodlands, Texas, USA, 11-13 April,(2011)

160

Chelton, H.M., “Darcy's Law Applied to Drilling Fluid Filtration”, PE 1649, (1967) Chellam,S., and Wiesner,R.M., “Particle Back-Transport and Permeate Flux Behavior in

Crossflow Membrane Filters”, Environ. Sci. Technol., 31 (3), pp 819–824,(1997) Chenevert,M.E., and Sharma,M.M.,“Maintaining shale stability by pore plugging,” U

Patent 0314549, (2009) Chenevert, M.E., “Filter Cake tructure Analysis Using the canning Electron Microscope”,

22208-MS, SPE, (1991) Chenevert,M.E., and Sharma,A.K.,“Permeability and effective pore pressure of shales,”

SPE/IADC 21918, (1991) Chilingarian,G.V., “ tudy of dispersing agents”, Journal of Sedimentary Research,

22(4),pp229-233,(1952) Chilingarian,G.V., and Vorabutr,P., “Drilling and Drilling Fluids”. Updated textbook

edition,Elsevier Amsterdam,(1983) Chin, W. C., “Formation invasion with applications to measurement-while-drilling, time-

lapse analysis, and formation damage”, Gulf Publishing Company, Houston, Texas,(1995)

Choi,S.U.S., Li, S., Yu, W., Thompson, L.J., “Anomalously increased effective thermal conductivities of ethylene glycol-based nanofluids containing copper nanoparticles,” Applied Physics Letters, 78( 6), pp. 718-720,(2001)

Clark, P.E., and Barkat, O.,“Analysis of Fluid-loss data”, SPE Production Engineering, 5(3),(1990)

Costa,A., “Permeability-porosity relationship: A reexamination of the Kozeny-Carman equation based on a fractal pore-space geometry assumption”,Geophysical Research letters,33(2006)

Crowe, C.W., “Laboratory study provided guidelines for selecting clay stabilizers”, PE, CIM/SPE International Technical Meeting, 10-13 June 1990, Calgary, Alberta, Canada,(1990)

Dampier,L.,http://www.landfood.ubc.ca/soil200/interaction/ions.htm,Soilweb,(2004) Darley, H.C.H., “Designing Fast Drilling Fluids”,JPT, pp. 465- 470,(1965) Deriszadeh,A., “Chemically and Electrically Induced welling in Clays and hale”,PhD

Thesis,University of Calgary,(2012) Dersizadeh,A., Husein,M and Harding,T., “Oil ands Produced Water Treatment by

Micellar Enhanced Ultrafiltration”, Environ. ci. Technol.,44, 1767–1772 ,(2010) de Zwart ,A.H., “Investigation of Clogging Processes in Unconsolidated Aquifers near

Water Supply Wells”,PhD Thesis, Delft University of Technology,(2007) Dickerson,P.J., and Rayborn, J.J., “Method of making a drilling fluid containing carbon

black in a dispersed state”,US 5114597 A, (1992) Dick,A.M., Heinz, J,T., Svoboda, F.C.,and Aston, M., “Optimizing the selection of

Bridging particles for reservoir drilling fluids”, SPE International Symposium on Formation Damage Control, 23-24 February 2000, Lafayette, Louisiana, SPE 58793 23,(2000)

Ding,Y., and Wen,D.,“Particle migration in a flow of nanoparticle suspensions”, Powder Technology,149,pp 84-92,(2005)

Donaldson,C.E., and Chernoglazov,V., “Characterization of drilling mud fluid invasion”, Journal of Petroleum Science and Engineering,1,pp-3-13, (1987)

161

Dyke,V.K and Baker,R., “Drilling Fluids, Mud pumps and Conditioning Equipment”, Univ. of Texas at Austin Petroleum; 1st edition,(1998)

Edwards,W.J., and Edwards,C.T., “Population Limiting Factors”, Nature Education, (2011)

Einstein, A., “On the theory of the brownian movement.” Analytical Phys.19, pp. 289- 306,(1906)

Ellen,H.,Geir,P., and Oystein B,T., “Testing of iron oxides as weight materials for drilling muds”, 6218-MS, SPE,(1976)

Fann 35 Viscometer Instruction Manual, Fann Company,(2008) Farajzadeh,R.,“Produced Water Re-Injection (PWRI) An Experimental Investigation into

Internal Filtration and External Cake Build up”,Msc Thesis, Delft University of Technology,(2004)

Ferguson,C.K., and Klotz,J.A., “Filtration from mud during drilling,” Trans, AIME, 201:29, (1954)

Fernandez,M.J., and Young, ., “Environmentally esponsible Water-Based Drilling Fluid for HTHP Applications”,AADE-10-DF-HO-37,(2010)

Fisk, J.V., and Jamison, D.E., “Physical Properties of Drilling Fluids at High Temperatures and Pressures”, PE Drilling Engineering, 4(4), pp.341-346,(1989)

Fordham E.J., Ladva, H.K.J., Hall, C., Baret, J-F. and Sherwood, J.D., “Dynamic Filtration of Benronite Muds Under Different Flow Conditions”, PE 18038 presented at the 63rd Annual Technical Conference and Exhibition of the Society of Petroleum Engineers held in Houston, TX, October 2-5, (1988)

Fraser,L.J.,Harrington,b.,Albarrazin,C.,Snyder,G., and Donham,F., “Use of Mixed Metal Oxide fluid to combat losses in porous and fractured formations: two case

histories,” AADE-03-NTCE-40, (2003) Fjelde,I., “Formation damages caused by emulsions during drilling with emulsified

drilling fluids”, PE105858, PE international symposium on oilfield chemistry, Houston, Texas, USA, 28 February-2 March, (2007)

Follett, E. A. C., “The retention of amorphous, colloidal 'ferric hydroxide' by kaolinites”, J. Soil Sci. 16,pp. 334- 341,(1965)

García,A., Espinosa,R., Delgado,L., Casals,E., González,E.,Puntes,V.,Barata,C., Font,X., and Sánchez,A., “Acute toxicity of cerium oxide, titanium oxide and iron oxide nanoparticles using standardized tests”, Desalination, 269, pp 136–141,(2011)

Gates, G.L., and Bowie, C.P., “Correlation of Certain Properties of Oil Well Drilling Fluids with Particle ize Distribution”, Unite States Bureau of Mines, R.I. 3645, pp.1-22,(1942)

Ghesmat,K.,Hassanzadesh,H.,Abedi,J and Chen,Z., “ Influence of nanoparticles on the dynamics of miscible Hele- haw flows”, Journal of Applied Physics,109,104907, (2011)

Griffo,A., and Keshavan,M., “High performance rock bit grease”,US Patent 20070254817 A1,(2007)

Grim, R. E., “Clay mineralogy: Science”, v.13S, No. 3507, pp.890-898, (1962) Guelberth, .C., and Chiras,D.,“The Natural Plaster Book: Earth, Lime, and Gypsum

Plasters for Natural Homes”, New Society publishers, Canada,(2003)

162

Hands,N., “Dill-in Fluid educes Formation Damage,increases Production ates”, Oil and gas Journal,13,(1998)

Happel, J., “Viscous flow in multiparticle systems: low motion of fluids relative to beds of spherical particles”, AIChE J, 4(197), (1958)

Henshaw,B.B., “ Cation-exchange constants for clay from electrochemical measurements”, Clays and Clay Minerals, 12(1),pp. 397-421 (1963)

Herzhaft, B., Ragouillaux, A., and Coussot, P., “How to unify low-shear-rate rheology and gel properties of drilling muds: a transient rheological and structural model for complex well applications”, The IADC/SPE Drilling Conference held in Miami, Florida, U.S.A., 21–23 February,(2006)

Huang,T.,and Crews,J.B., “Nanotechnology applications in viscoelastic surfactant simulation fluids”, SPE production & operations,512-517,November,(2008)

Husein,M., Zakaria,M., and Hareland,G., “Novel nanoparticle-containing drilling fluids to mitigate fluid loss”, PCT Application: PCT/CA2012050075 ,(2012a)

Husein,M., Zakaria,M., and Hareland,G., “Use of nanoparticles as a lubricity additive in well fluids”, PCT Application: PCT/CA2012/050688,(2012b)

Husein,M., Patruyo, L., Pereira-Almao,P., and Nassar,N., “H2S(g) Sorption from Oil Phases by Means of Ultradispersed Sorbents”, J. Colloid Interface Sci, 342, pp 253-260,(2010)

Husein,M.M., and Nassar,N.N., and “Nanoparticles Preparation using the single microemulsion scheme, “Current Nanoscience,4, pp 370-380,(2008)

Hoff,T.,Connor,O.R.,and Growcock,F.,“Drilling fluid selection to minimize formation invasion- A new test method”, AADE-05-NTCE-73, (2005)

Hoskin,W.T.,“Drilling fluid additive and method for improving lubricity or increasing rate of penetration in a drilling operation”,U Patent 0204067A1, (2010)

Hwang,K.J., Liu,H.C., and Lu,W.M., “Local properties of cake in cross-flow microfiltration of submicron particles”, J. Membr. Sci., 138, pp.181, (1998)

Irawan,S.,Sum,C.W., and Mohammad,B.K., “ Formulation of drilling fluids for high temperature well application using Sabah Bentonite”,Proceedings World Geothermal Congress,Bali,Indonesia,25-29 April,(2010)

Isambourg,P.,and Marti,J.,“Downhole simulaton cell for measurement of lubricity and differential pressure sticking”, PE/IADC Drilling conference, Amsterdam, Holland, SPE/IADC,52816,(1999)

Issaka,N.R., Buri,M.M.,Tobita,S.,Nakamura,S., and Owusu-Adjei,E., “Indigenous Fertilizing Materials to Enhance oil Productivity in Ghana”, InTech,(2012)

Ives, K.J., “ apid Filtration”, J. Water es., 4(3): 201,(1970) Jamaluddin,S.R., and Ahmad,A.K.Z., “The Effect Of CaCO3 Addition on the

Crystallization Behavior of ZnO Crystal Glaze Fired at Different Gloss Firing and Crystallization Temperatures”, Science of Sintering, 42,pp.345-355,(2010)

Javeri,M.S., Haindade,W.Z and Jere,B.C., “ Mitigating loss circulation and differential sticking problems using silica nanoparticles”, PE/IADC 145840, (2011)

Javora,P.H., and Qu,Q., “Well treatment fluids containing nanoparticles and methods of using same”, U Patent 0082230A1,(2009)

163

Jihua,C., and Sui,G., “ heological Behaviors of Bio-degradable Drilling Fluids in Horizontal Drilling of Unconsolidated Coal Seams” I.J. Information Technology and Computer Science, 3,pp 1-7, (2011)

Jimenez, A.M, Geolet,L.C., Chavez,C.J., and Espin,D., “Method for treating drilling fluid using nanoparticles”, US Patent 6,579,832,(2003)

Jones, RH., and Babson, E.C., “Evaluation of otary Drilling Muds”, API Drilling and Production Practice, pp. 22-33,(1935)

Ju,B.,Fan,T., and Ma,M.,“Enhanced oil recovery by flooding with hydrophilic nanoparticles”,China Particuology,4(1),pp.41-46,(2006)

Juppe,L.H.,Marchant,P.R.,Melbouci,M.,“Environmentally acceptable fluid polymer suspension for oil field services”,U 6620769 B1,(2003)

Kabalah-Amitai,L.,Mayzel,B.,Kauffmann,Y.,Fitch,N.A.,Bloch,L.,and Gilbert,UPA.P., “Vaterite Crystals Contain Two Interspersed Crystal Structures”, 340 (6131), pp.454-457,(2013)

Kanj,Y. M.,Funk,J.J., and Al-Yousif,Z., “Nanofluid coreflood experiments in the Arab-D.” SPE paper 12616,presented at the 2009 SPE Saudi Arabia Technical Symposium and Exhibition held in Saudia Arabia,Alkhobar,May 09-11,(2009)

Ke,M., and Foxenberg.W., “Lubricity of Brine Completion and Workover Fluids”, SPE/ICoTA Coiled Tubing and Well Intervention Conference and Exhibition, 23-24 March 2010, The Woodlands, Texas, USA,(2010)

Keblinski, P., Phillpot,S.R., Choi, S.U.S., and Eastman, J.A., “Mechanisms of heat flow in suspensions of nano-sized particles (nanofluids),” International Journal of Heat and Mass Transfer, 45(4), pp. 855-863, (2002)

Kercheville,J.D.,Hinds,A.A.,Clements,W.R., “ Comparison of environmentally acceptable materials with diesel oil for drilling mud lubricity and sptting fluid formulation”,IADC/ PE 14797,paper presented at the 1986 IADC/SPE Drilling conference held in Dallas,TX,February 10-12,(1986)

Kitchens,L.C., “Metallic nanoparticle synthesis within reverse micellar microemulsion systems”,PhD thesis, Auburn University,(2004)

Kleinstreuer, C., Zhang,Z., and Z, Li.,“Modeling air flow and particle transport/deposition in pulmonary airways”, Respratory Physiology & Neurobiology, 163, pp.128-138, (2008)

Knowles,M.K., and Freeman ,F. S. H. B., “Microscopy and microanalysis of crystalline glazes”,J. Microscopy, 215 pp.257-270,(2004)

Kokal,L.S., “Crude Oil Emulsions”, Petroleum Engineering Handbook, Volume I: General Engineering, Chapter 12,SPE, p536,(2006)

Kostansek,E., “Emulsion,” tandard Article, Kirk Othmer Encyclopedia of Chemical Technology,Vol.10, John Wiley & Sons Inc,pp 114,(2003)

Kostic,M.,Golubovic,M.,Hull,R.J.,and Choi, S.U.S., “One-step method for the production of nanofluids”,U Patent 7,718,033B1,(2010)

Kosynkin,V.D.,Ceriotti,G.,Wilson,C.K.,Lomeda,J.R.,Scorsone,T.J.,Patel,D.A.,Friedheim, E.J., and Tour,M.J.,“Graphene oxides as a high performance fluid-loss control additive in water based drilling fluids”, ACS Appl. Mater. Interfaces,4 (1), pp. 222–227,(2012)

164

Krishnamoorti,R.,“Extracting the benefits of Nanotechnology for the oil industry,”JPT,pp. 24-26, (2006)

Krueger, R.F., “Evaluation of Drilling Fluid Filter Loss Additives Under Dynamic Conditions”,JPT, pp. 90-98,(1963)

Kumar,A., “Fluid loss as function of position around the wellbore”, AADE-10-DF-10-18, (2010)

Lam,Y.C., Chen,X., Tan,K.W., Chai,J.C and Yu,S.C.M., “Numerical investigation of particle migration in Poiseuille flow of composite system”, Composites Sci. and Technol., 64 ,pp. 1001–1010,(2004)

Lai,H.C., Lob,L.S.,and Chiang,L.H.,“Adsorption/desorption properties of copper ions on the surface of iron-coated sand using BET and EDA analyses”,Chemosphere 41 pp.1249-1255, (2000)

Larsen, H.D., “Determination the Filtration Characteristics of Drilling Muds-Part I and Part II”, The Petroleum Engineer, 50-5,pp. 42-48, (1938)

Lee,I.,Han,W. .,Choi,J., and Kim,K., “Nanoparticle-directed crystallization of calcium carbonate”, Advanced Materials,13(21),pp 1617-1620,(2001)

Lee,J-K,Son,Y-H.,Tran,P.,Song,Y.,Martello,D.,and Chyu,M., “Catalytic properties of clay based hybrid materials for drilling operations”, PE 134021,(2010)

Lee,J-K., Sefzik,T., Son,Y-H., Phuoc, T.X., Soong,Y., Martello,D., and Chyu,M.K., “Magnetic nanoparticles for smart drilling fluids,” NTCE 18-04, AADE (2009)

Lécolier,E., Herzhaft ,B., ousseau,L., and Néau,L., “Development of a Nanocomposite Gel for Lost Circulation Treatment”, 94686-MS, SPE European Formation Damage Conference, 25-27 May 2005, Sheveningen, The Netherlands, (2005)

Leoppke,G.E.,Glowka,D.A.,and Wright,E.K., “Design and evaluation of Lost Circulation Materials for severe environments”,JPT,328,(1990)

Liu,H., Wei,Y.,Sun,Y and Wei,W., “Dependence of the mechanism of phase transformation of Fe(III) hydroxide on pH”,Colloids and Surfaces A: Physicochemical and Engineering Aspects, 252, pp. 201–205,(2005)

Lopez-Quintela, M.A., “Synthesis of nanomaterials in microemulsions: formation mechanism and growth control”,Curr. Opin. Coll. Int. Sci,8, pp. 137-144, (2003)

Lyons, W. C., “ tandard Handbook of Petroleum & Natural Gas Engineering”, Vol.1, 3rd Edition, Gulf Publishing Company, Texas, (1996)

Mackley,M.R., and N,E. Sherman., “Cross-flow Cake Filtration Mechanisms and Kinetics”, Chemical Engineering cience, 47(12), pp 3067-3084, (1992)

Maduka,C.J., “Experimental study of similarities in the filtration and resistivity properties of water based and oil based muds,” M c thesis, Texas Tech University,(2010)

Maghrabi, ., Wagle, V., Teke, K., Kukkarni,D and Kulkarni,K., “Low Plastic Viscosity Invert Emulsion Fluid ystem for HPHT”, AADE-11-NTCE-15,(2011)

Malshe,P.A.,Adhvaryu,A.,Verma,A., and McCluskey,H.P., “Nanoparticle based lubricants”, U Patent 0234149A1,(2008)

Manea,M.,“Designing of drilling fluids using nano scale polymer additives”, Revue Roumaine de Chimie,57(3),pp. 197-202,(2012)

Martine,M.J., and Ohmae,N., “Colloidal Lubrication: General Properties”,Nanolubricants, Chapter-1,John Wiley and Sons, England, (2008)

165

Mas,M., Tapin,T., Márquez,R., Gabay,R., Negrín,Z., Díaz,C and Bejarano,L., “A New High-Temperature Oil-Based Drilling Fluid”, Latin American and Caribbean Petroleum Engineering Conference, 21-23 April 1999, Caracas, Venezuela,(1999)

Mcdonogh,R.M., Fell,C.J.D., and Fane,A.G., “Surface-charge permeability in the ultrafiltration of non-flocculating colloids”, J. Membr. Sci., 21 pp. 285, (1984)

McGlothlin,R.E., and Baggett,C.J., “ Drilling Fluid”, U Patent 3,658,701, (1972) McLean,A.,Wilde,A.,Zamora,M and Rafferty,M., “The Top 10 Mud-Related Concerns in

Deepwater Drilling Operations-Revisited after 10 Years”,AADE-10-DF-HO- 04,(2010)

Mei,Z., Liu,S.Y., and Wang,L., “Preparation of positively charged oil/water nano- emulsions with a Sub-PIT method”, Journal of Colloid and Interface Science,361,pp. 565-572,(2011)

Messenger,U.J., “Lost Circulation”, Pennwell Publishing Company, Tulsa, Oklahoma, (1981)

Midoux,N.,Hoek,P.,Paillers,L.,Authelin,J., “Micronization of pharmaceutical substances in a spiral jet mill”,Powder Technology,104(2),pp 113-120,(1999)

Mitchell, J.K., and Soga, K.,“Fundamentals of oil Behavior,” John Wiley & Sons, New York, 3rd Edition, (2005)

Mitchell, J.K., and Greenberg, J.A.,“Chemico-Osmotic Effects in Fine- ained oils,” Journal of the Soil Mechanics and Foundation Division, ASCE, Vol. 99, No. SM4, pp. 307-322, (1973)

Miller,G., “Oil Based fluid composition”, U Patent 3,622,513, (1971) Milligan, D.I. and Weintritt, D.J., “Filtration of Drilling Fluids at Temperature of 300 T and

Above”,API Drilling and Production Practice, pp. 42-48,(1961) Mohan,K.K., Vaidyab,N. ., edd,G.M.,and Fogler, .H., “Water sensitivity of sandstones

containing swelling and non-swelling clays”, Colloids and urfaces A: Physicochemical and Engineering Aspects. 73,pp. 231-254,(1993)

Mokhatab,S.,Fresky,A.M.,and Islam, .M, “Application of Nanotechnology in Oil and Gas E&P”,JPT,pp.48-51,(2006)

Monteiro,O., and Quintero,L., “Graphene-containing fluids for oil and gas exploration and production”, WO 2012129302 A2,(2012)

Moore,P.L.,“Drilling practices manual,The petroleum publishing co. Tulsa,Ok,(1974) Moore,T.V., and Cannon,G.E., “Weighted oil base drilling fluid”,United States Patent

2055666,(1936) Moshkovita,A., Perfiliev,V.,Verdyan,A.,Lapsker,I.,Popovitz-Biro,R.,Tenne,R.,and

Rapoport,L.,“Sedimentation of IF-WS2 aggregates and a reproducibility of the tribological data”, Tribology International,40,pp 117-124,(2007)

Mosleh,M.,Atnafu, D. N.,Belk, H.J.,and Nobles,M.O., “Modification of sheet metal forming fluids with dispersed nanoparticles for lubrication,Wear,267,1220-1225,(2009)

Moyne, C., and Mourad, M.A.,“Electro-Chemo-Mechanical Couplings in Swelling Clays Derived from a Micro/Macro-Homogenization Procedure,” International Journal of Solids and Structures, Vol. 39, No. 25, pp. 6159-6190,(2002)

166

Mungun,N., “ Permeability reduction through changes in pH and salinity”,J.Pet.Tech.pp 1449-1453,Dec,(1965)

Nabhani,N., and Emami,M., “ The potential impact of nanomaterials in oil drilling industry”, Nanocon 2012,Brno Czech epublic,EU,23-25 October,(2012)

Nasr-El-Din, H.A., and Al-Humaidan,A.Y., Iron Sulfide Scale: Formation, Removal and Prevention, 68315-MS, International Symposium on Oilfield Scale, 30-31 January 2001, Aberdeen, United Kingdom, (2001)

Nassar,N., Husein,M., and Pereira-Almao,P., “Ultradispersed Particles in Heavy Oil: Part II, Sorption of H2S(g)”, Fuel Process. Technol.,91, 169-174, (2010) Nassar,N.N., and Husein,M.M., “ tudy and Modeling of Iron Hydroxide Nanoparticles

uptake by AOT (w/o) Microemulsions,Langmuir,23 (26),pp 13093-13103,(2007a) Nassar,N.N., and Husein,M.M., “Effect of microemulsion on copper oxide nanoparticles

uptake by AOT microemulsion,”Journal of Colloid and Interface cience,316 pp 442-450,(2007b)

Nelson, A.R., Geologic Analysis of Naturally Fractured Reservoirs, 2nd edition, Gulf Professional publishing, USA,(2001)

Nelson, P. H., “Pore-throat sizes in sandstones, tight sandstones, and shales”,American Association of Petroleum Geologists Bulletin, 93,pp.329 – 340,(2009)

Newman,K.,Lomond,P.,and McCosh.,K., “Advances in mixing technology improve drilling fluid preparation and properties”, AADE American Association of Drilling Engineers National technical conference & exhibition, New Orleans, Louisiana, NTCE-08-02,(2009)

Newman,k., Lomond,P., Jones,B., and McCosh,K., “Advances in Drilling Fluid Preparation and Conditioning at Fluid Mixing Facilities”, AADE-10-DF-HO-48, (2010)

nFluids Inc, “nFluids Explained”,www.nfluids.com,(2012) Njobuenwu, D.O., and Nna, E., “The effect of critical wetting agent concentration on

drilling fluids performance”,Journal of Science and Technology Research 4 (1),pp. 65-71,(2005)

Nicora, L.F., “High-Density Invert-Emulsion System with Very Low Solids Content to Drill ERD and HPHT Wells.” PE 65000, (2001)

Nimmo,J.R.,“Porosity and pore size distribution”, in Hillel,D.,ed. Encyclopedia of Soils in the Environment,London,Elsevier,V.3,pp-295-303,(2004)

Nyland, T., Azar, J. J., Becker, T. E., and Lummus, J. L., “Additive Effectiveness and Contaminant Influence Control on Fluid Loss control of Water-Based Muds”,SPE Drill. Engr., 6, pp. 195–203, (1988).

Nwaoji,O.C.,Hareland,G.,Husein,M.,Nygaard, R and Zakaria,F.M., “Nano-Particle Drilling Fluid Experimental Design Using Hydraulic Fracture Apparatus”, PE163434-M , PE/IADC Drilling Conference and Exhibition, Amsterdam, The Netherlands, 5–7 March (2013 )

Nwaoji,C.O., “Wellbore Strengthening- Nano-Particle Drilling Fluid Experimental Design Using Hydraulic Fracture Apparatus”, M c thesis, University of Calgary,(2012)

Ofite Lubricity Tester Instruction Manual, OFI Testing Equipment Inc,(2011) Okrajni, S.S., and Azar, J.J., “The Effects of Mud heology on Annular Hole Cleaning in

Directional Wells.” PEDE, pp 297-308,(1986)

167

Outmans,H.D., “Mechanics of static and dynamic filtration in the bore hole,” PE 491- PA,SPE Journal,pp-236-244,(1963)

Paiaman,M.A., and Al-Anazi,D.B., “Using nanoparticles to decrease differential pipe sticking and its feasibility in Iranian oil field, “Oil and Gas Business,(2008)

Pang, ., and harma, M.M.,“A Model for Predicting Injectivity Decline in Water Injection Well”. PE 28489,(1994)

Paiva,B.L.,Morales,R.A.,and Díaz,V. .F,“Organoclays: Properties, preparation and applications”,Applied Clay cience, 42, pp.8-24,(2008)

Pauling,L.,“The tructure of the Micas and elated Minerals”,Proc Natl Acad Sci U S A. 16(2),pp.123–129, (1930)

Payson Petroleum,“How does Drilling Fluid (Mud) work?”www. paysonpetro.com,(2011) Peng,J.S., “Filtration properties of water based drilling fluids”,PhD thesis, Heriot-Watt

University,UK,(1990) Phillips,R.J., Armstrong,R.C., Brown,R.A., Graham, L.A., Abbott,R.J., “A constitutive

equation for concentrated suspensions that accounts for shear-induced particle migration”,Phys. Fluids, A Fluid Dyn., 4, pp. 30–40, (1992)

Poulton,S.W.,and Raiswell,R., “ Chemical and physical characteristics of iron oxides in riverine and glacial meltwater sediments.Chem Geol.218,203-221,(2005)

Q’Max Technical Bulletin #7, “Differentially tuck Pipe”, Q’Max olutions Inc Qinfeng,Di., Shen,C.,Wang,Z.,Jing,B.,Gu,C.,and Qian,Y.,“Innovative Drag Reduction of

Flow in Rock's Micro-channels Using Nano Particles Adsorbing Method”, SPE International Oil and Gas Conference and Exhibition in China, 8-10 June 2010, Beijing, China,(2010)

aveendran, P., “Mechanisms of particle detachment during filter backwashing”, PhD Thesis, School of Civil Engineering, Georgia Institute of Technology, Atlanta, (1993)

Ravi,K., Patil, C.R., Patil, P.S, Pindiprolu, K.S., Roddy, C.,Convington, R.,Chatterji.J., and Brenneis,C., “Loss circulation compositions and associated methods”, U Patent 0162845 A1,(2011)

Rahman,A.M., and Oomori,T.,“In vitro regulation of CaCO(3) crystal growth by the highly acidic proteins of calcitic sclerites in soft coral, Sinularia Polydactyla”, Connect Tissue Research,50(5),pp 285-293,(2009)

Reed,J.K., “Ceramic high temperature lubricity agent”,WO 2008030810 A2,(2008) Reid, P.I., Meeten, G.H., Way, P.W., Clark, P., Chambers, and B.D.,Gilmour, A.,

Sanders, M.W., “Differential-sticking mechanisms and a simple wellsite test for monitoring and optimizing drilling mud properties. SPE Drilling and Completion,15 (2), (2000)

Rezaee,R.,Saeedi,Ali., and Clennell, B., “Tight gas sands permeability estimation from mercury injection capillary pressure and nuclear magnetic resonance data”,Journal of Petroleum Science and Engineering,88-89,pp.92-99,(2012)

Riley,M.,Stamatakis,E.,Young,S.,Hoelsher,P.K.,Stefano,D.G.,Ji,L.,Guo,Q.,and Friedheim,J.,“Wellbore stability in unconventional shale-the design of a nanoparticle fluid”, PE 153729,presented at PE oil and gas india cond. And exhibition held in Mumbai,India,28-30 March,(2012)

168

Roddy,W.C., Chatterji,J.,and Cromwell,R., “Well treatment composition and methods utilizing Nano-Particles”, U Patent 7,559,369B2,(2009)

Rodriguez,E.,Roberts,M.R.,Yu,H.,Huh,C., and Bryant,S.L., “ Enhanced migration of surface-treated nanoparticles in sedimentary rocks,” PE 124418 presented at the 2009 SPE Annual Technical Conference and Exhibition, New Orleans,LA,4-7 October,(2009)

RSC (Royal Society of Chemistry), “Growing riches”,http://www.rsc.org /chemistry world/Issues/2003/February/riches.asp,(2003)

Russel, W.B., Saville, D.A., and Schowalter,W.R., “Colloidal Dispersions”, Cambridge University Press,UK, p-373,(1999)

Saboori,R.,Sabbaghi,S.,Mowla,D., and Soltani,A., “ Decreasing of water loss and mud cake thickness by CMC nanoparticles in mud drilling”, Int.J.Nano Dimens, 3(2), pp101-104,(2012)

Sáchez,G.,León,N.,Esclapés,M.,Galindo,I.,Martínez,A.,Bruzual,J., and Siegert, I., “Environmentally safe oil based fluids for drilling activities,” Paper PE 52739 presented in Austin,Texas 28 February-3 March,(1999)

Sayyadnejad,A.M., Ghaffarian,R.H, and Saeidi.M., “Removal of hydrogen sulfide by Zinc oxide nanoparticles in drilling fluid,” Int. J. Environ. Sci. Tech., 5 (4), pp. 565-569,(2008)

Schmidt,D.D., Roos,A.F., and Cline,T.J., “ Interaction of water with organophilic clay in Base oils to build viscosity,“Paper PE 16683 presented at Dallas,T , petember 27-30,(1987)

Schremp, F.W.,and Johnson, V.L., “Drilling Fluid Filter Loss at High Temperature and High Pressures”. Pet.Trans. of AIME.,195, pp. 157-162,(1952)

Scoggins,C.W.,and Ke,M., “Method for increasing lubricity of brine-based drilling fluids and completion brines”,U Patent 8,071,510B2,(2011)

Scott,D.P., Robinson,H.L., “Economic Considerations and Impacts for Using Low Grade Barite”, AADE-10-DF-HO-19,(2010)

Sensoy,T., Chenevert,M.E., and Sharma,M.M., “Minimizing water invension in shale using nanoparticles”, PE 124429 presented at the 2009 PE Annual Technical Conference and Exhibition, New Orleans,LA,4-7 October,(2009)

Sethi,S ,“Transient permeate flux analysis,cost estimation, and design optimization in cross flow membrane filtration”, (1997)

Sharma,M.M., and Jiao,Di., “Formation damage due to static and dynamic filtration of water-based muds, Paper SPE 23823 presented in Louisiana, February 26 27,(1992)

helton,J., “ Experimental investigation of drilling fluid formulation and proceesing methods for a riser dilution approach to dual density drilling”, M c thesis, Louisiana State University,(2005)

Sherwood, J.D.,“Ionic Motion in a Compacting Filter-Cake,” Proceedings of the Mathematical and Physical Sciences, Vol. 437, pp. 607-627,(1992)

Smalley,E.R., and Yakobson, I.B., “The future of the fullerenes," Solid State Communications, 107, pp. 597-606, (1998)

Smith,C.G.A., and Hartman,H., “Clay Minerals and the Origin of Life”, Cambridge University Press,England,(1987)

169

impson,P.J., “ A new approach to oil-base muds for lower cost drilling”,Journal of Petroleum Tech, pp-643-650, (1979)

imon,A,K “ The role of different rheological models in accuracy of pressure loss prediction”, Rud.-geol.-naft. zb., Zagreb ,16, pp. 86-89, (2004)

Simon,G., Leonard, B., Trotter,N., Morgan,M., Young,G., Morrison,R and Bridges ,S., “Effects of Lost Circulation Material etention on Drilling Fluids and Drilling Waste Management”, AADE-10-DF-HO-23,(2010)

Song,L.F., and Elimelech,M., “Particle deposition onto a permeable surface in laminar- flow”, J. Colloid Interface Sci., 173,pp. 165, (1995)

Spier,R.M., Letti, AJ,L., Woiciechowski,L,A., and Soccol,R.C., “A simplified model for A.Niger FS3 growth during phyyase formation in solid state fermentation”, Braz. arch. biol. technol. vol.52 no. spe Curitiba Nov. (2009)

Srivatsa,T.J., and Ziaja,B.M.,“An Experimental Investigation on Use of Nanoparticles as Fluid Loss Additives in a Surfactant - Polymer Based Drilling Fluids”, International Petroleum Technology Conference, 7-9 February 2012, Bangkok,Thailand,(2012)

Srivatsa,T.J., “An Experimental Investigation on use of Nanoparticles as Fluid Loss Additives in a Surfactant–Polymer Based Drilling Fluid”, Texas Tech University, M.Sc thesis,(2010)

Streat,M.,Hellgardt,K and Newton,N.L.R., " Hydrous ferric oxide as an adsorbent in water treatment part 1. Preparation and physical characterization, Process Safety and Environmental Protection,86,pp.1-9,(2008)

Suri,A., and Sharma,M.M., “Strategies for Sizing Particles in Drilling and Completion Fluid”, PE Journal,9(1),pp.13-23,(2004)

Tai,Y.C., and Chen,K.C., “Particle morphology, habit, and size control of CaCO3 using reverse microemulsion technique”, Chem. Eng. ci.,vol. 63, pp. 3632-3642, (2008)

Tran,H.M., Abousleiman,N.Y., and Nguyen,X.V., “The Effects of Filter-Cake Buildup and Time-Dependent Properties on the tability of Inclined Wellbore”, SPE Journal, 16(4), pp. 1010-1028, (2011)

Tran,XP.,Lyons,KD.,Gupta,R.,andWalker,L.,NETL, http://www.netl.doe.gov/publications/ factsheets/rd/R&D108.pdf, (2007)

Tien, C., and Payatakes, A.C., “Advances in Deep Bed Filtration”, AIChE J., 25: 737, (1979)

Tipping,E., and Cooke,D., “The effects of adsorbed humic substances on the surface charge of goethite (a-FeOOH) in fresh waters”, Geochim,Cosmochim Acta 46,75-80,(1982)

Tipping,E.,“The adsorption of aquatic humic substances by iron oxides”,Geochim, Cosmochim Acta 45,191-199,(1981)

Tong,H.,Ma,W.,Wang,L.,Wan,P.,HU,J.,and Cao,L.,“Control over the crystal phase,shape, size and aggregation of calcium carbonate via a L-aspartic acid inducing process”,Biomaterials,25(17),pp.3923-9,(2004)

Tour, M.J, Schmidt, K.H., Lomeda, R.J., and Kosynkin. V.D,“Graphene compositions and drilling fluids derived thereform”, U Patent 0059871A1 ,(2011)

170

Tovar,J., Azar, J.J., Lummus,L.J., Sarma,S.D, and Pilehvari, A., “Formation Damage Studies on Reservoir Rocks using Water-Based and Oil-Based Muds,”Paper SPE 27349 presented in Louisiana 7-10 February,(1994)

Verma,A.,Jiang,W.,Abusafe,H.H.,Brown.D.W., and Malshe,P.A.,“Tribological behavior of deagglomerated active inorganic nanoparticles for advanced lubrication”,Tribology Transaction,51(5),pp. 673-678,(2008)

Vasii,l.L.,“Nanotechnology,possible solutions for drilling fluids”, Alberta Nanotech Showcase, (2008)

Vickers,S., Cowie,M., Jones,T., and Twynam,A.J., “A new methodology that surpasses current bridging theories to efficiently seal a varied pore throat distribution as found in natural reservoir formations”, WIERTNICTWO NAFTA GAZ,TOM 23/1pp.501-515, (2006)

Whitfill,L.D.,Tare,U.,and Hemphill., “Managing Circulation Losses with ynthetic Base Drilling Fluid”, AADE-03-NTCE-39 ,(2003)

Whitfill,D., “Lost Circulation Prevention and emediation”, IADC/SPE Asia Pacific Drilling Technology Conference and Exhibition, 25-27 August 2008, Jakarta, Indonesia, IADC/SPE 115039,(2008)

Wilcox,M., Baldwin,R., Garcia-Hernandez,A., and Brun,K., “Guideline for gas turbine inlet air filtration systems,” in Proceedings of the Gas Machinery esearch Council, (2010)

Williams,M.,and Canon, G.E., “Evaluation of filtration properties of drilling muds,” The Oil Weekly,pp. 25-32,(1938)

Wu,J., Proett,A.M., Torres-Verd¡n,C.,and epehrnoori,K., “The Influence of Water-Base Mud Properties and Petrophysical Parameters on Mudcake Growth, Filtrate Invasion, and Formation Pressure”, ociety of Petrophysicists and Well-Log Analysts, 46(1),(2009)

Wu,YY.,Tsui,C.W, and Liu,C.T, “Experimental analysis of tribological properties of lubricating oils with nanoparticle additives”,Wear,262,pp819-825,(2007)

Xu,X., Yu,Z.,Zhu,Y., and Wang,B., “Effect of sodium oleate adsorption on the colloidal stability and zeta potential of detonation synthesized diamond particles in aqueous solutions, Diamond Relat. Mater. 14,pp.206-212,( (2005)

Xuan,Y., and Li,Q., “Investigation on convective heat transfer and flow features of nanofluids,” Journal of Heat Transfer, 125(1), pp. 151-155,(2003)

Xuan,Y., and Li,Q., “Heat transfer enhancement of nanofluids,” Int. J. Heat & Flow, 21(1), pp. 58-64,(2000)

Yang,Z.,Liu,Y.,Zhao,X.,Song,T.,Yan,J., and Jia,W., “Research and application of nanoscale emulsion lubricating material for drilling fluid in Daqing oil field”,IADC/ PE 161899,1-7,(2012)

Yao,C.L.,Qi,CX.,Z,JM., and XU,WH.,“Unusual morphology of calcium carbonate controlled by amino acids in agarose gel”,J.Chil.Chem. oc.55(2),pp 270-273, (2010)

Ying,Z.,“High performance drilling fluids with submicron-size particles as the weighting agent”,U Patent 8252729 B2,(2012)

Zamani, A., and Maini, B., “Flow of Dispersed Particles through Porous Media – Deep Bed Filtration”, J. Pet. ci. Eng., 69, pp.71-88,(2009)

171

Zamora ,M.,Broussard,P.N., and Stephens,M.P., “The Top 10 Mud-Related Concerns in Deepwater Drilling Operations”, SPE International Petroleum Conference and Exhibition in Mexico, 1-3 February 2000, Villahermosa, Mexico,(2000)

Zakaria,M., Husein,M., and Hareland,G., “Novel Nanoparticle-Based Drilling Fluid with Improved Characteristics”, PE156992, International Oilfield Nanotechnology, Conference and Exhibition, Noordwijk, The Netherlands, June 12–14, (2012)

Zakaria,F.M, Mostafavi,V., Hareland,G., Husein,M., “Design and Application of Novel Nano Drilling Fluids to Mitigate Circulation Loss Problems during Oil Well Drilling Operations” presented at the World Nano Conference and Expo 2011 in Boston, Massachusetts, U.S.A on June 13-16, (2011)

Zhang,M.,Wang, .,Fu, i., ia,Y., “Performance and anti-wear mechanism of CaCO3 nanoparticles as a greem additive in poly-alpha-olefin”, Tribology International, 42,pp.1029-1039,(2009)

Zielińska-Jurek,A., eszczyńska,J., Grabowska,E., and Grabowska,A.,“Nanoparticles Preparation Using Microemulsion ystems”, InTech, (2012)

Zwager, D., “Airbased drilling technique for faster and easier drilling process and maximizing production from geothermal wells (in Indonesian)”, PT Air Drilling, Indonesia, (2007)

172

Appendix A: Classification of lost-circulation zones

Mud losses vary in type, severity, and location in the hole. Mud losses occur to the

following types of formations (Messenger,1981):

1. Unconsolidated or highly permeable formations (porous sands and gravels)

The permeability of the porous formation that takes the whole mud or cement must

exceed 10 darcies. Gravels and shallow sands often reveal such permeabilities. The

deeper sands seldom exceeds about 3.5 darcies and therefore they are not often

consider as loss zones unless they are not fractured.

2. Natural fractures

Loss is evidenced by gradual lowering of the mud in the pits. If drilling is continued and

more fractures are exposed, complete loss of returns may be experienced. Natural

fractures reservoirs are the reservoir that contains fractures created by the stress that

exceed the rupture strength of the rock (Nelson,2001). These fractures may exist in the

deeper formation with little or no width. That’s why the mud losses to them are small

until the fractures are widened.

3. Induced fractures

There may be some cases when the horizontal fracture can be induced. One of the

most common is in shale. Loss is usually sudden and accompanied by complete loss of

returns.

4. Cavernous formations

Caverns form mainly in limestones. Loss of returns may be sudden and complete. It will

make the sealing more difficult.

173

Appendix B: Lost circulation materials size selection methods

There have been several methods given in Table B.1 for the selection of the lost

circulation materials based on their size to keep the mud loss at minimum.

Table B.1: Lost circulation materials selection methods.

Halliburton Method

(Whitfill,2008)

The particle size distribution is equal to the

estimated fracture width to offset uncertainty in

the estimation. Enough particles smaller and

larger than the fracture are present to plug

smaller and larger fracture width.

Vickers et al.,(2006)

For minimal fluid loss in the reservoir, the

following criteria should be met

D(90) = largest pore throat

D(75) < 2/3 of largest pore throat

D(50) +/- 1/3 of the mean pore throat

D(25) 1/7 of the mean pore throat

D(10) > smallest pore throat

IPT (Ideal Packing Theory)

(Dick et al.,2000)

The IPT addresses either pore sizing from thin

section analysis or permeability information,

combined with PSD of bridging material, to

determine ideal packing sequence.

Cargnel and Luzardo,(1999) The criterion of selection of particle size of

bridging agents is: 1/7 DPore throat < Dparticle

< 1/3 DPore throat. This yields a small invasion of

solids into the porous media.

Abrams’ Median Particle-Size Rule

(Abram,1977)

The median particle size of the bridging

material has to be equal or slightly greater

than 1/3 the median pore size/fracture size of

the formation.

174

Appendix C: Diffusion coefficient and Peclet number

The diffusion coefficient and peclet number shown in Figures 5.8 and 5.9 are given in the following tables C.1-C.4. Table C.1: Effect of particle sizes of DF (dp =2-200 μm in DF) on Peclet number which is a control sample of Fe(OH)3 NP-based fluid.

Table C.2: Effect of Fe(OH)3 NPs size in DF ranges from 0.001-0.3 μm (1-300 nm) on Peclet number.

175

Table C.3: Effect of particle sizes of DF (dp =2-200 μm in DF) on Peclet number which

is a control sample of CaCO3 NP-based fluid.

Table C.4: Effect of CaCO3 NPs size in DF ranges from 0.001-0.3 μm (1-300 nm) on Peclet number.